🧪 Test?View on arXiv
Uncovering the Blind Spots of OKS-mAP: A Unified Protocol for Occlusion, Joint-Angle, and Temporal Evaluation of 2D Pose Estimators
Not provided in the abstract
pose estimationevaluation metricsocclusionjoint-angle analysis
2611.12911
Builder Relevance
8h ago80%
Abstract
The paper proposes a supplementary protocol to address the limitations of the OKS-mAP metric in 2D human pose estimation.
Reality Card
Core Claim
The proposed protocol reveals that aggregate mAP overestimates occluded-scene performance and introduces metrics for joint-angle error and temporal stability.
Method / Result
Delta_occ is negative in every combination tested, indicating that aggregate mAP masks occlusion-specific differences.
Limitations
The paper does not specify the authors, which may limit reproducibility and verification of results.
Paper to code
Verified implementation resources so builders can test the paper’s claims instead of stopping at the abstract.
No verified implementation link has been attached yet. AIBuzzHub will keep this panel separate from unverified search results.