<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Image Search on HyperCrux.com</title>
    <link>https://hypercrux.com/tags/image-search/</link>
    <description>Recent content in Image Search on HyperCrux.com</description>
    <generator>Hugo</generator>
    <language>en-us</language>
    <lastBuildDate>Wed, 07 Oct 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://hypercrux.com/tags/image-search/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Finding the Shot You Half Remember: A Photographer&#39;s Archive Searched by Image Embeddings</title>
      <link>https://hypercrux.com/finding-the-shot-you-half-remember-a-photographers-archive-searched-by-image-embeddings/</link>
      <pubDate>Wed, 07 Oct 2026 00:00:00 +0000</pubDate>
      <guid>https://hypercrux.com/finding-the-shot-you-half-remember-a-photographers-archive-searched-by-image-embeddings/</guid>
      <description>&lt;p&gt;Imagine a commercial photographer with about 60,000 keepers from twelve years of shoots. A client calls. They want the shot of the red chairs on the stage from one of their events, or something like it, and wider if she has one. She can picture it. She can&amp;rsquo;t place which event it was, and her keywords from back then are patchy.&lt;/p&gt;&#xA;&lt;p&gt;Image embeddings are good at this kind of memory. A CLIP model such as ViT-B/32 turns each photo into 512 numbers, and photos that look alike end up close together. It can turn a sentence into numbers in the same space, so &amp;ldquo;red chairs on a stage&amp;rdquo; lands near photos of red chairs on a stage. The rest of what she knows, which client it was for and that she only wants her four- and five-star picks, are facts, and facts belong in fields and links.&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
