# Motion Question Answering on HuMMan-QA Source: [Motion Question Answering on HuMMan-QA]() Collection index: [llms.txt]() Data updated: 2026-06-23T07:36:15.615Z Evaluation of model performance on the HuMMan-QA dataset for Motion Question Answering tasks, reporting detailed accuracy breakdowns across query types (Action, Direction, Body Part). 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 | Motion Question Answering | | Dataset (including any stated split) | HuMMan-QA | | Metric | Overall Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 5 of 5 | | 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 - [Overall Accuracy](): Higher is better; unit: PERCENT. - [Accuracy (Action, All)](): Higher is better; unit: PERCENT. - [Accuracy (Action, Before)](): Higher is better; unit: PERCENT. - [Accuracy (Action, After)](): Higher is better; unit: PERCENT. - [Accuracy (Action, BTW)](): Higher is better; unit: PERCENT. - [Accuracy (Direction, All)](): Higher is better; unit: PERCENT. - [Accuracy (Direction, Before)](): Higher is better; unit: PERCENT. - [Accuracy (Direction, After)](): Higher is better; unit: PERCENT. - [Accuracy (Body Part, All)](): Higher is better; unit: PERCENT. - [Accuracy (Body Part, Before)](): Higher is better; unit: PERCENT. - [Accuracy (Body Part, After)](): Higher is better; unit: PERCENT. - [Accuracy (Body Part, BTW)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Overall Accuracy | Result date | Report paper | | --- | --- | --- | --- | | FiGMo | 82.4 | Not recorded | 6a3a1314206542251d7c6fd5 | | FiGMo | 79.2 | Not recorded | 6a3a1314206542251d7c6fd5 | | IMoRe II | 73 | Not recorded | 6a3a1314206542251d7c6fd5 | | IMoRe I | 71.9 | Not recorded | 6a3a1314206542251d7c6fd5 | | NSPose | 69.1 | Not recorded | 6a3a1314206542251d7c6fd5 | ### FiGMo Overall Accuracy: 82.4. Result ID: 6a3a3767fbc373592c341d7e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Motion representation":"3D","Fine-tuning protocol":"fine-tuned on HuMMan-QA"}. This record does not assert independent reproduction. ### FiGMo Overall Accuracy: 79.2. Result ID: 6a3a3767fbc373592c341d7d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Motion representation":"2D","Fine-tuning protocol":"fine-tuned on HuMMan-QA"}. This record does not assert independent reproduction. ### IMoRe II Overall Accuracy: 73. Result ID: 6a3a3767fbc373592c341d7c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Fine-tuning protocol":"fine-tuned on HuMMan-QA"}. This record does not assert independent reproduction. ### IMoRe I Overall Accuracy: 71.9. Result ID: 6a3a3767fbc373592c341d7b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Fine-tuning protocol":"fine-tuned on HuMMan-QA"}. This record does not assert independent reproduction. ### NSPose Overall Accuracy: 69.1. Result ID: 6a3a3767fbc373592c341d7a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Fine-tuning protocol":"fine-tuned on HuMMan-QA"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.