🔥 6x Linkedln Top Voice | AI Research Scientist & Chief Data Scientist at IBM | Generative AI Expert | Author - Hands-on Time Series Analytics with Python | IBM Quantum ML Certified | 11+ Years in AI | MLOps | IIMA |
𝗗𝗮𝘆-𝟭𝟮𝟮 Computer Vision Learning 𝗗𝟮𝗗𝗲𝘁:Towards High Quality Object Detection and Instance Segmentation by Follow me for a similar post: 🇮🇳 Ashish Patel Interesting Facts : 🔸 This is a paper in CVPR 2020 with over 15 citations. 🔸 It Outperforms with MaskRCNN, PANet. ------------------------------------------------------------------- 𝗔𝗺𝗮𝘇𝗶𝗻𝗴 𝗥𝗲𝘀𝗲𝗮𝗿𝗰𝗵 : https://bit.ly/3aW0iDq Code : https://lnkd.in/e2UtY69 ------------------------------------------------------------------- 𝗜𝗠𝗣𝗢𝗥𝗧𝗔𝗡𝗖𝗘 🔸 a) Overall architecture of our two-stage method. The RoI feature of each candidate proposal P, generated from RPN is passed through two different branches: dense local regression (b) and classification (c). Instead of treating RoI feature as a single global vector, our dense local regression treats them as k × k local features extracted from k × k subregions within RoI. #computervision #artificialintelligence #deeplearning #innovation
🔥 6x Linkedln Top Voice | AI Research Scientist & Chief Data Scientist at IBM | Generative AI Expert | Author - Hands-on Time Series Analytics with Python | IBM Quantum ML Certified | 11+ Years in AI | MLOps | IIMA |
2yFor previous post visit this github : https://github.com/ashishpatel26/365-Days-Computer-Vision-Learning-Linkedin-Post