<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Dashrath Mali | Prathmesh Bele</title><link>https://prathmeshbele.github.io/authors/dashrath-mali/</link><atom:link href="https://prathmeshbele.github.io/authors/dashrath-mali/index.xml" rel="self" type="application/rss+xml"/><description>Dashrath Mali</description><generator>HugoBlox Kit (https://hugoblox.com)</generator><language>en-us</language><lastBuildDate>Sun, 01 Jan 2023 00:00:00 +0000</lastBuildDate><item><title>Node and Edge Embedding Learning for Static Obstacle Segmentation</title><link>https://prathmeshbele.github.io/publications/static-obstacle-segmentation/</link><pubDate>Sun, 01 Jan 2023 00:00:00 +0000</pubDate><guid>https://prathmeshbele.github.io/publications/static-obstacle-segmentation/</guid><description>&lt;p&gt;Proposed a novel GNN architecture formulating free-space mapping as a scalable edge-prediction problem. Architected a dual-path Graph Attention Network to independently process node and edge features. Implemented class-aware edge retention via feature fusion, achieving parity with inflexible rule-based legacy systems.&lt;/p&gt;</description></item></channel></rss>