Papers › Learning Temporal Pose Estimation from Sparsely-Labeled Videos

Learning Temporal Pose Estimation from Sparsely-Labeled Videos

6 Jun 2019NeurIPS 2019 12arXiv:1906.04016archive 2025-07-28

Gedas Bertasius, Christoph Feichtenhofer, Du Tran, Jianbo Shi, Lorenzo Torresani

Modern approaches for multi-person pose estimation in video require large amounts of dense annotations. However, labeling every frame in a video is costly and labor intensive. To reduce the need for dense annotations, we propose a PoseWarper network that leverages training videos with sparse annotations (every k frames) to learn to perform dense temporal pose propagation and estimation. Given a pair of video frames---a labeled Frame A and an unlabeled Frame B---we train our model to predict human pose in Frame A using the features from Frame B by means of deformable convolutions to implicitly learn the pose warping between A and B. We demonstrate that we can leverage our trained PoseWarper for several applications. First, at inference time we can reverse the application direction of our network in order to propagate pose information from manually annotated frames to unlabeled frames. This makes it possible to generate pose annotations for the entire video given only a few manually-labeled frames. Compared to modern label propagation methods based on optical flow, our warping mechanism is much more compact (6M vs 39M parameters), and also more accurate (88.7% mAP vs 83.8% mAP). We also show that we can improve the accuracy of a pose estimator by training it on an augmented dataset obtained by adding our propagated poses to the original manual labels. Lastly, we can use our PoseWarper to aggregate temporal pose information from neighboring frames during inference. This allows our system to achieve state-of-the-art pose detection results on the PoseTrack2017 and PoseTrack2018 datasets. Code has been made available at: https://github.com/facebookresearch/PoseWarper.

PaperPDFConference PDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

facebookresearch/PoseWarper officialmentioned in papermentioned on GitHubpytorch report
shuuchen/PoseWarper.pytorch mentioned on GitHubpytorch report
open-mmlab/mmpose pytorchApache-2.0 report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Multi-Person Pose EstimationOptical Flow EstimationPose Estimation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Multi-Person Pose Estimation PoseTrack2017 PoseWarper Mean mAP 77.94 #2 of 3 Archive leaderboard report
Multi-Person Pose Estimation PoseTrack2018 PoseWarper Mean mAP 78 #3 of 4 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections