Papers › Detection, Pose Estimation and Segmentation for Multiple Bodies: Closing the Virtuous Circle

Detection, Pose Estimation and Segmentation for Multiple Bodies: Closing the Virtuous Circle

2 Dec 2024arXiv 2024 12arXiv:2412.01562archive 2025-07-28

Miroslav Purkrabek, Jiri Matas

Human pose estimation methods work well on separated people but struggle with multi-body scenarios. Recent work has addressed this problem by conditioning pose estimation with detected bounding boxes or bottom-up-estimated poses. Unfortunately, all of these approaches overlooked segmentation masks and their connection to estimated keypoints. We condition pose estimation model by segmentation masks instead of bounding boxes to improve instance separation. This improves top-down pose estimation in multi-body scenarios but does not fix detection errors. Consequently, we develop BBox-Mask-Pose (BMP), integrating detection, segmentation and pose estimation into self-improving feedback loop. We adapt detector and pose estimation model for conditioning by instance masks and use Segment Anything as pose-to-mask model to close the circle. With only small models, BMP is superior to top-down methods on OCHuman dataset and to detector-free methods on COCO dataset, combining the best from both approaches and matching state of art performance in both settings. Code is available on https://mirapurkrabek.github.io/BBox-Mask-Pose.

PaperPDFConference PDFCode

Code

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

Human Instance SegmentationPose EstimationPose-Based Human Instance SegmentationSegmentation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
2D Human Pose Estimation OCHuman BBox-Mask-Pose 2x Test AP 48.3 #1 of 11 Archive leaderboard report
2D Human Pose Estimation OCHuman BBox-Mask-Pose 2x Validation AP 48.6 #1 of 11 Archive leaderboard report
Human Instance Segmentation OCHuman BBox-Mask-Pose 2x AP 32.4 #1 of 18 Archive leaderboard report
Human Instance Segmentation OCHuman RTMDet-ins-l AP 26.5 #10 of 18 Archive leaderboard report
Keypoint Detection OCHuman BBox-Mask-Pose 2x Test AP 48.3 #1 of 10 Archive leaderboard report
Keypoint Detection OCHuman BBox-Mask-Pose 2x Validation AP 48.6 #1 of 10 Archive leaderboard report
Pose Estimation OCHuman BBox-Mask-Pose 2x Test AP 48.3 #5 of 19 Archive leaderboard report
Pose Estimation OCHuman BBox-Mask-Pose 2x Validation AP 48.6 #5 of 19 Archive leaderboard report
Pose Estimation OCHuman MaskPose-b Test AP 45.0 #9 of 19 Archive leaderboard report
Pose Estimation OCHuman MaskPose-b Validation AP 45.3 #9 of 19 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