Papers › Rethinking on Multi-Stage Networks for Human Pose Estimation

Rethinking on Multi-Stage Networks for Human Pose Estimation

1 Jan 2019arXiv:1901.00148archive 2025-07-28

Wenbo Li, Zhicheng Wang, Binyi Yin, Qixiang Peng, Yuming Du, Tianzi Xiao, Gang Yu, Hongtao Lu, Yichen Wei, Jian Sun

Existing pose estimation approaches fall into two categories: single-stage and multi-stage methods. While multi-stage methods are seemingly more suited for the task, their performance in current practice is not as good as single-stage methods. This work studies this issue. We argue that the current multi-stage methods' unsatisfactory performance comes from the insufficiency in various design choices. We propose several improvements, including the single-stage module design, cross stage feature aggregation, and coarse-to-fine supervision. The resulting method establishes the new state-of-the-art on both MS COCO and MPII Human Pose dataset, justifying the effectiveness of a multi-stage architecture. The source code is publicly available for further research.

PaperPDFCode

In Syntology View this paper on Syntology, its page in Syntology's graph. That page lists the repositories linked to the paper, the abstract and the calls for agents.

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

For agents, on Syntology's MCP service (how to connect):

Code

megvii-detection/MSPN officialmentioned in papermentioned on GitHubpytorchSyntology: not harvested report
chenyilun95/tf-cpn mentioned on GitHubtfMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Learning Delicate Local Representations for Multi-Person Pose Estimation (arXiv:2003.04030; that paper's own run record: 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 2 violated, 0 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. report
fenglinglwb/MSPN mentioned on GitHubpytorchSyntology: not harvested report
hyperionfalling/lightpose mentioned on GitHubpytorchSyntology: not harvested report
open-mmlab/mmpose pytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to RTMO: Towards High-Performance One-Stage Real-Time Multi-Person Pose Estimation (arXiv:2312.07526; that paper's own run record: 2 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
xiuyu0000/papers_with_examples mindsporeSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Lite-HRNet: A Lightweight High-Resolution Network (arXiv:2104.06403; that paper's own run record: 13 ran (of which 2 constructed an object rather than computing a result; 11 with no instrument failure: 0 honoured, 0 violated, 11 with no contract checked; 2 where Syntology's instrument failed) · 16 unverified); nothing checks that they implement this paper's method. report
yangyucheng000/MSPN mindsporeSyntology: not harvested 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

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

Tasks

Keypoint DetectionPose Estimation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Keypoint Detection COCO (Common Objects in Context) MSPN(384x288) Test AP 76.1 #6 of 24 Archive leaderboard report
Keypoint Detection COCO test-challenge MSPN+* AP 76.4 #2 of 8 Archive leaderboard report
Keypoint Detection COCO test-challenge MSPN+* AP50 92.9 #2 of 8 Archive leaderboard report
Keypoint Detection COCO test-challenge MSPN+* AP75 82.6 #2 of 8 Archive leaderboard report
Keypoint Detection COCO test-challenge MSPN+* APL 88.6 #2 of 8 Archive leaderboard report
Keypoint Detection COCO test-challenge MSPN+* AR 82.2 #2 of 8 Archive leaderboard report
Keypoint Detection COCO test-challenge MSPN+* AR50 96 #2 of 8 Archive leaderboard report
Keypoint Detection COCO test-challenge MSPN+* AR75 87.7 #2 of 8 Archive leaderboard report
Keypoint Detection COCO test-challenge MSPN+* ARL 83.2 #2 of 8 Archive leaderboard report
Keypoint Detection COCO test-challenge MSPN+* ARM 77.5 #2 of 8 Archive leaderboard report
Keypoint Detection COCO test-dev MSPN AP 76.1 #3 of 16 Archive leaderboard report
Keypoint Detection COCO test-dev MSPN AP50 93.4 #3 of 16 Archive leaderboard report
Keypoint Detection COCO test-dev MSPN AP75 83.8 #3 of 16 Archive leaderboard report
Keypoint Detection COCO test-dev MSPN APL 81.5 #3 of 16 Archive leaderboard report
Keypoint Detection COCO test-dev MSPN APM 72.3 #3 of 16 Archive leaderboard report
Keypoint Detection COCO test-dev MSPN AR 81.6 #3 of 16 Archive leaderboard report
Keypoint Detection COCO test-dev MSPN AR50 96.3 #3 of 16 Archive leaderboard report
Keypoint Detection COCO test-dev MSPN AR75 88.1 #3 of 16 Archive leaderboard report
Keypoint Detection COCO test-dev MSPN ARL 87.1 #3 of 16 Archive leaderboard report
Keypoint Detection COCO test-dev MSPN ARM 77.5 #3 of 16 Archive leaderboard report
Pose Estimation COCO minival MSPN AP 75.9 #1 of 1 Archive leaderboard report
Pose Estimation COCO test-dev MSPN AP 76.1 #18 of 47 Archive leaderboard report
Pose Estimation COCO test-dev MSPN AP50 93.4 #18 of 47 Archive leaderboard report
Pose Estimation COCO test-dev MSPN AP75 83.8 #18 of 47 Archive leaderboard report
Pose Estimation COCO test-dev MSPN APL 81.5 #18 of 47 Archive leaderboard report
Pose Estimation COCO test-dev MSPN APM 72.3 #18 of 47 Archive leaderboard report
Pose Estimation COCO test-dev MSPN AR 81.6 #18 of 47 Archive leaderboard report
Pose Estimation MPII Human Pose MSPN PCKh-0.5 92.6 #9 of 46 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