<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Computer Vision |</title><link>https://relo02.github.io/tags/computer-vision/</link><atom:link href="https://relo02.github.io/tags/computer-vision/index.xml" rel="self" type="application/rss+xml"/><description>Computer Vision</description><generator>Hugo Blox Builder (https://hugoblox.com)</generator><language>en-us</language><lastBuildDate>Mon, 01 Dec 2025 00:00:00 +0000</lastBuildDate><image><url>https://relo02.github.io/media/icon_hu_982c5d63a71b2961.png</url><title>Computer Vision</title><link>https://relo02.github.io/tags/computer-vision/</link></image><item><title>Histopathology Image Classification — 4th / 170</title><link>https://relo02.github.io/projects/histopathology/</link><pubDate>Mon, 01 Dec 2025 00:00:00 +0000</pubDate><guid>https://relo02.github.io/projects/histopathology/</guid><description>&lt;p&gt;Image classification of molecular structures in tissue images for the &lt;strong&gt;Artificial Neural Networks
&amp;amp; Deep Learning&lt;/strong&gt; course at Politecnico di Milano — &lt;strong&gt;ranked 4th of 170 teams&lt;/strong&gt; on a hidden test set.&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;&lt;strong&gt;Backbone:&lt;/strong&gt; transfer learning with a &lt;strong&gt;ResNet-50&lt;/strong&gt; pre-trained via &lt;strong&gt;RetCCL&lt;/strong&gt; for
histopathology-specific feature extraction.&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;Pipeline:&lt;/strong&gt; reproducible &lt;strong&gt;PyTorch Lightning&lt;/strong&gt; training with &lt;strong&gt;WeightedRandomSampler&lt;/strong&gt; for class
imbalance, &lt;strong&gt;AdamW&lt;/strong&gt;, label smoothing, cosine annealing, and rotation/flip/color-jitter augmentation.&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;&lt;code&gt;Python&lt;/code&gt; · &lt;code&gt;PyTorch&lt;/code&gt; · &lt;code&gt;PyTorch Lightning&lt;/code&gt; · &lt;code&gt;CNNs&lt;/code&gt;&lt;/p&gt;</description></item></channel></rss>