<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Wide-Baseline Multi-Camera Calibration using Person Re-Identification · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/2104.08568/</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/2104.08568/feed.xml" rel="self" type="application/rss+xml" /><item><title>Wide-Baseline Multi-Camera Calibration using Person Re-Identification — mentioned by eloisius</title><link>https://news.ycombinator.com/item?id=32917304</link><guid isPermaLink="false">urn:hackerpapers:mention:2104.08568:32917304</guid><pubDate>Tue, 20 Sep 2022 20:25:21 GMT</pubDate><description>&lt;p&gt;I can at least share some research that has been useful for what I&amp;#x27;m working on: https://arxiv.org/pdf/2104.08568.pdf Face reidentification is the approach this paper uses to find corresponding views of the same person, but that&amp;#x27;s not necessary if you have the ability to match the objects in your scene otherwise. For example, doing 3D triangulation to geometrically verify random matches using RANSAC.&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/2104.08568/"&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>
