The BBox-Mask-Pose (BMP) method integrates detection, pose estimation, and segmentation into a self-improving loop by conditioning these tasks on each other.
This approach enhances all three tasks simultaneously.
Using segmentation masks instead of bounding boxes improves performance in crowded scenarios, making top-down methods competitive with bottom-up approaches.
Key contributions:
- MaskPose: a pose estimation model conditioned by segmentation masks instead of bounding boxes, boosting performance in dense scenes without adding parameters
- Download pre-trained weights below
- PMPose: a pose estimation model conditioned by segmentation masks AND predicting full description of each keypoint. Combination of MaskPose and ProbPose (CVPR'25).
- BBox-MaskPose (BMP): method linking bounding boxes, segmentation masks, and poses to simultaneously address multi-body detection, segmentation and pose estimation
- Fine-tuned RTMDet adapted for itterative detection (ignoring 'holes')
- Download pre-trained weights below
- Support for multi-dataset training of ViTPose, previously implemented in the official ViTPose repository but absent in MMPose.
For more details, see the
GitHub repository.
See details of each model below.
What's new?
PMPose combines MaskPose-1.1.0 and
ProbPose (CVPR'25). It is conditioned by masks and has superior in-crowd performance as MaskPose and also precicts proabilities and visibilities as ProbPose.
1@InProceedings{BMPv2,
2 author = {Purkrabek, Miroslav and Kolomiiets, Constantin and Matas, Jiri},
3 title = {BBoxMaskPose v2: Expanding Mutual Conditioning to 3D},
4 booktitle = {arXiv preprint arXiv:to be added},
5 year = {2026}
6}
1@InProceedings{Purkrabek2025ICCV,
2 author = {Purkrabek, Miroslav and Matas, Jiri},
3 title = {Detection, Pose Estimation and Segmentation for Multiple Bodies: Closing the Virtuous Circle},
4 booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)},
5 month = {October},
6 year = {2025}
7}
1@InProceedings{Kolomiiets2026CVWW,
2 author = {Kolomiiets, Constantin and Purkrabek, Miroslav and Matas, Jiri},
3 title = {SAM-pose2seg: Pose-Guided Human Instance Segmentation in Crowds},
4 booktitle = {Computer Vision Winter Workshop (CVWW)},
5 year = {2026}
6}
