Ashish Patel 🇮🇳’s Post

𝗗𝗮𝘆-𝟭𝟲𝟯 Computer Vision Learning 𝗦𝗚-𝗡𝗡: Sparse Generative Neural Networks for Self-Supervised Scene Completion of RGB-D Scans by Technical University of Munich Follow me for similar post:  🇮🇳 Ashish Patel Interesting Facts : 🔸 This is a paper in CVPR 2020 with over 27 citations. 🔸 It Outperforms with the 3D-EPN, etc. ------------------------------------------------------------------- 𝗔𝗺𝗮𝘇𝗶𝗻𝗴 𝗥𝗲𝘀𝗲𝗮𝗿𝗰𝗵 : https://lnkd.in/edKYJND ------------------------------------------------------------------- 𝗜𝗠𝗣𝗢𝗥𝗧𝗔𝗡𝗖𝗘 🔸 SG-NN presented a self-supervised approach for completion of RGB-D scan geometry that enables training solely on incomplete, real-world scans while learning a generative geometric completion process capable of predicting 3D scene geometry more complete than any single target scene seen during training. ✔️ Sparse generative approach to generating a sparse TSDF representation of a surface enables much higher output geometric resolution than previous on large-scale 3D scenes. #computervision #artificialintelligence #technology