Ashish Patel 🇮🇳’s Post

𝗗𝗮𝘆-𝟭𝟲𝟬 Computer Vision Learning In 𝗣𝗲𝗿𝗳𝗲𝗰𝘁 𝗦𝗵𝗮𝗽𝗲: Certifiably Optimal 3D Shape Reconstruction from 2D Landmarks by Massachusetts Institute of Technology Follow me for similar post :  🇮🇳 Ashish Patel 𝗜𝗻𝘁𝗲𝗿𝗲𝘀𝘁𝗶𝗻𝗴 𝗙𝗮𝗰𝘁𝘀 : 🔸 This is a paper in CVPR 2020 with over 65 citations. 🔸 It Outperforms with the OccNet, DMC, PSGN, etc. ------------------------------------------------------------------- 𝗔𝗺𝗮𝘇𝗶𝗻𝗴 𝗥𝗲𝘀𝗲𝗮𝗿𝗰𝗵 : https://lnkd.in/eA-uY5z ------------------------------------------------------------------- 𝗜𝗠𝗣𝗢𝗥𝗧𝗔𝗡𝗖𝗘 🔸Problem of 3D shape reconstruction from 2D landmarks extracted in a single image. Adopt the 3D deformable shape model and formulate the reconstruction as a joint optimization of the camera pose and the linear shape parameters. 🔸First contribution is to apply Lasserre’s hierarchy of convex Sums-of-Squares (SOS) relaxations to solve the shape reconstruction problem and show that the SOS relaxation of order 2 empirically solves the original non-convex problem exactly. 🔸Second contribution is to exploit the structure of the polynomial in the objective function and find a reduced set of basis monomials for the SOS relaxation that significantly decreases the size of the resulting semidefinite program (SDP) without compromising its accuracy. #computervision #artificialintelligence #innovation