<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Learning Topology from Synthetic Data for Unsupervised Depth Completion · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/2106.02994/</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/2106.02994/feed.xml" rel="self" type="application/rss+xml" /><item><title>Learning Topology from Synthetic Data for Unsupervised Depth Completion — mentioned by alexklwong</title><link>https://news.ycombinator.com/item?id=28648310</link><guid isPermaLink="false">urn:hackerpapers:mention:2106.02994:28648310</guid><pubDate>Fri, 24 Sep 2021 22:01:59 GMT</pubDate><description>&lt;p&gt;Want to leverage synthetic data for 3d reconstruction, but don&amp;#x27;t want to deal with the photometric domain gap? TLDR; Here is a gif showing an overview of our approach: https://github.com/alexklwong/learning-topology-synthetic-da... Check out the extended version of our ICRA 2021 talk for Learning Topology from Synthetic Data for Unsupervised Depth Completion: https://www.youtube.com/watch?v=zGKH-OKPJD4 This is joint work with Safa Cicek and Stefano Soatto at the UCLA Vision Lab. In the talk, we will walk you through on how we learn dense topology from sparse geometry e.g. point clouds. This…&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/2106.02994/"&gt;Paper and discussion history&lt;/a&gt;. A mention is not an endorsement.&lt;/p&gt;</description><category>cs.CV</category></item></channel></rss>
