<?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 Yunqi Guo on Medium]]></title>
        <description><![CDATA[Stories by Yunqi Guo on Medium]]></description>
        <link>https://medium.com/@yunqiguo?source=rss-150c15c2183------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*pB8dVAUtrVFw-hIn.</url>
            <title>Stories by Yunqi Guo on Medium</title>
            <link>https://medium.com/@yunqiguo?source=rss-150c15c2183------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Mon, 16 Nov 2026 18:29:53 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@yunqiguo/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[Why Spatial Photos and Videos Offer More Immersion than 3D]]></title>
            <link>https://medium.com/@yunqiguo/why-spatial-photos-and-videos-offer-more-immersion-than-3d-39432f8b36f9?source=rss-150c15c2183------2</link>
            <guid isPermaLink="false">https://medium.com/p/39432f8b36f9</guid>
            <category><![CDATA[immersive]]></category>
            <category><![CDATA[ar]]></category>
            <category><![CDATA[spatial-computing]]></category>
            <category><![CDATA[apple-vision-pro]]></category>
            <category><![CDATA[apple]]></category>
            <dc:creator><![CDATA[Yunqi Guo]]></dc:creator>
            <pubDate>Sat, 15 Jun 2024 17:34:37 GMT</pubDate>
            <atom:updated>2024-06-15T17:34:37.054Z</atom:updated>
            <content:encoded><![CDATA[<p>The release of Vision OS 2 offers advanced spatial photo and video experiences, allowing users to convert 2D images into spatial photos. Capturing these photos is now not exclusive to the iPhone 15 Pro and Vision Pro.</p><p>HEIC images for spatial photos still store the left/right eye views but do not store depth/LiDAR information. However, Vision Pro offers a more immersive experience than traditional 3D. Here are two techniques, in my opinion, that Apple uses to deliver this experience.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*qUR7dj_1uvgATq7MnjXnDA.png" /><figcaption>Image content for spatial photo</figcaption></figure><h3>1. The Window Frame Concept in Spatial Photos and Videos</h3><ul><li><strong>Unique Attribute</strong>: When the viewer’s head moves, the content within the frame also shifts.</li><li><strong>3D Window Simulation</strong>: This movement simulates the experience of looking through a 3D window, making the content feel more lifelike.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*8Y-9Ipy6n3fGqZbDI2CJ9Q.png" /><figcaption>Image from <a href="https://developer.apple.com/videos/play/wwdc2024/10166/">WWDC24</a></figcaption></figure><p>To understand this better, let’s recall the difference between viewing through a window frame and a movie screen:</p><ul><li><strong>Window Frame</strong>: When you look through a window frame, the scene changes as you move your head, altering your perspective.</li><li><strong>Movie Screen</strong>: When observing a picture on a movie screen, the image remains static regardless of your movement.</li></ul><p><strong>With Vision Pro, the frame and the 3D image are set apart. This illusion tricks our brain into treating the frame as a window frame</strong>, <strong>making the objects behind it appear more like 3D, even though each eye is still seeing a 2D image. </strong>This enhances the depth and realism of the content, providing a more engaging visual experience.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/400/1*iMlyJhse_siDlWnfuLOLVw.gif" /><figcaption>Window frame for the spatial content</figcaption></figure><h3>2. Varying Fields of View (FoV) for Different Spatial Content</h3><ul><li><strong>Daily Life Adaptation</strong>: We are accustomed to the size and field of view (FoV) of objects in our daily lives. This natural adaptation plays a crucial role in how we perceive spatial content.</li><li><strong>Sensitivity to FoV in 3D Space</strong>: Unlike viewing through a 2D screen, our eyes are highly sensitive to the sizes of objects in 3D space, especially when the objects are not framed.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/536/1*F2gM56o_zmJ8-eJ1Qez2kg.png" /><figcaption>Image from <a href="https://www.pcgamingwiki.com/wiki/Glossary:Field_of_view_%28FOV%29">Glossary: Field of view (FOV)</a></figcaption></figure><p>Spatial photos and videos exploit this sensitivity to create a more immersive experience. Vision Pro enhances this experience by using the FoV information stored in the photo metadata to display the image or video in its original FoV during playback. This attention to detail ensures that the content appears as it would in real life, further enhancing the sense of immersion.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4tK0INbZ8R9IyNRl9rTDSQ.png" /><figcaption>Displaying image captured with a 15mm (Horizon 100°) lens</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*FVZZuIashu4yx8gcVP8vOA.png" /><figcaption>Displaying image captured with a 35mm (Horizon 55°) lens. For both images, the viewer stays at the same place. The sizes are automatically adjusted by Vision Pro.</figcaption></figure><p>Note that these two tricks can function independently. The window frame trick is applied in the window view, while the corresponding FoV is used in the immersive view. However, these techniques can also work together. In the case of panorama photos and Apple Immersive movies, both tricks are utilized.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7DVI2K_ZJtNjmv3ZEHknlQ.png" /><figcaption>Panorama photo on Vision Pro (Image from <a href="http://apple.com/">apple.com</a>)</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*gG1LO41WziXzLHhxStHfjA.png" /><figcaption>Apple Immersive (Image from <a href="http://apple.com/">apple.com</a>)</figcaption></figure><p>With these two tricks, Apple Vision Pro’s spatial photos and videos offer a more immersive and realistic experience compared to traditional 3D content. This could be an exciting development for both creators and consumers.</p><h3>Reference</h3><ol><li><a href="https://developer.apple.com/documentation/ImageIO/Creating-spatial-photos-and-videos-with-spatial-metadata">Creating spatial photos and videos with spatial metadata</a></li><li><a href="https://medium.com/antaeus-ar/what-is-spatial-video-on-iphone-15-pro-and-vision-pro-e20ac2b957f">What Is Spatial Video On iPhone 15 Pro And Vision Pro</a></li></ol><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=39432f8b36f9" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Why We Keep Sponsoring Accessibility in Tech Conferences]]></title>
            <link>https://medium.com/@yunqiguo/why-we-keep-sponsoring-accessibility-in-tech-conferences-d217fc8a2530?source=rss-150c15c2183------2</link>
            <guid isPermaLink="false">https://medium.com/p/d217fc8a2530</guid>
            <category><![CDATA[non-profit-organization]]></category>
            <category><![CDATA[research]]></category>
            <category><![CDATA[accessibility]]></category>
            <category><![CDATA[sponsorship]]></category>
            <category><![CDATA[deaf]]></category>
            <dc:creator><![CDATA[Yunqi Guo]]></dc:creator>
            <pubDate>Sat, 15 Jun 2024 06:49:19 GMT</pubDate>
            <atom:updated>2024-06-15T06:54:13.336Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*UPl8Qq1ORq79u0jKPdgm0g.jpeg" /></figure><h3>What We Did</h3><p>In line with our commitment to accessibility, our team, <a href="https://anysign.net/">AnySign</a>, proudly sponsored <a href="https://sigmobile.org/mobicom/2023/">MobiCom’23</a> and <a href="https://cps-iot-week2024.ie.cuhk.edu.hk/">CPS-IoT-Week’24</a> in accessibility. We provided sign language interpreter support for deaf presenters at these prestigious tech conferences. This ensured clear communication between the tech community and the deaf attendees and presenters.</p><h3>Why We Do It</h3><h3>The Need for Sign Language Interpreters</h3><p>Conferences featuring deaf audience members or presenters require interpreters. This is both a legal requirement under the Americans with Disabilities Act (ADA) and crucial for effective communication and idea sharing. Imagine attending a conference in a language you do not understand; without an interpreter, you would miss out on valuable content. Similarly, a deaf individual faces significant challenges without the presence of a sign language interpreter. For the audience, understanding a deaf presenter without interpretation is equally difficult.</p><h3>Regional Challenges</h3><p>Sign language translation across different regions presents a unique challenge. Each region and its respective deaf communities have their own unique sign languages. For instance, the US and the UK, despite being English-speaking countries, use American Sign Language (ASL) and British Sign Language (BSL), respectively. This diversity in sign languages complicates the task of finding suitable interpreters.</p><h3>Real-World Challenges</h3><p>Our experiences underscore these challenges. At MobiCom 2023, the committee had difficulty finding the ASL interpreter for the conference in Europe. Similarly, for CPS Week 2024 in Hong Kong, despite extensive efforts, none of the registered sign language interpretation organizations in Hong Kong could provide ASL interpreters. Additionally, many tech conferences are not yet fully aware of accessibility requirements and needs, often leading to budget constraints that hinder the provision of necessary accessibility services.</p><h3>How We Made It</h3><p>Despite the difficulties, we succeeded in providing interpreters for the sessions. To our knowledge, we supported the first Deaf presenters at these conferences, even though papers discussing deaf and sign language support have been featured for decades.</p><p><strong>Our success hinges on our collaboration with the community and Deaf teams.</strong> For MobiCom’23, my friend Issa introduced us to our interpreter, Zach. In Hong Kong, our interpreter, Jenny, was introduced by our friend Melody’s family. <strong>[<em>Ad: If you’re in LA, you can visit their pizza shop, </em></strong><a href="https://www.instagram.com/pi00ala/"><strong><em>@Pi00a</em></strong></a><strong><em>. They serve a fantastic fusion of Eastern and Western flavors.</em>]</strong></p><h3>The Vision and Mission of AnySign</h3><p>Our mission at AnySign is to fill the gap between tech research teams and accessibility users. There has long been a disconnect between accessibility research and actual users, leading to trust issues within the Deaf communities. To address this, we established AnySign, an interdisciplinary team dedicated to bridging this gap. We convey the actual needs of users to academic research teams and actively promote the significance of communication channels. While there is still a long way to go, we remain committed to this mission.</p><h3>Plan</h3><p><strong>Work with More Groups:</strong> Our plan involves collaborating with official groups and increasing involvement from various countries and schools. By broadening our network and fostering collaboration, we aim to enhance accessibility at tech conferences globally.</p><p><strong>Promote Cross-Region Sign Language Interpretation:</strong> As cross-region communications become increasingly intensive, we are exploring ways to support cross-region sign language interpretation. This includes:</p><ul><li><strong>Supporting individuals who know multiple sign languages (primarily deaf individuals).</strong></li><li><strong>Producing sign language textbooks in different languages</strong>, such as a Chinese version of the American Sign Language (ASL) Dictionary and an English version of the Chinese Sign Language (CSL) Dictionary.</li></ul><p>This initiative is still in the planning stages, and we welcome your comments and opinions.</p><h3>Key Idea in Our Workflow: User in the Loop</h3><p>Ensuring that users are involved at every stage of our workflow is central to our approach. This user-centric focus helps us stay aligned with the actual needs and challenges faced by the communities we aim to serve.</p><p>Thank you for reading. We hope this blog highlights the importance of accessibility at tech conferences and encourages others to help make these events more inclusive. For further discussion, please feel free to contact <a href="/service/mailto:contact@anysign.net">contact@anysign.net</a>.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d217fc8a2530" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Install Mediapipe in Blender]]></title>
            <link>https://medium.com/@yunqiguo/install-mediapipe-in-blender-d75177123410?source=rss-150c15c2183------2</link>
            <guid isPermaLink="false">https://medium.com/p/d75177123410</guid>
            <category><![CDATA[3d-graphics]]></category>
            <category><![CDATA[blender]]></category>
            <category><![CDATA[mediapipe]]></category>
            <dc:creator><![CDATA[Yunqi Guo]]></dc:creator>
            <pubDate>Thu, 19 Aug 2021 08:06:47 GMT</pubDate>
            <atom:updated>2021-08-19T08:06:47.533Z</atom:updated>
            <content:encoded><![CDATA[<p>Mediapipe provides lightweight and fast 3D mocap functions. Using Mediapipe in Blender is efficient and affordable. Here are the steps to install Mediapipe in Blender. I’m working on Mac OS, but the process for Linux should be the same.</p><h4><strong>Replace Blender’s python environment with Anaconda</strong></h4><ul><li>Install Anaconda in your system: <a href="https://docs.anaconda.com/anaconda/install/">https://docs.anaconda.com/anaconda/install/</a></li><li>Find Blender’s python path and version in the console:</li></ul><pre>PYTHON INTERACTIVE CONSOLE 3.7.7 (default, Jul  8 2020, 14:54:17)  [Clang 11.0.0 (clang-1100.0.33.8)]<br>...<br>&gt;&gt;&gt; import sys<br>&gt;&gt;&gt; sys.path<br>[..., &#39;/Applications/Blender.app/Contents/Resources/2.92/python/lib/python3.7&#39;, ...]</pre><ul><li>I’m using Blender 2.92 and python 3.7.7, located under /Applications/Blender.app/Contents/Resources/2.92/python/</li><li>Make sure to create a Conda environment with the same python version as Blender</li></ul><pre>conda create -n blender python=3.7.7</pre><ul><li>Rename the folder of the Blender Python environment:</li></ul><pre>mv path/to/blender/python path/to/blender/_python</pre><ul><li>Link the folder of Conda Python:</li></ul><pre>ln -s path/to/conda/python path/to/blender/python</pre><ul><li>Now you can use Conda to install libraries for Blender</li></ul><h4>Install Mediapipe and dependencies</h4><ul><li>Active your Conda environment and install Mediapipe</li></ul><pre>conda activate blender<br>pip install mediapipe</pre><ul><li>Now, you can use Mediapipe in the same python outside Blender. But it’s not enough for the Blender environment. So I have to run the following commands to install the other dependencies <strong>using pip and Conda.</strong></li></ul><pre>pip install requests<br>pip install opencv-python opencv-contrib-python imutils numpy dlib<br>pip install mathutils<br>conda install -c anaconda six<br>conda install -c anaconda python-dateutil<br>conda install -c conda-forge matplotlib<br>conda install -c anaconda cmake<br>conda install -c conda-forge dlib</pre><p>Some of the libraries might be not required by Mediapipe, but they are helpful for running the demos in <a href="https://youtu.be/EUR6vsE0k6E">https://youtu.be/EUR6vsE0k6</a> and <a href="https://youtu.be/O7nNO3FLkLU">https://youtu.be/O7nNO3FLkLU</a> for facial mocap.</p><p>I’m still testing the full-body mocap with Mediapipe. Will update it later.</p><h4>Reference</h4><ul><li><a href="https://medium.com/@hansongfang/use-anaconda-python-in-blender-8e3f9040f9ca">https://medium.com/@hansongfang/use-anaconda-python-in-blender-8e3f9040f9ca</a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d75177123410" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>