Ashish Patel 🇮🇳’s Post

𝗗𝗮𝘆-𝟭𝟰𝟳 Computer Vision Learning 𝗔𝗧𝗦𝗦: Bridging the Gap Between Anchor-based and Anchor-free Detection via Adaptive Training Sample Selection by CBSR, NLPR, CASIA, UCAS, AIR, CAS, BUPT Follow me for a similar post:  🇮🇳 Ashish Patel Interesting Facts : 🔸 This is a paper in CVPR 2020 with over 140 citations. 🔸 It Outperforms with RetinaNet, RepPoints, CenterNet, CornetNet-lite, CornetNet, Grid R-CNN, ExtreameNet etc. ------------------------------------------------------------------- 𝗔𝗺𝗮𝘇𝗶𝗻𝗴 𝗥𝗲𝘀𝗲𝗮𝗿𝗰𝗵 : https://lnkd.in/eeteXym Code: https://lnkd.in/eUFyAnm ------------------------------------------------------------------- 𝗜𝗠𝗣𝗢𝗥𝗧𝗔𝗡𝗖𝗘 🔸  Target detection has been dominated by anchor-based detectors for several years. Due to the proposal of FPN and Focal loss, anchor-free detectors have become popular again.  🔸 ATSS used two types of approach anchor-based or anchor-free detector. The author proposes an Adaptive Training Sample Selection (ATSS), which automatically divides positive and negative samples according to the feature statistics of the target. This method significantly improves the detection performance of the anchor-free and anchor-based detectors and makes up for the difference between the two. difference.   #computervision #artificialintelligence