<?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 Dheera Venkatraman on Medium]]></title>
        <description><![CDATA[Stories by Dheera Venkatraman on Medium]]></description>
        <link>https://medium.com/@dheeranet?source=rss-3de225deda00------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*1MM_luCyhcpaWBCY.</url>
            <title>Stories by Dheera Venkatraman on Medium</title>
            <link>https://medium.com/@dheeranet?source=rss-3de225deda00------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Mon, 16 Nov 2026 23:29:03 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@dheeranet/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[Snapchat vs. Art Museum]]></title>
            <link>https://medium.com/@dheeranet/snapchat-vs-art-museum-5ad89de3fce0?source=rss-3de225deda00------2</link>
            <guid isPermaLink="false">https://medium.com/p/5ad89de3fce0</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[snapchat]]></category>
            <category><![CDATA[social-media]]></category>
            <category><![CDATA[computer-vision]]></category>
            <category><![CDATA[art]]></category>
            <dc:creator><![CDATA[Dheera Venkatraman]]></dc:creator>
            <pubDate>Tue, 28 May 2019 19:13:05 GMT</pubDate>
            <atom:updated>2019-05-28T19:48:36.188Z</atom:updated>
            <content:encoded><![CDATA[<h3>Snapchat Filters vs. Art Museum</h3><p>I’m sure most of us have seen our friends posting hilarious gender-swapped versions of themselves all over social media, thanks to a new filter by Snapchat.</p><p>Now I have never used Snapchat before — I just don’t know why I’d use it, personally — and I don’t usually go to museums. But that’s all only because I’m not usually a trend follower. On the other hand, I thought that registering a Snapchat account and taking it to an art museum might be fun, just to test the limits of the filter on “selfies” from a few hundred years ago. So I headed over to Stanford’s Cantor Arts Center to find out.</p><p>A couple of the paintings had somewhat impressive results. It worked particularly well on male-to-female conversions where the only modifications really needed were some face cleaning, face shaping, and long hair.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*_ndY4sbGdO3Sg8IpFie9xA.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*cGLnvEfyHu4HHU0NTK3Nlw.jpeg" /></figure><p>It failed miserably though when it comes to beards, and understandably so, as a computer vision engineer myself — this is an in-painting problem, where you’d take the beard out and want to paint a chin and neck in its place. While there have been some impressive recent results in both still image [<a href="http://www.chuanxiaz.com/project/pluralistic/">0</a>, <a href="https://neurohive.io/en/state-of-the-art/facebook-ai-research-presents-eye-in-painting-with-exemplar-gan/">1</a>], and video [<a href="https://arxiv.org/abs/1905.02884">2</a>] in-painting, and I’m sure some of these might perform well on these images if one were to take the beard out of the image, I doubt Snapchat had the resources to incorporate such methods into a phone app. Nevertheless the results are interesting, to say the least.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*VUbyddcB7CNGApusfuHqAw.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NfJqNZIYE0DIP5UqgmCSfQ.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4cQ8uuAqPk90yx_Qma_BPQ.jpeg" /></figure><p>Female to male conversions seem to amount largely to eyebrow thickening and roughening up the face a bit. Snapchat’s filter evidently doesn’t really know to modify anything but the face though.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bw0BT0ugys4Sc9X5FurUoA.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NohxuNq6CKtIS5Fcv4d7AQ.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fJvIOhS98tScWGrE-6TScg.jpeg" /></figure><p>Enough with the paintings. Art museums also usually have 3D sculptures, to which Snapchat mostly decided to give them some real human hair without modifying much else. Here are a couple of Ming Dynasty sculptures from China and their gender-swapped alter egos. (An interesting side note: the Buddhist deity Guanyin — not pictured here — evolved from male to female over the centuries and passage through various cultures.)</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ClRrzAIk1wvR7P6rECZ2cQ.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NxzC4f85hliJhiUqQCuJxg.jpeg" /></figure><p>Lastly here are a few other statues, and and as you can see it really didn’t do well with that last African face mask.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*AhguQQaRes0jEh-doYBQUA.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*a35NDYJzAVEkjiqvSi-V3g.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Eu0Hc7TFiVj19ecqcupLaA.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*K8l1S8ndq30StO4XIo10Qw.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Y24mAk4v6ggeWLSaCJZ1NQ.jpeg" /></figure><p>On one hand, we do see that Snapchat’s filter has learned some gender stereotypes (the most obvious one here being that “women have long hair, men have short hair”), but some of the results are comical in its inability to in-paint beards or deal with more abstract interpretations of the human body. Maybe future versions of the filter will improve on this, or maybe the fad will pass and we’ll see other interesting identity-changing filters, including age, hairstyle, or outfit changing.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5ad89de3fce0" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[ROSshow: Visualize robot sensor data with ASCII art]]></title>
            <link>https://medium.com/@dheeranet/rosshow-visualize-robot-sensor-data-with-ascii-art-cb6b8a1fe760?source=rss-3de225deda00------2</link>
            <guid isPermaLink="false">https://medium.com/p/cb6b8a1fe760</guid>
            <category><![CDATA[robots]]></category>
            <category><![CDATA[data-visualization]]></category>
            <category><![CDATA[robotics]]></category>
            <category><![CDATA[hacking]]></category>
            <category><![CDATA[programming]]></category>
            <dc:creator><![CDATA[Dheera Venkatraman]]></dc:creator>
            <pubDate>Fri, 29 Mar 2019 18:53:12 GMT</pubDate>
            <atom:updated>2019-03-29T18:54:19.737Z</atom:updated>
            <content:encoded><![CDATA[<p>Have you ever ssh’ed into a robot to debug whether a sensor is outputting what it should, and used rostopic echo? Have you ever copy-pasted a GPS location from the terminal into Google Maps just to see where it is? Have you ever been lazy to fire up rviz, port forwards, X forwards, or dig up HDMI cables to plug into your robot? If so, ROSshow is for you.</p><p>ROSshow renders visualizations of ROS topics, including time-series graphs, 3D point clouds, laser scans, maps, and more, from right inside the terminal. ROSshow uses Braille to render dots with sub-character resolution, but also has a strict ASCII (-a) mode that you can use if you happen to be on a non-Unicode terminal.</p><p>It’s free and open-source — please use it, contribute, send feedback, and help make ROS and robots better!</p><p>Get it here: <a href="http://github.com/dheera/rosshow">http://github.com/dheera/rosshow</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*UodEZfr4hU-MBXRHyskoHA.png" /><figcaption>sensor_msgs/NavSatFix</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*pkbyKh7Zg1_rFFckIAlZNg.png" /><figcaption>sensor_msgs/CompressedImage</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*QWihN3KTFn9HOYb-RFL-Ng.png" /><figcaption>std_msgs/Float32 (time series plot)</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*BRcHqow1XZTpfsGyaDUXTw.png" /><figcaption>sensor_msgs/PointCloud2 (Velodyne data!)</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/975/1*Wg7ajiAsfqyYmQwAGeKX6w.png" /><figcaption>sensor_msgs/LaserScan</figcaption></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=cb6b8a1fe760" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Time travelling in China]]></title>
            <link>https://medium.com/@dheeranet/time-travelling-in-china-84f4c719aac1?source=rss-3de225deda00------2</link>
            <guid isPermaLink="false">https://medium.com/p/84f4c719aac1</guid>
            <category><![CDATA[travel]]></category>
            <category><![CDATA[history]]></category>
            <category><![CDATA[photography]]></category>
            <dc:creator><![CDATA[Dheera Venkatraman]]></dc:creator>
            <pubDate>Thu, 07 Apr 2016 03:24:33 GMT</pubDate>
            <atom:updated>2016-04-07T04:10:36.009Z</atom:updated>
            <content:encoded><![CDATA[<p>When most people travel, they think of travelling in space. Whether it’s hiking through a national park or learning about a distant culture, it’s usually about the destination. This time, I decided to travel in time. Here are some scenes from China as they looked over the past hundred years. The old pictures were pulled from various books and websites. I then re-took each of them, matching the angle and focal length as closely as possible.</p><p>It’s interesting to see what humans are able to change and what humans are utterly incapable of changing. On one hand, a city may rise out of nowhere in just a matter of a decade or two, showing how powerful China’s economy has become, and how advanced engineering has enabled us to build so rapidly. At the same time, the profile of the mountains and rivers don’t budge, a constant reminder that we are still powerless compared to Mother Nature.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*PcBeVGwLPi51hR1x9i38lg.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*P60CPBMz3Zh8LBds4mf2gA.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*qB0-kHJL4xhy7w5uCmhWDg.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2HiQB9AtWcf5dRt_tcfa9A.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OoRlTvNuykFzEFC6_dqj9g.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bzmbl2Oc17ATekbPvXxPCw.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jcKglx3BG7mphuDHuoRC2w.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*22KPwEBBlsOWAlfRgkdgrg.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-_OOdjF_GCettcHuu_kYBA.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*sAmDiOXXBcPsBHhYp1im2g.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*0_f9fzZ7Ix1sdlqgsFMJ_A.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*nc7TfIsfsTMj9PTrH3vHSQ.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jpjsxWAgwQCpBM82Ig9W7w.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*tABtJ-nrv0W-fwbBT5ZAxQ.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZHDj8s4ix_IQDN1Jgf8nSQ.jpeg" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=84f4c719aac1" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>