Perceiving Systems, Computer Vision

SPIN: Human pose and shape from an image

2019-10-01


SPIN is a state-of-the-art deep network for regressing SMPL body shape and pose parameters directly from an image. SPIN uses a novel training method that combines a bottom-up deep network with a top-down, model-based, fitting method. SMPLify model fitting is used in the loop with the DNN training to provide SMPL parameters used in the training loss. Code is available.

SPIN is a state-of-the-art deep network for regressing SMPL body shape and pose parameters directly from an image. SPIN uses a novel training method that combines a bottom-up deep network with a top-down, model-based, fitting method. SMPLify model fitting is used in the loop with the DNN training to provide SMPL parameters used in the training loss. Code is available.

Author(s): Kolotouros, Nikos and Pavlakos, Georgios and Black, Michael J. and Daniilidis, Kostas
Department(s): Perceiving Systems
Publication(s): Learning to Reconstruct {3D} Human Pose and Shape via Model-fitting in the Loop
Authors: Kolotouros, Nikos and Pavlakos, Georgios and Black, Michael J. and Daniilidis, Kostas
Release Date: 2019-10-01
Repository: https://github.com/nkolot/SPIN
External Link: https://www.seas.upenn.edu/~nkolot/projects/spin/