# Online Pose Estimation on Sub-JHMDB (test) Source: [Online Pose Estimation on Sub-JHMDB (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:46:25.121Z Evaluation of online (real-time) human pose estimation performance on the test split of the Sub-JHMDB dataset, reporting joint accuracy, average performance, and latency metrics (Max/Mean Delay, RT Frames). A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Online pose estimation | | Dataset (including any stated split) | Sub-JHMDB (test) | | Metric | Head Acc | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 6 of 6 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Head Acc](): Higher is better; unit: PERCENT. - [Shoulder Acc](): Higher is better; unit: PERCENT. - [Elbow Acc](): Higher is better; unit: PERCENT. - [Wrist Acc](): Higher is better; unit: PERCENT. - [Hip Acc](): Higher is better; unit: PERCENT. - [Knee Acc](): Higher is better; unit: PERCENT. - [Ankle Acc](): Higher is better; unit: PERCENT. - [Average Accuracy](): Higher is better; unit: PERCENT. - [Max Latency (ms)](): Lower is better; unit: TIME. - [Mean Latency (ms)](): Lower is better; unit: TIME. - [RT Frames (%)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Head Acc | Result date | Report paper | | --- | --- | --- | --- | | main network + stream-aware | 99.4 | Not recorded | 68fa6f0a1e209131cea43848 | | main network + SAN | 99.1 | Not recorded | 68fa6f0a1e209131cea43848 | | Skip-Convolution | 98.7 | Not recorded | 68fa6f0a1e209131cea43848 | | DKD | 98.3 | Not recorded | 68fa6f0a1e209131cea43848 | | SimpleBaseline | 97.5 | Not recorded | 68fa6f0a1e209131cea43848 | | main network + random policy | 97.5 | Not recorded | 68fa6f0a1e209131cea43848 | ### main network + stream-aware Head Acc: 99.4. Result ID: 69907fa83a6c31484900ff5b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"policy":"stream-aware"}. This record does not assert independent reproduction. ### main network + SAN Head Acc: 99.1. Result ID: 69907fa83a6c31484900ff5c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"policy":"SAN"}. This record does not assert independent reproduction. ### Skip-Convolution Head Acc: 98.7. Result ID: 69907fa83a6c31484900ff59. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DKD Head Acc: 98.3. Result ID: 69907fa83a6c31484900ff57. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### SimpleBaseline Head Acc: 97.5. Result ID: 69907fa83a6c31484900ff58. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### main network + random policy Head Acc: 97.5. Result ID: 69907fa83a6c31484900ff5a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"policy":"random policy"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.