Skip to main content
. 2025 Aug 11;16:1596408. doi: 10.3389/fneur.2025.1596408

Figure 6.

Two bar charts compare performance metrics on the AFLW and PoseTrack datasets for different models. Metrics include Accuracy, Recall, F1 Score, and AUC. Models like HRNet, SimpleBaseline, DarkPose, OpenPose, DEKR, PRTR, and an unnamed model labeled “Ours” are shown. “Ours” performs highest in both datasets across all metrics. Performance is measured as a percentage, with the left chart focusing on AFLW and the right on PoseTrack.

Analysis of the performance of SOTA techniques on the AFLW and PoseTrack datasets.