{"url":"/dataset/posetrack","name":"PoseTrack","full_name":null,"description_markdown":"The **PoseTrack** dataset is a large-scale benchmark for multi-person pose estimation and tracking in videos. It requires not only pose estimation in single frames, but also temporal tracking across frames. It contains 514 videos including 66,374 frames in total, split into 300, 50 and 208 videos for training, validation and test set respectively. For training videos, 30 frames from the center are annotated. For validation and test videos, besides 30 frames from the center, every fourth frame is also annotated for evaluating long range articulated tracking. The annotations include 15 body keypoints location, a unique person id and a head bounding box for each person instance.\r\n\r\nSource: [Simple Baselines for Human Pose Estimation and Tracking](https://arxiv.org/abs/1804.06208)\r\nImage Source: [https://posetrack.net/](https://posetrack.net/)","description_withheld":null,"homepage":"https://posetrack.net/","introduced_date":"2018-01-01","introduced_date_note":null,"introduced_by":{"paper":"/paper/posetrack-a-benchmark-for-human-pose","title":"PoseTrack: A Benchmark for Human Pose Estimation and Tracking","first_author":"Mykhaylo Andriluka","url":null},"license":{"name":"CC BY-NC 4.0","url":"https://creativecommons.org/licenses/by-nc/4.0/"},"modalities":[{"name":"Images","url":"/datasets/modality/images"},{"name":"Videos","url":"/datasets/modality/videos"},{"name":"Tracking","url":"/datasets/modality/tracking"}],"tasks":[{"name":"Pose Tracking","url":"/task/pose-tracking","datasets_with_task":"/datasets/task/pose-tracking"},{"name":"Multi-Person Pose Estimation","url":"/task/multi-person-pose-estimation","datasets_with_task":"/datasets/task/multi-person-pose-estimation"},{"name":"Multi-Person Pose Estimation and Tracking","url":"/task/multi-person-pose-estimation-and-tracking","datasets_with_task":"/datasets/task/multi-person-pose-estimation-and-tracking"}],"languages":[],"variants":["PoseTrack2017","PoseTrack2018","PoseTrack"],"data_loaders":[{"repo":"https://github.com/open-mmlab/mmpose","url":"https://github.com/open-mmlab/mmpose/blob/master/docs/tasks/2d_body_keypoint.md#posetrack18","frameworks":["pytorch"]}],"num_papers_in_archive":103,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/pose-tracking-on-posetrack2017","task":"Pose Tracking","dataset_variant":"PoseTrack2017","rows":10,"metrics":["MOTA","mAP"],"first_row_in_archive_order":{"model":"DetTrack","paper":"/paper/combining-detection-and-tracking-for-human","metrics":{"MOTA":"64.09","mAP":"74.14"},"code_links":[]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/pose-tracking-on-posetrack2018","task":"Pose Tracking","dataset_variant":"PoseTrack2018","rows":5,"metrics":["MOTA","mAP","IDF1","IDs"],"first_row_in_archive_order":{"model":"DetTrack","paper":"/paper/combining-detection-and-tracking-for-human","metrics":{"MOTA":"64.3","mAP":"73.5"},"code_links":[]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/multi-person-pose-estimation-on-posetrack2018","task":"Multi-Person Pose Estimation","dataset_variant":"PoseTrack2018","rows":4,"metrics":["Mean mAP"],"first_row_in_archive_order":{"model":"Poseidon","paper":"/paper/poseidon-a-vit-based-architecture-for-multi","metrics":{"Mean mAP":"87.8"},"code_links":[{"title":"CesareDavidePace/poseidon","url":"https://github.com/CesareDavidePace/poseidon"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/multi-person-pose-estimation-on-posetrack2017","task":"Multi-Person Pose Estimation","dataset_variant":"PoseTrack2017","rows":3,"metrics":["Mean mAP"],"first_row_in_archive_order":{"model":"DCPose","paper":"/paper/deep-dual-consecutive-network-for-human-pose","metrics":{"Mean mAP":"79.2"},"code_links":[{"title":"Pose-Group/DCPose","url":"https://github.com/Pose-Group/DCPose"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/multi-person-pose-estimation-and-tracking-on-1","task":"Multi-Person Pose Estimation and Tracking","dataset_variant":"PoseTrack2018","rows":1,"metrics":["MOTA"],"first_row_in_archive_order":{"model":"Refine","paper":"/paper/learning-to-refine-human-pose-estimation","metrics":{"MOTA":"58.4"},"code_links":[]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/poseidon-a-vit-based-architecture-for-multi","title":"Poseidon: A ViT-based Architecture for Multi-Frame Pose Estimation with Adaptive Frame Weighting and Multi-Scale Feature Fusion","date":"2025-01-14","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/humans-in-4d-reconstructing-and-tracking","title":"Humans in 4D: Reconstructing and Tracking Humans with Transformers","date":"2023-05-31","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":5,"samples_ran":4,"samples_unverified":1,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/do-different-tracking-tasks-require-different","title":"Do Different Tracking Tasks Require Different Appearance Models?","date":"2021-07-05","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":17,"samples_ran":6,"samples_unverified":11,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/deep-dual-consecutive-network-for-human-pose","title":"Deep Dual Consecutive Network for Human Pose Estimation","date":"2021-03-12","rows_on_this_dataset":2,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":13,"samples_ran":11,"samples_unverified":2,"pointer_only_for_licence":13,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/combining-detection-and-tracking-for-human","title":"Combining detection and tracking for human pose estimation in videos","date":"2020-03-30","rows_on_this_dataset":2,"code_links":0,"syntology":null},{"paper":"/paper/15-keypoints-is-all-you-need","title":"15 Keypoints Is All You Need","date":"2019-12-05","rows_on_this_dataset":1,"code_links":0,"syntology":null},{"paper":"/paper/learning-temporal-pose-estimation-from","title":"Learning Temporal Pose Estimation from Sparsely-Labeled Videos","date":"2019-06-06","rows_on_this_dataset":2,"code_links":3,"syntology":null},{"paper":"/paper/pose-estimator-and-tracker-using-temporal","title":"Pose estimator and tracker using temporal flow maps for limbs","date":"2019-05-23","rows_on_this_dataset":2,"code_links":0,"syntology":null},{"paper":"/paper/lighttrack-a-generic-framework-for-online-top","title":"LightTrack: A Generic Framework for Online Top-Down Human Pose Tracking","date":"2019-05-07","rows_on_this_dataset":1,"code_links":2,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":16,"samples_ran":2,"samples_unverified":14,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/deep-high-resolution-representation-learning","title":"Deep High-Resolution Representation Learning for Human Pose Estimation","date":"2019-02-25","rows_on_this_dataset":1,"code_links":39,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":25,"samples_ran":8,"samples_unverified":17,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/efficient-online-multi-person-2d-pose","title":"Efficient Online Multi-Person 2D Pose Tracking with Recurrent Spatio-Temporal Affinity Fields","date":"2018-11-29","rows_on_this_dataset":1,"code_links":0,"syntology":null},{"paper":"/paper/learning-to-refine-human-pose-estimation","title":"Learning to Refine Human Pose Estimation","date":"2018-04-21","rows_on_this_dataset":2,"code_links":0,"syntology":null},{"paper":"/paper/simple-baselines-for-human-pose-estimation","title":"Simple Baselines for Human Pose Estimation and Tracking","date":"2018-04-17","rows_on_this_dataset":2,"code_links":27,"syntology":null},{"paper":"/paper/pose-flow-efficient-online-pose-tracking","title":"Pose Flow: Efficient Online Pose Tracking","date":"2018-02-03","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/detect-and-track-efficient-pose-estimation-in","title":"Detect-and-Track: Efficient Pose Estimation in Videos","date":"2017-12-26","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/posetrack-a-benchmark-for-human-pose","title":"PoseTrack: A Benchmark for Human Pose Estimation and Tracking","date":"2017-10-27","rows_on_this_dataset":2,"code_links":2,"syntology":null}],"syntology_totals":{"read_at":"2026-09-24T18:15:14+00:00","papers_with_samples":5,"samples_harvested":76,"samples_ran":31,"samples_unverified":45,"pointer_only_for_licence":13,"papers_with_no_sample_that_ran":0,"note":"the per-paper counts above, summed; not a rate"},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks."}