Back
My works mainly lie in inferring human structures from RGB inputs, which starts from 2D keypoint estimation, towards more complex tasks like 3D skeleton inference and SMPL-based human pose & shape estimation. Along this road, we find that high-level tasks, like human body estimation, can benefit a lot from low-level inferred structures, like 3D skeletons, and vice versa. Furthermore, in our latest work, "Human Pose Regression with Residual Log-likelihood Estimation", we unified all the above HPS tasks in a direct regression paradigm, replacing generally accepted heatmap without loss of accuracy. This work has been accepted by ICCV'21 as Oral.
Jiefeng Li (Shanghai Jiao Tong University)
Ph.D. student
Jiefeng Li is a second-year PhD student at Shanghai Jiao Tong University, advised by Prof. Cewu Lu. His research interest lies in Computer Vision, with a focus on understanding and digitizing human. He is the co-first author of AlphaPose and has published papers about human pose estimation and model-based human reconstruction in CVPR, ICCV, ECCV and BMVC.