<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>A Learned Stereo Depth System for Robotic Manipulation in Homes · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/2109.11644/</link><description>Collected HN comments and submissions linking this paper.</description><language>en</language><lastBuildDate>Mon, 02 Nov 2026 11:16:22 GMT</lastBuildDate><atom:link href="https://hackerpaper.kaipfister.de/p/2109.11644/feed.xml" rel="self" type="application/rss+xml" /><item><title>A Learned Stereo Depth System for Robotic Manipulation in Homes — mentioned by GistNoesis</title><link>https://news.ycombinator.com/item?id=30160636</link><guid isPermaLink="false">urn:hackerpapers:mention:2109.11644:30160636</guid><pubDate>Tue, 01 Feb 2022 10:12:20 GMT</pubDate><description>&lt;p&gt;The title of the webpage is LIDAR 3d scanner. I don&amp;#x27;t have an IPhone to try the app, so based on the videos and the information I could gather online. It seems they have multiple operating modes. One mode use rgb image only and standard photogrammetry like meshroom. In the other mode they use the information from the LIDAR sensor to help with depth and camera registration, with some sort of RGB-D SLAM algorithms. Looking for point-cloud fusion algorithms and the pcl library should get you started. There is probably some deep-learning involved but it&amp;#x27;s not yet end-to-end deep learning, in…&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/2109.11644/"&gt;Paper and discussion history&lt;/a&gt;. A mention is not an endorsement.&lt;/p&gt;</description><category>cs.RO</category></item></channel></rss>
