<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Alex Hornstein on Medium]]></title>
        <description><![CDATA[Stories by Alex Hornstein on Medium]]></description>
        <link>https://medium.com/@alexhornstein?source=rss-70871fc45f5c------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*SYKFls9QDa2SvhCxJxE4TQ.png</url>
            <title>Stories by Alex Hornstein on Medium</title>
            <link>https://medium.com/@alexhornstein?source=rss-70871fc45f5c------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Fri, 20 Nov 2026 03:52:10 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@alexhornstein/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Comic 001: a first pass at a data-driven narrative]]></title>
            <link>https://medium.com/wild-data-comics/comic-001-a-first-pass-at-a-data-driven-narrative-0bd20b3862a7?source=rss-70871fc45f5c------2</link>
            <guid isPermaLink="false">https://medium.com/p/0bd20b3862a7</guid>
            <category><![CDATA[storytelling]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[comics]]></category>
            <category><![CDATA[animals]]></category>
            <category><![CDATA[wildlife]]></category>
            <dc:creator><![CDATA[Alex Hornstein]]></dc:creator>
            <pubDate>Fri, 20 Feb 2026 03:29:08 GMT</pubDate>
            <atom:updated>2026-02-20T03:29:08.933Z</atom:updated>
            <content:encoded><![CDATA[<p>I’m going to make a series of comics based on data over the coming weeks. This is the first one, and I’m going to go through how my program and I parse the tracking data and art-direct the comic, panel-by-panel. But first, the comic:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*r_7sfpB1sRdN2P3GuzWW0g.jpeg" /></figure><p>When I think about finding these narratives, before I even get to the data, I have to talk about the larger context of the data: what animal species are we tracking, and why are we tracking it in the first place? For these goats, the context starts with a wildfire in Chile in February, 2024.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/852/1*rUyxwwMYsRyEF7sgGX3e3w.jpeg" /></figure><p>In 2024, in the midst of an unprecedented drought and heat wave, Chile <a href="https://en.wikipedia.org/wiki/2024_Chilean_wildfires">suffered a series of devastating wildfires</a>. The combination of unusually hot and dry weather, in addition to a large amount of accumulated, flammable underbrush led to these first fires burning out of control, and reaching the hills outside of the major city of Valparaiso, burning thousands of homes and causing major damage and loss of life.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*LVgO4crtz0vHTWpNHY7iJg.jpeg" /></figure><p>In response to the fires, Chilean group <a href="http://buenacabra.cl/">Buena Cabra</a> set out to systematize the use of goats grazing underbrush to create natural firebreaks. Their pilot project is on a hillside some 500km south of Valparaiso, where they have a herd of goats which they are monitoring in order to understand how quickly they work, and the proper ways to field the goats in order to achieve the desired results. Two of those goats are wearing CollarID tracking collars, and that’s the data source for this first project. These goats might not know it, but they are part of humanity’s attempt to adapt to a changing planet.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/960/1*x8I_YUCaRBydgJeGCl0Vog.jpeg" /></figure><p>The particular goats, which are just labeled with ID0002 and ID0005, are both domestic goats that are part of a small herd. They’re kept in a shed at night, and are let out to graze the hillside during the day. I have a call coming up with the folks from Buena Cabra — the info I have so far is second hand, through Collar ID, who provided the tracking data. I’m very curious, when we do get to talk directly, to see how my blind interpretation of the data matches up with the reality on the ground.</p><p>That context matters, both in the context of how I interpret the goat’s data, and how I describe the landscape that the goats are in. The wildfires are recent enough that the scars of the burn are still visible, and when the data was taken, a year and a half after the fires, the vegetation was in an early stage of regrowth. I don’t have visual <em>data</em> to back that up, just an oral context, but that information will go into the landscape description for each panel.</p><p>The data itself is quite thorough — collar ID trackers collect quite a bit of data about the environment the animal is in, as well as information about how the animal moves through its environment. In this dataset, that includes:</p><ul><li>GPS position <em>sampled every five minutes</em></li><li>3-axis accelerometer <em>sampled continuously @ 50hz</em></li><li>Audio of the ambient environment <em>sampled two minutes on-two mins off</em></li><li>Ambient light levels <em>sampled continuously</em></li><li>Temperature <em>sampled continuously</em></li></ul><p>My first crack at reading that data was basically just to write a bunch of code that would load and parse it all.</p><h3>GPS data</h3><p>GPS data tells you both a lot and a little. My first approach was to load a list of lat/long points and draw them on a map:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*IiAoC_ILj44FTXSlQGfXsA.png" /><figcaption>three months of GPS track, wandering around the same hillside</figcaption></figure><p>That doesn’t say much to me. I looked at it a few other ways, first as a heatmap of all the days combined, and then as an animated heatmap, showing popular areas day by day</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7RKCBpJCOtcV0qhS2_Ld8A.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*78K4w1-g0FIXneabd4WfmA.gif" /></figure><p>These are nicer to look at than the track data, but it still really only shows me one thing: the goats mostly hung out on this one hill.</p><h3>Scene Descriptions</h3><p>We eventually want to describe the place where the goats are — this will be the background for the panels, so I made a function that takes a GPS position and creates a written scene description, along with a 3D model of the terrain and a satellite view of the terrain at that point, based on contour map and satellite imagery libraries.</p><p>For GPS point (-37.347201, -72.916191), that gives me the following output:</p><blockquote>The time is 2025–08–28 11:47:19–04:00. At (-37.34720, -72.91619), the ground elevation is about 169.6 (DEM units, usually meters). The surrounding terrain within ~800 m is very steep (median slope ~90.0°) and generally slopes down toward S (~181°). Local relief is ~171.1 (min 117.1, max 288.2), which is very rugged. Relative to its neighborhood, the point sits on near the local average elevation (TPI ~+4.1). Terrain roughness (TRI-like) is ~20.6.</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OLMzDn6xNG1MCLpRlA2E7w.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/1*9LCUP8h9yCfewHAfQA_xKg.png" /></figure><p>Now, this is fine, but not super informative. The big thing is really that I have some data and a text description, and if I add something like “I want you to illustrate the scene from the point of view of a <em>&lt;goat&gt;</em>, in <em>&lt;comic, pen and ink style&gt;</em>, of the scene at GPS point (-37.347201, -72.916191), where the terrain….”, that is a reasonable prompt for an AI image generator, and entirely based on data. It’s telling the story, “the goat was here”. Not al interesting, but it’s the simplest data-driven panel I can make. I plopped it into an image generator as an experiment, along with the terrain model and satellite image for reference, and got this:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Sr-N-g53qJTsEuLkNuYdYg.png" /></figure><p>This isn’t right — the goat is in a much more desolate, rocky environment than the satellite image represents. I decided I needed to give an explicit description of the vegetation, rather than trusting the AI to interpret the satellite image correctly. In order to do that, I found the Worldcover library, which can create a description of vegetation for a given time, based on data from the <a href="https://sentinels.copernicus.eu/copernicus/sentinel-2">sentinel-2 satellite</a>.</p><p>No sooner did I get into that than I realized, while this would be great for future animal stories, it wouldn’t help for this one, because the worldcover datasets were from 2021 and 2022, and these goats were responding to a wildfire which hit Chile in 2024. I would have to manually add a description here,</p><blockquote>this landscape has recently burned in a wildfire and, the vegetation should be native Chilean vegetation which is regrowing after the burn</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2EmIrcefLeX-QzfLpwhlog.png" /></figure><p>That’s more like it!</p><h3>Finding a narrative in the data</h3><p>The next question is, what can I find in the data that is worth telling? Let’s start by not just look at where the animal is, but also <em>how</em> it’s moving around, by looking at distance covered, velocity and acceleration, based on the GPS data:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*b80x9k6TtZ93xeU9zIk5gQ.png" /></figure><p>That green line in velocity is a rolling average, smoothing out some GPS jitter. The final trace below shows the delta T between GPS points — if the collar wasn’t able to get a GPS fix, it won’t save data until it gets another one, so data in or near that range is not particularly meaningful.</p><p>Talking with Patrick, who runs collar ID, about the daily rhythm of the goats, he gave me some helpful <strong><em>context </em></strong>— the goats are kept in a shed, and are released by hand each day, where they walk around the hill. The shed is enclosed, and that’s why there are big gaps in GPS fixes — the GPS may not be able to see the satellites, or might have more error in its position data than it would otherwise.</p><p>I decided to look at one other data source to map when the goats were in the shed — the ambient light levels.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*lIxrF8VOBwLcsWeYfyDA1Q.png" /></figure><p>I can’t <em>begin</em> to describe how critical plotly graphs were for <em>me</em> to look at the data, scrub through for interesting moments, and understand what was going on there. If you can’t see it, you can’t think about it!</p><p>This raises one more question, maybe the biggest question — how do you actually tell a story based on data? As a first experiment, I decided to go through the data for notable shifts — sudden changes in levels, outliers, etc., and call each one of those a ‘Moment of Interest’, with an accompanying text description of what was happening at that moment.</p><p><strong>Moments of Interest</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/750/1*XtjmhGvsNqtOxHQsZ6QpRw.png" /></figure><p>For day one, I went through and pulled in moments of interest by hand. Eventually, this needs to get automated, but you gotta walk before you can run.</p><p>Here’s the list:</p><ul><li>the goats are dark and still, sleeping in the shed at 4:30am</li><li>the shed first starts to get dimly lit, 5:05am</li><li>the moment goats go out from the shed into the light, 2:17pm</li><li>a characteristically ambient, slow moment from the afternoon, 4:03 pm</li><li>a sudden acceleration and sustained velocity — something spooked the goat, maybe *chased* it — velocity was up for 10 mins. 5:07pm</li><li>goats are still and outside, light is falling, 8:38pm</li></ul><h3>Panel layout</h3><p>I made a set of possible panel layouts for the comic. For this one, I picked a five-panel layout I called ‘expository”</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/612/1*D1s_whHr0VvxCpviBuOuqQ.png" /></figure><p>I imagine that there will always be as many or more moments of interest than there are panels, so the challenge will be to draw a narrative based on N moments of interest — something that fits a panel layout I give it. Eventually, I will want to write a little program that parses an SVG and figures out how many panels there are automatically, and then fills in each one with the generated imagery. For now, I am that program.</p><p>In this case, there were five panels and six moments of interest. I decided that both the shed scenes were redundant, so I skipped the wake-up scene in the shed.</p><p>The other questions, when it comes to the storytelling, are cinematography and scene content: how does the camera frame the scene, and how do I figure out what is in each frame? This really is going to be the key programming question as this project continues — how do I use the sensors and context to create an accurate and thorough description of the scene? If it’s always pictures of a goat, it’s only good for #goatSelfie on goatstagram. I need to create a separate logic that acts like an AI director of photography — how do I frame each image so that it‘s not monotonous, but it also makes sense in the larger story?</p><h3>Making the art</h3><p>The next thing is to actually generate prompts for each moment of interest, as well as reference images that I need to send the image generator to give context for the characters. As it turns out, I have photographs of each animal wearing the collar, and I can use that as a character reference, giving me a data-driven, consistent character.</p><p>My little loop, which goes through each moment of interest, creates a directory with a scene description based on the animal’s GPS point and ambient environment data. If I add a custom, manually entered description of the action, scene composition or camera framing, that gets pushed into the ‘customDescription.txt” file. Satellite, Terrain and the character reference image get sent to the image generator, along with the description or custom description prompt</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*t5Af_rqvAhLBnU90XwhRww.png" /></figure><p>For the first image, I have a manually written custom scene description, but based on the data — the audio data at this moment is a very comforting chorus of goat snores, the only direct link to other characters in the story. The other thing I know is the light level, lack of GPS position, and stillness of accelerometer, a good indication that the goats are snoozing. I sent over:</p><blockquote>I want you to draw an image, in pen and ink comic style, landscape orientation, 2:1 aspect ratio, of a series of goats sleeping on eachother in a small, darkened shed. Shot should be medium-wide, showing goats resting on eachother, centered on the head of our protoganist goat, based on the two photos below</blockquote><p>and got back</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4gHfcaAPuTiBs6s61rJY_g.png" /></figure><p>….</p><p>As I went through all five panel generators, I get five individual images:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ziITAIklRGUtye71AO7wnQ.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jBolI8qsZ6tZ9nKaFYSAOw.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-OwBNMTeVbInknc9rWhXiQ.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*h__F0XFKyj7CZeVZ27fRlw.png" /></figure><h3>Putting it all together</h3><p>The last step is to place them in the comic layout, and give it a title</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*r_7sfpB1sRdN2P3GuzWW0g.jpeg" /></figure><p>This is a fine comic for a first pass, and I’m happy with this result, but I wouldn’t want to read one of these comics everyday for a month. Don’t get me wrong - I’m very happy that I went through the process of making this comic, because it informs the methods for making more, for automating it, and for improving the narrative. I’m saying that honestly — it’s important to get a version 1.0 out, and it’s fine to criticize your own work, as long as you keep on going. Having made my v1.0, I believe I can do much better. Stay tuned for the next experiments!!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=0bd20b3862a7" width="1" height="1" alt=""><hr><p><a href="https://medium.com/wild-data-comics/comic-001-a-first-pass-at-a-data-driven-narrative-0bd20b3862a7">Comic 001: a first pass at a data-driven narrative</a> was originally published in <a href="https://medium.com/wild-data-comics">Wild Data Comics</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Wild Data Comics: a seed of an idea]]></title>
            <link>https://medium.com/wild-data-comics/wild-data-comics-a-seed-of-an-idea-bad21502a29e?source=rss-70871fc45f5c------2</link>
            <guid isPermaLink="false">https://medium.com/p/bad21502a29e</guid>
            <category><![CDATA[wildlife]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[comics]]></category>
            <category><![CDATA[storytelling]]></category>
            <category><![CDATA[big-data]]></category>
            <dc:creator><![CDATA[Alex Hornstein]]></dc:creator>
            <pubDate>Mon, 16 Feb 2026 18:45:46 GMT</pubDate>
            <atom:updated>2026-02-17T00:14:47.173Z</atom:updated>
            <content:encoded><![CDATA[<p>A few months ago, I was talking with Patrick from <a href="https://collarid.org:2087/">CollarId</a> about why the world should care about wildlife tracking collars. I had a funny thought — in the modern world, where we have all the computation, data, and technological ability, what about making a program that makes daily comics about individual wild animals, driven by the data coming from trackers on those animals?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/960/1*x8I_YUCaRBydgJeGCl0Vog.jpeg" /><figcaption>A domestic goat in Chile with a CollarID sensor</figcaption></figure><p>There’s a clear reason why <strong>scientists</strong> care about these collars — right now, on earth, ~170,000 wild animals across ~1000 species have some form of tracking device on them, and scientists use that information to get a close, individualized picture of those animals’ lives — everything from how they eat to how they socialize and how they migrate. The broader problem is that most of that information ends up in science papers, and most of the world doesn’t read science papers. Every so often, one of the pieces of information that we learn from deploying trackers will turn into a catchy Science Communication Fact, and that gets circulated, without context, in nature documentaries and pop science articles, generally without any sense of how hard it was to learn this Fact, and the incredible level of technology, biology and field work that had to happen in order for us to rattle off this Fact.</p><p>Some groups, like <a href="https://myfahlo.com/">Fahlo</a> or <a href="https://www.atlanticwhiteshark.org/sharktivity-app">Sharktivity</a>, have found the first level of public engagement with trackers — people are fascinated by charismatic animals, so take the animal tracking data that is coming out of research programs and have people follow the lives of those animals, making a ~real time-ish connection. If people feel inspired to support the research, so much the better!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/503/1*n7t0uTWMgi_wHaEgiMiX9Q.png" /><figcaption>Sharktivity app showing white shark sightings along Cape Cod over the past year</figcaption></figure><p>I’m a huge fan of Steve Irwin (<a href="https://www.youtube.com/playlist?list=PL70iApDFZRdU_7pPhIJC-DFEBGDhLqYEr">the Crocodile Hunter</a>) and these <a href="https://www.goodreads.com/book/show/130256781-the-crocodile-hunter">two</a> <a href="https://www.amazon.com/Crocodile-Hunter-Steve-Irwins-Life/dp/B0G1LYZC4L">biographies</a> focusing on his life. The idea for the Crocodile Hunter started in a bar — Steve was in a bar in Australia, a nature documentary on the TV, and everyone was just ignoring it. He loved nature documentaries, but he realized at that moment that they were just preaching to the choir. If you wanted to make people care about nature, it wasn’t enough to show beautiful imagery from nature to people who cared enough to go to the video store or theater and seek it out: you had to make it fun to watch in a bar. That was the origin of the Crocodile Hunter character — sure, it was ultimately a powerful vessel for conservation, but its power comes from being, at its core, really really fun to watch.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/940/1*xWWfRhab_gjAdj6woJ4oOQ.jpeg" /><figcaption>this sure looks more fun than that map</figcaption></figure><p>This got me wondering, is there a more compelling way to use wildlife tracking data to tell stories about animals? Talented sensor designers are making increasingly sophisticated sensors that give us an unprecedented lens into the <a href="https://www.medtronic.com/en-us/our-impact/wildlife-conservation.html">inner physiology</a>, <a href="https://www.biomark.com/pit-tags/">motion</a>, and <a href="https://collarid.org:2087/">ambient surroundings</a> of the animal. Scientists are using this information to get glimpses into previously unseen parts of animals’ lives, but outside of science papers, this information is mostly unseen.</p><p>This will be the first of many posts I write about my experiments as I work to transform wildlife tracking data into compelling comic narratives, while still remaining true to the animals’ experience. After this first conversation, I spent some time building some initial data-driven storytelling experiments, which I’ll share over the coming days. Today, I am starting my residency at the <a href="http://recurse.com/">Recurse center</a>, where I’ll be focusing on this project for the next six weeks, exploring the various facets of this type of wildlife storytelling and publishing my results as I go.</p><p>I’ll be working with data from many different sources, using a host of programming and analytical techniques to represent animal behavior, state of mind, and activities. The end result should be compelling — I want to create stories that capture the essence of life for wild animals, and that can appeal to people of all backgrounds, ages, and interests and draw them in to the animals’ world. As I work on this, I’ll be posting regular updates on my progress to this publication, including the comics I create, an in-depth look at how I work with the data, and thoughts on how to sift out compelling narratives from raw tracking data.</p><p>Stay tuned!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=bad21502a29e" width="1" height="1" alt=""><hr><p><a href="https://medium.com/wild-data-comics/wild-data-comics-a-seed-of-an-idea-bad21502a29e">Wild Data Comics: a seed of an idea</a> was originally published in <a href="https://medium.com/wild-data-comics">Wild Data Comics</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Holograms of the Golden Age]]></title>
            <link>https://medium.com/@alexhornstein/holograms-of-the-golden-age-6be06384576c?source=rss-70871fc45f5c------2</link>
            <guid isPermaLink="false">https://medium.com/p/6be06384576c</guid>
            <category><![CDATA[hologram]]></category>
            <category><![CDATA[stereo-photography]]></category>
            <category><![CDATA[stereograms]]></category>
            <category><![CDATA[light-field]]></category>
            <category><![CDATA[looking-glass]]></category>
            <dc:creator><![CDATA[Alex Hornstein]]></dc:creator>
            <pubDate>Thu, 24 Oct 2019 12:55:19 GMT</pubDate>
            <atom:updated>2019-10-24T17:17:11.348Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/480/1*as7akCRfA4vxMpQOVkiOkg.gif" /><figcaption>New York City Filmmaker, 1872</figcaption></figure><p>I’ve been fascinated by putting imagery from the real world into the Looking Glass. Ever since we started making 3D displays, I’ve loved the idea of a digital shoebox diorama — a tiny scene from another time and place playing out inside this magical display. I’ve always loved the large-scale dioramas in the Museum of Natural History in New York, and the artistry and life in that kind of display. I’ve always dreamt of a Looking Glass being used in this kind of context, and recently, I took the time do an exploration into bringing historical photographs into a 3D Looking Glass Diorama.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1TmwQpkpVoh_l4QOyqePQw.jpeg" /><figcaption>Illustration Credit: Nicc Balce, @robotnicc</figcaption></figure><p>To get a scene into a Looking Glass, you either need a 3D model of a scene, or you need pictures of the scene from more than one perspective. Not many people were making 3D models a hundred years ago, but as luck would have it, people have been shooting stereo photos all over the world since their invention in 1838, and there are countless stereograms from history that are publicly available now. The <a href="https://www.loc.gov/pictures/collection/stereo/">Library of Congress</a> has an extensive collection of stereograms that are in the public domain and are free to download, so I started exploring their collection. Let me just say, this is an *insane* collection — there are 29,500 stereo photos from all over the world, most over 100 years old. There are photos of joy and war, of dancers and animals, poverty and love, celebration and storms. Someone took a stereo photo of an alligator eating a person. This was a collection of bygone humanity, brought together from countless photographers around the world, covering a huge diversity of cultures and human experience. I was hooked.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/1*H2QCxwnlYFY0srtehDFRMg.jpeg" /></figure><p>I started sifting through the collection and pulling scenes into the Looking Glass. It’s just a delight to see a moment from 100 years ago brought back on me desk — I feel more connected to these people in the past, and they go from being black-and-white relics on a page to a lively miniature scene in front of me. It feels like I’m looking at real people, rather than pictures of real people.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/384/1*iubYaPNimDSxgupgRykLig.gif" /><figcaption>Men Smoking Hookah at a Tea House in Mosul, Iraq</figcaption></figure><p>To get a stereo card in to a Looking Glass, I had to first convert the left and right images to a depth map, a way to represent how far every pixel is from the camera. There are lots of useful tools for bringing a stereo photo into a Looking Glass — <a href="http://stereo.jpn.org/eng/">Stereo Photo Maker</a> and <a href="https://triaxes.com/products/3d-software/stereotracer/">Stereo Tracer</a> both let you just drag in a stereo photo and see it in a Looking Glass a few seconds later. The challenge, though, is working with old stereograms — there are artifacts in individual left and right images, and issues with contrast and exposure wreak havoc on stereo matching algorithms, like you can see in this conversion of an extremely charming ostrich ride in 1900s Paris:</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2Fr9CNZ395PNw%3Ffeature%3Doembed&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3Dr9CNZ395PNw&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2Fr9CNZ395PNw%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/9dd681b3f4af6208c93f828c1a8e238e/href">https://medium.com/media/9dd681b3f4af6208c93f828c1a8e238e/href</a></iframe><p>After doing a ton of conversions, I started looking for easier ways to convert old photographs. Stereo photos take some effort to convert — they are very sensitive to rotation, as well as any blemishes in the images, so I was spending a lot of time cleaning up the image before and after the conversion. Around this time, I had been experimenting with a recent google research project called the <a href="https://github.com/google/mannequinchallenge">Mannequin Challenge</a> that uses AI to generate a depth map from a single 2D image. This sped up the conversion work from minutes to a few seconds, and it started feeling less like technical work and more like exploring through the halls of an art or history museum. I put some of my favorite images below.</p><p>The more I look through these old photos, the more I’m drawn to them. One after another, I swipe through moments from the lives of people who lived more than a century ago. These flashes of memory are re-animated, brought back into the world with the help of this magical box on my desk. I love how delightful, relatable and familiar these miniature worlds are, as they drop, one by one, into my shoebox diorama that is overflowing with stories: my own personal, boundless museum.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/480/1*VsrfGb94pf4cif8-GSBFyw.gif" /><figcaption>Eskimo Children on Sled Pulled by Pet Bear, location unknown (1904)</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jJUaerXmskpsFe6KVUG6HQ.jpeg" /><figcaption>Hula Dancer, Honolulu, Hawaii (1923)</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/480/1*GDSh8nRTltFfb8SmxdNHUQ.gif" /><figcaption>Dancing on the Surf, Atlantic City, US(1987)</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*IgTcTg-u3dFpMH9BqcAQdA.jpeg" /><figcaption>Children playing in Peking, China (1902)</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bRHMt2PpydMiBuJN2KHdbA.jpeg" /><figcaption>Playing in the surf, location unknown (1906)</figcaption></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6be06384576c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[A Holographic Approach to Conservation]]></title>
            <link>https://medium.com/through-the-looking-glass/a-holographic-approach-to-conservation-ddf555e4dcdd?source=rss-70871fc45f5c------2</link>
            <guid isPermaLink="false">https://medium.com/p/ddf555e4dcdd</guid>
            <category><![CDATA[conservation]]></category>
            <dc:creator><![CDATA[Alex Hornstein]]></dc:creator>
            <pubDate>Wed, 05 Jun 2019 15:52:06 GMT</pubDate>
            <atom:updated>2020-02-01T21:27:33.593Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*vnFR7B6-Nr2CL4fCb07Sgg.jpeg" /><figcaption>A Cahow from Nonsuch Island, Bermuda, in a Looking Glass holographic display</figcaption></figure><p>We’re living in a time when a huge number of species across the planet are at risk of going extinct. As humans and our impact continues to expand across the planet, the viable habitats for some of the most vulnerable species shrink to tiny areas, small oases in the planet where the species can continue to survive. One of these species is the Cahow, or Bermuda Petrel, an extremely endangered, pigeon-sized seabird whose destiny is intertwined with a tiny spec of habitat called Nonsuch Island in Bermuda.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*KGaLUfATj59A3UZuF4HwMA.png" /><figcaption>different-scale maps showing Nonsuch Island, the only 14 acres on the planet where Cahows can breed</figcaption></figure><p>Nonsuch is a tiny island, about 14 acres in size. Its size and isolation are the saving grace for the Cahow — a bird that evolved before rats or people came to Bermuda and has the unusual ecological feature of nesting in holes in the ground for the five very vulnerable months that it takes to go from egg to fledged chick. Once people and rats came to Bermuda, the population of Cahows was driven into near-extinction, and now, after decades of effort, there are only 135 mating pairs of the bird in existence. All of them nest on Nonsuch and a few nearby rocks.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RExs6efnTcATD33zb_Pytw.png" /></figure><p>Cahows are in the Petrel family — birds that specialize in living in the middle of nowhere, and they don’t survive in captivity. When a Cahow chick fledges, it flies on its first attempt and doesn’t touch land again for five years. Nesting Cahows regularly travel from Bermuda to the Gulf of Maine to feed on their favorite squid, and they make the 1,500 mile round-trip in about five days. Scientists tracking the Cahow with GPS tags clocked the birds traveling at 100mph (with the wind at their backs, but still!) during the trip. These are birds that have evolved to survive at edge of the world.</p><p>That’s the fundamental challenge for Cahows — they like to live in the middle of nowhere, and there’s not much nowhere left. It’s only thanks to the work of a few dedicated conservationists that Nonsuch can stay isolated from the inexorable flow of human expansion, and remain a viable habitat for the Cahow.</p><p>This work is admirable, but how do you make people care about the Cahow? We learn about a lot of animals through zoos, but these birds aren’t meant for zoos. You can see them in the wild, but Nonsuch is not a big place, and you don’t <em>want</em> that many people coming out to Nonsuch — maybe a couple thousand people alive today have seen a Cahow in person, and if you tried to bring ten times that many people out to Nonsuch, you would drive the Cahows into extinction. This is where I came into the story.</p><p>I had built my first holographic video camera, and was presenting it at a show last year when I met JP Rouja, a Conservationist and Tech Developer who runs <a href="http://www.nonsuchisland.com/">Nonsuch Expeditions</a>, an organization that works to document and spread awareness of Bermuda’s unique biodiversity and the Cahows. I’m one of the co-founders of a holographic display company called <a href="http://lookingglassfactory.com/">Looking Glass</a>, and for the past nine months, I’ve been working on a number of techniques to capture digital holographic images and video of the natural world, that we can play back in our displays. Holograms present the illusion that something is there in front of you when in fact it’s not — it could be a recording, or a live scene being streamed from another part of the world. I am particularly interested in capturing holographic video of the natural world, and had been working with various conversationist organizations to film different animals, large and small, with cameras that I designed. JP had actually already built infrastructure on Nonsuch to stream <a href="http://www.nonsuchisland.com/live-cahow-cam/">live 2D footage of the Cahows</a> in their nests, and had cultivated a large community of Cahow fans the world over. Over several conversations, JP and I worked out a plan to take the live presence setup to the next level by live-streaming holographic footage of Cahows from their natural habitat on Nonsuch out into the world, where they could be viewed, life-sized, in our displays.</p><p>A brief note on holographic cameras — this isn’t something that you can buy. If you want one, you have to build the hardware and software to operate it. Holograms are all about capturing a scene from many different perspectives, simultaneously, and Cahow nests are one of the most challenging places to film. They are small, pitch-black holes in the ground, and you can only film the Cahow chicks in infrared. There is a limited window in time where we can install a camera in a nest, when there are no Cahows nesting, and then it just has to work. Perfectly. Over the course of four months, I designed and built a prototype holographic Cahow-cam that would be capable of capturing holographic footage of a Cahow chick.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*zWzJBwjYkixVU4v_gZpA_Q.jpeg" /><figcaption>Testing the Cahow cam in the lab — that little black strip in the front is actually 32 separate cameras, rolling simultaneously.</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yHtur_Z_hTvAvlpZeR1qMw.jpeg" /><figcaption>Cahow cam, capturing footage of a chick during an ornithologist’s health check</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*lf2cvB-5e9qkJd3VBswr7Q.jpeg" /><figcaption>Shooting holographic B-roll of scientist Jeremy Madeiros doing a health check on a Cahow chick</figcaption></figure><p>Once the camera was ready, I brought it down to Bermuda and Nonsuch Expeditions took us out to Nonsuch Island to shoot the first holographic footage of the Cahow. Over the course of several days, I worked with JP and the scientists working there to film Cahow chicks. For this first experiment, we shot test footage in a lit, artificial burrow while the scientists were performing routine health checks on the Cahow chicks, and then returned the chicks to their nests.</p><p>After processing the footage, we were able to do what we set out to do — to show a life-size, holographic replica of a Cahow in one of our displays. And it looks wonderful.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/480/1*SEvp6qM-A62ZGt6HTvo7Vw.gif" /><figcaption>holographic Cahow chick in its holographic nest</figcaption></figure><p>This is just the beginning, of course — a proof of concept of a third way to represent wild animals in human environments. It is my hope that this can develop further into a range of techniques for filming and viewing wild animals in their natural habitat, as an alternative to viewing animals in captivity or bringing large numbers of humans out to the fragile ecosystems where these animals live. I want to see this concept grow out into the world, and to continue working with conservationist partners like Nonsuch Expeditions to film and live-stream more species in the future. For an island that is unique in its ability to hatch extraordinary things, I’m thrilled to be working on Nonsuch to bring one more extraordinary idea out into the world.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*L3aa_-IdGRs0SVQ0DfCm7A.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ddf555e4dcdd" width="1" height="1" alt=""><hr><p><a href="https://medium.com/through-the-looking-glass/a-holographic-approach-to-conservation-ddf555e4dcdd">A Holographic Approach to Conservation</a> was originally published in <a href="https://medium.com/through-the-looking-glass">Through the Looking Glass</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[History, Volume I]]></title>
            <link>https://medium.com/@alexhornstein/history-volume-i-7fead6ba48a0?source=rss-70871fc45f5c------2</link>
            <guid isPermaLink="false">https://medium.com/p/7fead6ba48a0</guid>
            <category><![CDATA[3d-printing]]></category>
            <dc:creator><![CDATA[Alex Hornstein]]></dc:creator>
            <pubDate>Sun, 15 Jul 2018 21:34:54 GMT</pubDate>
            <atom:updated>2026-01-15T00:23:48.935Z</atom:updated>
            <content:encoded><![CDATA[<p>Whoever said that necessity is the mother of invention wasn’t trying to make a volumetric display. No, panic was the mother of this invention, panic and a deep desire to see a science-fiction fantasy become reality. And this is the story of the invention and of the last two years of designing, prototyping, testing, and occasionally panicking.</p><p>Flashback to three years ago: Shawn Frayne and I are running independent invention labs in Hong Kong and Manila. We’re working together on machines that make low-cost solar panels for rural, unelectrified areas. The work is fascinating, but it doesn’t exactly pay the bills, so we spend a third of our time doing contract invention for a Fortune 500 company, and that pays for everyone’s time. It’s not just us, too — Shawn has four other people working his lab, I have two in mine. This arrangement had been working well for a couple years, and gave us a stable way to work well into the futur…oh, what’s that? The fortune 500 company is having financial problems and just gave us a month’s notice that they’re cutting off our contract? Cue the panic!</p><p>Panic gets a bad rap — most people think of panic as a disorderly, emotional mess, but panic over the future, looking down the gun barrel of onrushing financial disaster, losing an extremely talented engineering team that you’ve worked with for years and the increasingly real possibility that you’re going to have to go back home to live in your parents’ basement, or worse, get a *job* — that kind of panic is like a laser beam for the mind. Our reaction to panic is to try to invent our way out — if we don’t like the hand we were dealt, invent a new game, then convince a bunch of people to come play the new game rather than the old one with the crappy cards.</p><p>As I mentioned, in the last hand we were holding, we had spent a lot of time working with low-cost solar panels. There’s a step in making solar panels where you bond a protective layer of glass to the fragile silicon solar cells. The sturdy glass protects the silicon, but it also raises a problem — how do you keep sunlight from bouncing off of the glass and missing the solar cells? There’s an optical trick used in every solar panel in the world to turn this glass sandwich completely clear, letting light pass through as if the sandwich weren’t there at all. That trick is called index-matching.</p><p>Somewhere in there, the lightbulb started firing. A trick that makes glass disappear!</p><p>[…. intro intro intro, make it more interesting]</p><p><strong>Part 1, Volumetric Printing</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*YX6ejNBUq4NGXqOkimcQVQ.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/350/1*YXZMRSIZ_Khc3b-d4iTHDw.gif" /></figure><p>Three years ago, we started experimenting with a new idea. We thought we could bring scenes from a virtual world into the physical world by slicing up the 3D scene like slicing an egg with an eggslicer, and then printing each slice onto a sheet of thin, clear plastic. By stacking up the sheets of plastic, we could reconstruct the 3D scene in the real world.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/831/1*9IPWh174WWgvyNX66j2fZw.gif" /></figure><p>Getting these prints to work was tough, and it took us the better part of a year to perfect. Half of the problem was the software. Slicing like this takes smooth, curved surfaces like a person’s face and turns them into discrete slices, and if we don’t compensate for this effect, the prints will look like they’re made of legos.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/639/1*3l1pk8oLE3ZJiimzc9WNKA.jpeg" /></figure><p>A big part of making prints look good was learning to blend between different layers. Over months of experimenting with software techniques and prints, we were able to bring some information from the neighboring slices onto a given slice, and to “feather” the edges of the neighboring information by making geometry farther from the slicing plane more transparent, and geometry closer to the slicing plane more opaque. This blending sm0othed out the lego-like appearance, making surfaces look curved and realistic in the prints.</p><p>For anyone interested in the ins and outs of slicing color 3D content, our original slicing software is open-source and free for download <a href="https://github.com/enjrolas/glasscutter">here on github.</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/300/1*rrT0MFy7yiVcMJgjXUt3-w.gif" /></figure><p>The other big part of the trick is a technique called index-matching — imagine stacking up twenty overhead transparencies. If you look at them, they don’t look clear, they actually only let a tiny bit of light through, and they look milky and mostly opaque. This is because every time that a ray of light travels through a boundary between two materials with different indices of refraction, some of that light is reflected back, and is refracted, passing through the boundary at a slightly different angle. Even if you only lose 5% of the light at every boundary, after going through 20 sheets (with 40 boundaries between plastic and air), less than 13% of the light will still pass through. Index matching means that we fill in all the air gaps with a material with the same index of refraction as the plastic, and as far as the light is concerned, it’s just one big block of clear plastic. As luck would have it, one of the clearest plastics available is acrylic, and it has an index of refraction very close to mineral oil, or as your local pharmacy calls it, baby oil.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/922/1*hcEA_qVGLCiQYhWnPJOt6g.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/922/1*pmAdKf_Dk1ZVLZpdnwpD5w.png" /></figure><p>Part 2, Moving Pictures</p><p>The prints were good, but they weren’t a great business, and more importantly, they were static, and that meant that they weren’t the sci-fi hologram of our dreams. After a year of making prints, we decided to scrape together all the cash we could spare and try to make a dynamic volumetric display.</p><p>A volumetric display is basically what we think of when we think of a hologram in a sci-fi movie. It’s a true, three-dimensional volume that’s full of pixels, or as us 3D nerds love to call ’em, <em>voxels</em>. The concept is just like how we recreated a 3D scene in the prints by stacking up a bunch of slices, but instead of dots of ink in a volume, we’d need to make points of light in a volume.</p><p><strong>take 1, swept-volume displays</strong></p><p>We did the first thing that everyone does when they try to make their first volumetric display: we got a high-speed projector (we used one of TI’s lightcrafter development kits, which can show 1-bit color depth images at 1,200 hz) and we wobbled a plane back and forth in front of the projector. At every different position of the plane, we’d project a different image, like a CAT scan in reverse. The general term for this type of display is a <em>swept-volume</em> display, because you’re “sweeping” an imaging surface through the display volume, drawing voxels as you go.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*hmq04xNXhWik8f11Ka41zA.gif" /></figure><p>This worked! Here’s our very first volumetric display prototype, showing a smiley face and not syncing at all between the motor and the projector.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fplayer.vimeo.com%2Fvideo%2F189589094&amp;url=https%3A%2F%2Fvimeo.com%2F189589094&amp;image=https%3A%2F%2Fi.vimeocdn.com%2Fvideo%2F599961016_295x166.jpg&amp;key=d04bfffea46d4aeda930ec88cc64b87c&amp;type=text%2Fhtml&amp;schema=vimeo" width="320" height="240" frameborder="0" scrolling="no"><a href="https://medium.com/media/43a6145c305c96c50d714d657fdfba05/href">https://medium.com/media/43a6145c305c96c50d714d657fdfba05/href</a></iframe><p>That jackhammer sound is a feature, not a bug. It turns out that it’s fairly loud and noisy to move a large object back and forth several inches, 20 times a second. But worse than that, the high-speed projectors were super expensive. We firmly believed that if we were going to succeed, we had to make a display that was cheap enough for people to get and tinker with. That meant that we couldn’t use a $1,000 projector as a component in our product.</p><p>We started looking for the highest framerate displays we can find. There really weren’t any cheap &gt;1000 frames per second displays that we could find, so we started playing with making them ourselves. We made a quick version of replacing the high-res, expensive projector with a low-res, cheap 30x30 pixel LED screen.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fplayer.vimeo.com%2Fvideo%2F189589093&amp;url=https%3A%2F%2Fvimeo.com%2F189589093&amp;image=https%3A%2F%2Fi.vimeocdn.com%2Fvideo%2F599961032_960.jpg&amp;key=d04bfffea46d4aeda930ec88cc64b87c&amp;type=text%2Fhtml&amp;schema=vimeo" width="1080" height="1920" frameborder="0" scrolling="no"><a href="https://medium.com/media/4002d1cac6ca5aa50241f6d01a60c246/href">https://medium.com/media/4002d1cac6ca5aa50241f6d01a60c246/href</a></iframe><p>This also worked, but all that circuitry to make a high refresh-rate LED screen was big and bulky and is hard to get small and cheap and fast, and we didn’t want to re-engineer an existing 2D displays if we could avoid it. While pondering this can of technological worms, we had another idea: <em>what if we could take an off-the-shelf high-res, low frame-rate projector and add a mechanism to turn it into a low-res, high frame-rate projector.</em></p><p><strong>iiintroducing lightfolding!</strong></p><p>This idea felt really powerful. It meant that we could avoid the expense of a fancy high-speed projector and the engineering slog of making a custom high-speed 2D projector or screen. We called this idea <strong>lightfolding</strong>, and that idea of “gearing” between 2D resolution and depth resolution became a common theme in our thinking.</p><p>The first lightfolding prototype was made of toilet paper tubes, mirrors and hot glue. We made a kind of rotating periscope that would spin around in a circle and “grab” images from a projector that were projected along the circumference of the circle and spit them out the same spot in the center of the circle. Those images along the circumference were high-speed frames, and this mechanism was the first simple add-on that could turn a regular, el-cheapo projector into the high-speed projector that we needed to make a swept-volume display work.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*pcAQflljJy_uK5wBVjgh3w.gif" /></figure><p>The first design was just a proof of concept, though — to make a working display, we couldn’t have the mirrors move past the projected image — that would mean that the high-speed image would have motion blur. We decided instead to make a system with twenty pairs of mirrors that would catch twenty high-speed frames within a single projected image. We called this monster the Stargate.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fplayer.vimeo.com%2Fvideo%2F189591626%3Fh%3D73a8264f69%26app_id%3D122963&amp;dntp=1&amp;display_name=Vimeo&amp;url=https%3A%2F%2Fvimeo.com%2F189591626&amp;image=https%3A%2F%2Fi.vimeocdn.com%2Fvideo%2F599964461-a9d1bdc66d5b83ccb9e894fac01acfd0c1406c8b1ecb03e27225edb3fbfdca7a-d_1280&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=vimeo" width="1280" height="720" frameborder="0" scrolling="no"><a href="https://medium.com/media/5b58179078d18b3cc3e5dab9595c7e2d/href">https://medium.com/media/5b58179078d18b3cc3e5dab9595c7e2d/href</a></iframe><p>In the stargate display, a regular projector shines an image with twenty slices arranged in a circle. Those twenty projected slices fall onto twenty pairs of mirrors, which redirect each slice towards a unique plane at the center of the circle. A giant subwoofer shakes a piece of vellum back and forth a couple inches, scattering the light from each projected plane. And finally, there’s a spinning shutter that only allows light from one slice through at a time. I wrote some software that controlled the shutter’s position and generated an audio signal for the subwoofer that kept their relative position in sync. This whole idea was exactly as complicated as it sounds, and the machine was a huge beast with a tiny 1&quot;x1&quot;x1&quot; display volume, but it was our first full-color dynamic volumetric display. It was awesome! We looked at this pile of wires and hot glue and occasionally smoldering electronics and we saw the future. And that made us unique.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*AHLGwSuXaFBESe465DOsNw.gif" /></figure><p>The complexity of the Stargate scared us, though — we figured that we would need a couple years to improve the system to the point where it could be reliable enough to turn into a product. Fast, mechanically oscillating planes are really tricky, and electrically synchronizing the phase of two high-speed mechanica oscillators (the display plane and the shutter) is tricky. If we were going to make this system work, we’d have to get rid of some of the moving parts. We’d need that time to iterate through hardware designs, and to start getting a software pipeline that could take a world in a virtual 3D environment, slice it up, and draw it in our display.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*qQ30ByKcVZ57PHc4kO8h8w.jpeg" /><figcaption>Shawn with Stargate at 4am in the Hong Kong lab</figcaption></figure><p>The prospect of two years of difficult invention sounded awesome, but one thing stuck in our craw — what if we spent the next two years on this idea, but we were off the mark? What if we misread the holographic zeitgeist and made something that didn’t capture the sci-fi dream? Two years is a lot of time to waste. We wanted a way to prove to ourselves as much as anyone that there were people out there who were interested in what we were doing.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1XbbIrhpkgXtVfROE235JA.gif" /></figure><p>It made us take another look at an idea that we had discarded earlier: LED cubes. LED cubes are a type of volumetric display with an LED at each voxel. They’re great because they’re relatively easy to build by hand and have no moving parts. They’d been popular among makers and electronics hobbyists for years. LED cubes have a fundamental flaw, though — because they have an LED at every voxel, the cost scales with the cube of the resolution. Even if you could get a color LED at the insanely cheap price of $.01, a modest resolution of 100 x 100 x 100 voxels would cost you $1000 in LEDs alone! However, we thought, maybe we were being too insistent on high-resolution. People on the web and at maker faires loved LED cubes — they’d spend hundreds of hours building and programming their cubes, and would share it on youtube or their personal site. At the time, everyone made LED cubes by taking through-hole LEDs, bending the leds and soldering them to neighboring LEDs to form a cubic structure. Everybody would engineer their cube differently, which was awesome, but also meant that nobody could share code between cubes because they all had different hardware. And more than anything else, we were impressed by the hundreds of people we kept finding who were so into LED cubes that they would spend this huge effort making them, because it was a kind of 3D display that they could have and play with right now.</p><p>Our lead electrical engineer Samtim picked up a pencil and sketched out a design that would dramatically simplify the construction of an LED cube. Instead of manually soldering leads of LEDs, we could use traditional automated pick-and-place assembly to place a series of LEDs onto thin, reed-like circuit boards, and then we’d plug a bunch of those reeds into a motherboard to make a 3D array of LEDs. Instead of using “dumb” LEDs and separate driver circuits, we’d use a new, popular variety of “smart” LEDs with integrated drivers, so we could just pass them an RGB color value and they would take care of turning that color, saving us a huge mess of circuitry and wiring. Robots would do all the soldering, and instead of 200 hours of assembly, it would take half an hour to put one together. Samtim opened up Eagle, designed a circuit board, and a week later, we were testing our first LED cube.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/942/1*s91A1Gwn_aXijZ5v6iLWwA.gif" /><figcaption>L3D Cube Assembly</figcaption></figure><p>This was my first time working with these smart LEDs, and I was a little bit dumb about it. We had designed an 8x8x8 LED cube, with 512 LEDs that each could draw 60mA at 5V, so I multiplied those numbers out and decided that I needed a power supply that could supply an insane 30 Amps at 5 Volts. Foolishly, I bought one and hooked it up to our first prototype cube. I wrote some code that should turn all the LEDs a solid color, programmed it, and…yikes! 150W of LED light is like having a couple streetlights on your desk on full blast, shining at your face. We all immediately fled to the other side of the room and fashioned little eskimo snow goggles out of crap lying around the lab and squinted at this weird concoction through pinholes. After a moment, I turned the power supply off and on again, and the massive surge in current caused the supply’s voltage to oscillate, which promptly blew out the entire cube in a fell swoop. On to prototype number two!</p><p>Later, we got a bit smarter and realized that we could current-limit the LEDs without them complaining. We made seven revisions of this design, but they were all very similar to the first cube we made.</p><p>Shawn had a smart idea in this process. Rather than pushing for months to perfect the cube and make a video and launch a kickstarter, we should just take some photos and put it up on amazon to see if anyone would buy it. At the time, we had exactly one cube. We put it up, and about thirty minutes later, someone bought it! We were stunned — how on earth did they even find us? Surprised and excited, we put our one cube in a box, mailed it to the customer, and quickly started making several more.</p><p>Within a couple weeks, we had sold twenty cubes and we were working as hard as we could to keep producing the cubes. We figured that the experiment was complete — there were plenty of people out there who were interested in LED cubes. We pulled the cubes off Amazon and started preparing for a more complete launch on Kickstarter. A couple months later, we launched <a href="http://kickstarter.com/projects/lookingglass/l3d-cube-the-3d-led-cube-from-the-future">our Kickstarter campaign</a>, and for our modest company, it was a smash hit.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fplayer.vimeo.com%2Fvideo%2F112570393&amp;url=https%3A%2F%2Fvimeo.com%2F112570393&amp;image=https%3A%2F%2Fi.vimeocdn.com%2Fvideo%2F497590455_1280.jpg&amp;key=d04bfffea46d4aeda930ec88cc64b87c&amp;type=text%2Fhtml&amp;schema=vimeo" width="1280" height="720" frameborder="0" scrolling="no"><a href="https://medium.com/media/1036f78d730537a633faeadbc0d0c062/href">https://medium.com/media/1036f78d730537a633faeadbc0d0c062/href</a></iframe><p>The L3D cube was great as a hobbyist kit, and we started seeing some really talented people building and sharing some awesome programs for the cube. It was great to see so much enthusiasm over volumetrics, but the 512-voxel L3D cube was still far too low-resolution to feel like the sci-fi dream of holograms. An interesting side effect of a successful launch was that we started making money from selling kits, enough to keep pushing on R&amp;D to make the dream real.</p><p>Part 3, More Voxels</p><p>Putting a single LED at every voxel is a dead-end for a cheap volumetric display — the law of cubes is a harsh master. We started a hunt for cheaper 2D pixels that we could map into 3D space. The cheapest 2D pixels in the world are on an LCD screen — thanks to mobile devices, we can get millions of pixels in a small screen for tens of dollars. After LCDs, a common source of cheap pixels is pico projectors, small handheld projectors that can project millions of pixels. They’re not as cheap as LCDs, but the projection is very handy when it comes to pulling pixels up into the third dimension.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*MUO8xxxinMbeP6CP_Mtu9g.gif" /></figure><p>We started experimenting with an idea that we called Nata de Coco, after the tropical drink with small chunks of translucent coconut flesh floating inside. We made a static Looking Glass print with an array of milky blocks carefully scattered throughout the volume of the print and we pointed a pico projector at the print. Each pixel of the pico projector lit up a certain block, mapping that pixel to a location in 3D space. By turning on a certain pixel in the projector, we could light up that 3D voxel in the print.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ILpjUTzE7gyFvhr0wwrA-Q.jpeg" /></figure><p>This technique was promising — we could use our unique printing technology to precisely control the arrangement and level of scattering of the blocks in our volume. We were quickly able to get to a comparable level of resolution as our largest LED cube — 16x16x16 voxels.</p><p>We had LEDs on the brain, so our thoughts started moving towards a territory favored by cheapskates and motor junkies: swept-volume persistence of vision displays.</p><p>The idea here is that LEDs are really bright and can turn off and on really fast. If you can somehow move an LED around quickly — faster than the eye can perceive, you can paint multiple points in space with a single LED, saving you money. Moving LEDs back and forth is tricky, mechanically speaking, but spinning LEDs around in a circle was just simple enough that it might work. We hatched on an idea to take just 256 LEDs — half of the number that were in the L3D cube — and spin them around really quickly, getting more than a hundred times more voxels out of the same LEDs. We called this concept the Hypertube.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*n2nKzc_P_HBYd9IaDPQaLw.gif" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/843/1*k07thX3CoCL775NSMa8Zeg.gif" /></figure><p>Our first design had over 50,000 voxels. We had a number of challenges to get it working — we had to get power to a circuit board spinning more than 20 times a second, we had to synchronize the LED display with the position of the LEDs. We had to build a whole host of new mechanical and electrical hardware, firmware and software, and run through <a href="https://vimeo.com/190470161">a bunch of tests to get things calibrated.</a> But once we did, well, the results were pretty awesome.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fplayer.vimeo.com%2Fvideo%2F190470216&amp;url=https%3A%2F%2Fvimeo.com%2F190470216&amp;image=https%3A%2F%2Fi.vimeocdn.com%2Fvideo%2F601138811_295x166.jpg&amp;key=d04bfffea46d4aeda930ec88cc64b87c&amp;type=text%2Fhtml&amp;schema=vimeo" width="568" height="320" frameborder="0" scrolling="no"><a href="https://medium.com/media/7795dfa0406c8653bf06c2f8ede78c23/href">https://medium.com/media/7795dfa0406c8653bf06c2f8ede78c23/href</a></iframe><p>The challenging thing with spinning prototypes is getting video data to a spinning display. There isn’t an easy way to do this, so in the first prototype, we pre-loaded volumetric data onto an SD card, and an on-board processor read the data and sent it out to the LEDs. Getting live volumetric data from a computer or other source would need a new architecture.</p><p>While we were talking about a new prototype, I wrote a simple volumetric simulator to look at <a href="https://vimeo.com/190470161#t=1m32s">a volumetric scene at various voxel resolutions</a>. After playing around with the simulator, we came up with a test that we called the <em>sheen test</em> — in this test, we took a 3D scan of Charlie Sheen, pulled it into the simulator and showed it to a stranger. If they said, “oh man! That’s Charlie Sheen!” then the resolution passed the sheen test. If not, it wasn’t high enough resolution to recognize a person. By trial and error, we found that the minimum voxel resolution to recognize a person’s face was around 250,000 voxels.</p><p>The thing keeping our resolution low was the LEDs. In the first prototype, we used LPD8806 LED drivers that have a 4,000 Hz refresh rate. That’s enough to show 200 radial slices at a 20Hz volumetric refresh rate. The limited our resolution. Additionally, the LPD8806 driver chips were physically large, and that limited our resolution. After a bunch of searching, we decided that the way to get the fastest and tightest pitch LEDs was to make our own LED screen using the smallest RGB LEDs in the world and some custom LED drivers. We built the fastest LED screen in the world, with 64x64 LED pixels that could refresh at 12,000 Hz at 1-bit color depth.</p><p>To get real-time data to the LEDs, we had to look at high-bandwidth wireless links. A local wifi network has the bandwidth to get the full volumetric data across, if we apply lossless compression to the data. To handle receiving wifi data and sending it to the screen, we put a Beaglebone Black, a small linux computer with a wifi interface onto the back of the LED screen. The sheer weight of the components that we had to spin around twenty times a second was starting to terrify us, so we put the whole thing into a shatterproof polycarbonate housing that would theoretically protect us if the proverbial shit hit the not-so-proverbial fan.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fplayer.vimeo.com%2Fvideo%2F190499707&amp;url=https%3A%2F%2Fvimeo.com%2F190499707&amp;image=https%3A%2F%2Fi.vimeocdn.com%2Fvideo%2F601178036_1280.jpg&amp;key=d04bfffea46d4aeda930ec88cc64b87c&amp;type=text%2Fhtml&amp;schema=vimeo" width="1280" height="720" frameborder="0" scrolling="no"><a href="https://medium.com/media/2c2eeb11f0d4a634e70bec23224f7b97/href">https://medium.com/media/2c2eeb11f0d4a634e70bec23224f7b97/href</a></iframe><p>Our first test with this new display looked pretty good, even at 3am on a cold and rainy morning in Zhong Mu Tou, but mannnn it was hard to get working. All the infrastructure in this version of the Hypertube was extremely difficult — we had even more layers of technology: software, DMA-level code on the beaglebone, and firmware on the LED driving chips, and all of it had to play together at very high speeds, and of course, spinning around twenty times a second.</p><p>The complexity of this design path was its undoing. At the time, we only had two people on the team who could write software or firmware: myself and Samtim. Even with the help of some extremely talented contractors, we found ourselves looking at a task on par with re-engineering a graphics card: we had to break down a 3D scene into raw voxel data, compress it, send it over wifi to our tiny spinning linux computer, decompress it, send the voxel data out to our LED driving microprocessors, and clock that data out to the LEDs. Yikes!</p><p>One thing that often shows up in the evolutionary progression of species is that a slow, complicated species get out-competed by simpler, more adaptable species. The same thing happens in invention. While we were slogging through engineering the complex dinosaur of the Hypertube, one of the other engineers on the team, Alvin, quietly did a little test that turned out to be a mammal. He called it the Hypercube, and we’ll get into that once I get around to writing part 2,</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7fead6ba48a0" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>