Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

HEMlets Pose: Learning Part-Centric Heatmap Triplets for Accurate 3D Human Pose Estimation

About

Estimating 3D human pose from a single image is a challenging task. This work attempts to address the uncertainty of lifting the detected 2D joints to the 3D space by introducing an intermediate state - Part-Centric Heatmap Triplets (HEMlets), which shortens the gap between the 2D observation and the 3D interpretation. The HEMlets utilize three joint-heatmaps to represent the relative depth information of the end-joints for each skeletal body part. In our approach, a Convolutional Network (ConvNet) is first trained to predict HEMlests from the input image, followed by a volumetric joint-heatmap regression. We leverage on the integral operation to extract the joint locations from the volumetric heatmaps, guaranteeing end-to-end learning. Despite the simplicity of the network design, the quantitative comparisons show a significant performance improvement over the best-of-grade method (by 20% on Human3.6M). The proposed method naturally supports training with "in-the-wild" images, where only weakly-annotated relative depth information of skeletal joints is available. This further improves the generalization ability of our model, as validated by qualitative comparisons on outdoor images.

Kun Zhou, Xiaoguang Han, Nianjuan Jiang, Kui Jia, Jiangbo Lu• 2019

Related benchmarks

TaskDatasetResultRank
3D Human Pose EstimationMPI-INF-3DHP (test)
PCK75.3
559
3D Human Pose EstimationHuman3.6M (Protocol #1)
MPJPE (Avg.)39.9
440
3D Human Pose EstimationHuman3.6M Protocol 1 (test)
Dir. Error (Protocol 1)34.4
183
3D Human Pose EstimationHuman3.6M (subjects 9 and 11)--
180
3D Human Pose EstimationHuman3.6M v1 (test)
Avg Performance39.9
58
3D Human Pose EstimationHumanEva
Walk S1 Error13.5
32
3D Human Pose EstimationHuman3.6M Protocol 2 (subjects 9 and 11)
Avg Error (mm)32.1
19
3D Human Pose EstimationMPI-INF-3DHP Universal, height-normalized skeletons 1.0/2.0 (test)--
8
Showing 8 of 8 rows

Other info

Follow for update