Research map / 3D vision / 3D generation
Adding Knowledge to Unsupervised Algorithms for the Recognition of Intent
Stuart Synakowski, Qianli Feng, Aleix M. Martínez
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add Concept 2 (external-force motion is non-intentional) to C1 (ablation) | intent algorithm · intent-maya | classification accuracy: 0.500 -> 0.667 | Improved |
| add Concept 2 to C1 (ablation) | intent algorithm · intent-mocap | classification accuracy: 0.500 -> 0.571 | Improved |
| add Concept 3 (EFM caused by SPM is intentional) to C1+C2 (ablation) | intent algorithm · intent-maya | classification accuracy: 0.667 -> 0.667 | No clear effect |
| add Concept 3 to C1+C2 (ablation) | intent algorithm · intent-mocap | classification accuracy: 0.571 -> 0.534 | Hurt |
| add Concept 4 (inertia of intentionality) to C1+C2+C3 (ablation) | intent algorithm · intent-maya | classification accuracy: 0.667 -> 0.950 | Improved |
| add Concept 4 to C1+C2+C3 (ablation) | intent algorithm · intent-mocap | classification accuracy: 0.534 -> 0.827 | Improved |
| add Concept 4 to C1+C2+C3 (ablation) | intent algorithm · intent-youtube | classification accuracy: 0.501 -> 0.785 | Improved |
| add Concept 3 when Concept 4 is present (ablation) | intent algorithm · intent-maya | classification accuracy: 0.783 -> 0.950 | Improved |
| add Concept 3 when Concept 4 is present (ablation) | intent algorithm · intent-youtube | classification accuracy: 0.735 -> 0.785 | Improved |
| occlude one random joint per frame (noisy center of mass) (ablation) | intent algorithm · intent-mocap | classification accuracy: 0.827 -> 0.712 | Hurt |
| occlude one random joint per frame (noisy center of mass) (ablation) | intent algorithm · intent-youtube | classification accuracy: 0.785 -> 0.624 | Hurt |
| occlude one random joint per agent (ablation) | intent algorithm · intent-mocap | classification accuracy: 0.827 -> 0.808 | No clear effect |
Closest papers by meaning
- What Actions Are Needed for Understanding Human Actions in Videos? (ICCV 2017 · similarity 0.86)
- About Time: Advances, Challenges, and Outlooks of Action Understanding (IJCV 2025 · similarity 0.86)
- Oops! Predicting Unintentional Action in Video (CVPR 2020 · similarity 0.86)
- Intend to Move: A Multimodal Dataset for Intention-Aware Human Motion Understanding (NeurIPS 2025 · similarity 0.85)
- AVA: A Video Dataset of Spatio-Temporally Localized Atomic Visual Actions (CVPR 2018 · similarity 0.85)
- Learning Features by Watching Objects Move (CVPR 2017 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
