MUMBAI, India, July 13 -- Intellectual Property India has published a patent application (202341087789 A) filed by Mercedes-Benz Group AG on December 21, 2023, for Method And System For Performing Static Obstacle Segmentation Using Neural Network.
Inventors include Mali Dasharathbhai Dahyalal; Jeevitha Shetty; Laxmi Kant Sahoo; Prachi Solanki; Apurva Kulkarni; and Prathmesh P Bele.
The application for the patent was published on July 10, 2026, under issue no. 28/2026.
Abstract: Abstract METHOD AND SYSTEM FOR PERFORMING STATIC OBSTACLE SEGMENTATION USING NEURAL NETWORK 5 The present disclosure relates to a method and system for performing static obstacle segmentation using a neural network. The method comprises receiving an input data comprising set of features corresponding to plurality of nodes and one or more edges associated with fused data points. The method comprises determining embedding indicative of updated feature information for each of plurality of nodes and the one or more 10 edges using trained graph attention network. The graph attention network comprises plurality of graph attention operation layers. Each graph attention operation layer performs updation of embeddings of each node and edges. Then, for each edge of the one or more edges, integration of respective determined embedding with embedding of respective end nodes is performed. Thereafter, probability of retaining each edge is predicted based on 15 integration using a neural network for performing static obstacle segmentation.
Disclaimer: Curated by HT Syndication.
