Research map / Video / Video segmentation and tracking
Learning Transferable Subspace for Human Motion Segmentation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 domain-invariant projection P to reconstruction (ablation) | transfer subspace clustering · Keck (MAD source) | clustering accuracy: 0.5224 -> 0.5415 | Helped |
| use source data only as dictionary instead of source plus target (ablation) | transfer subspace clustering · Keck (MAD source) | clustering accuracy: 0.4530 -> 0.5415 | Helped |
| add Gaussian noise N(0,1) to source dictionary (ablation) | transfer subspace clustering · MAD (Weizmann source) | clustering accuracy: 0.5736 -> 0.5251 | Hurt |
| shuffle the order of the source dictionary sequence (ablation) | transfer subspace clustering · MAD (Weizmann source) | clustering accuracy: 0.5736 -> 0.5710 | No clear effect |
| concatenate source and target data as input to TSC instead of target only | TSC · MAD, Keck, Weizmann | clustering accuracy: 0.5556 -> 0.5418 on MAD | Hurt |
| use transferable subspace clustering instead of TSC | temporal subspace clustering · MAD | NMI: 0.7721 -> 0.8286 | Helped |
| use transferable subspace clustering instead of TSC | temporal subspace clustering · Weizmann | clustering accuracy: 0.6111 -> 0.6208 | No clear effect |
Closest papers by meaning
- Probabilistic Temporal Subspace Clustering (CVPR 2017 · similarity 0.93)
- Multi-Mutual Consistency Induced Transfer Subspace Learning for Human Motion Segmentation (CVPR 2020 · similarity 0.90)
- Learned Trajectory Embedding for Subspace Clustering (CVPR 2024 · similarity 0.88)
- Unsupervised Action Segmentation by Joint Representation Learning and Online Clustering (CVPR 2022 · similarity 0.87)
- Learning Segmentation from Point Trajectories (NeurIPS 2024 · similarity 0.87)
- Exploring a Principled Framework for Deep Subspace Clustering (ICLR 2025 · similarity 0.86)
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.