Integral Human Pose Regression

An Integral Pose Regression System for the ECCV2018 PoseTrack Challenge

“…Promising extensions of the heat map approach have been presented for 3D pose estimation [6]. Most recently, Sun et al [7] replaced the argmax post-processing with an integral formulation to tackle the quantization problem and enable end-to-end learning. This approach currently achieves the highest 3D pose estimation performance.…”

“…For the ECCV 2018 PoseTrack Challenge, we present a 3D human pose estimation system based mainly on the integral human pose regression [7] method.…”

“…Specifically, 2D human pose data from MPII [3] and COCO [4] are used for both camera model estimation and 2D/3D mixed data training. COCO object bounding box data is used to train a person box detector for a two-stage top-down pose estimation paradigm similar to [7,8]. ImageNet [9] classification data is employed for pre-training the backbone networks.…”

“…These two components are used for data pre-processing. Third, the core CNN based learning component applies Integral Regression [7] to perform 3D human pose estimation.…”

“…Person Box Detection Instead of predicting the 3D key points from the original image directly, we follow a two-stage top-down paradigm similar to [7,8].…”