Research map / Video / Video segmentation and tracking
Behavior Discovery and Alignment of Articulated Object Classes from Unstructured Video
Luca Del Pero, Susanna Ricco, Rahul Sukthankar, Vittorio Ferrari
Read the paperCheck your idea against 137,217 papers
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 pauses+periodicity partitioning instead of whole shots (ablation) | behavior discovery · tigers | average interval uniformity: 0.78 -> 0.87 | Improved |
| add pauses+periodicity partitioning instead of whole shots (ablation) | behavior discovery · horses | average interval uniformity: 0.82 -> 0.89 | Improved |
| add pauses+periodicity partitioning instead of whole shots (ablation) | behavior discovery · dogs | average interval uniformity: 0.72 -> 0.88 | Improved |
| add periodicity to pause-based partitioning (ablation) | behavior discovery · dogs | number of intervals: 115 -> 219 | Improved |
| add periodicity to pause-based partitioning (ablation) | behavior discovery · tigers | number of intervals: 719 -> 885 | Improved |
| add MBH to PoTs for CMP extraction (ablation) | CMP extraction · tigers | alignable CMPs (%): 43% -> 51% | Improved |
| add MBH to PoTs for CMP extraction (ablation) | CMP extraction · horses | alignable CMPs (%): 47% -> 49% | Improved |
| extract CMPs using PoT-based clustering instead of uniform sampling (ablation) | CMP extraction · tigers | alignable CMPs (%): 19% -> 51% | Improved |
| extract CMPs using PoT-based clustering instead of uniform sampling (ablation) | CMP extraction · horses | alignable CMPs (%): 26% -> 49% | Improved |
| add foreground-mask regularization (+FG) to trajectory and SIFT homography alignment (ablation) | sequence alignment · tigers and horses | precision-recall | Improved |
| add time-varying TPS (TTPS) on top of TM+FG homography (ablation) | sequence alignment · tigers and horses | precision-recall | Improved |
| use TTPS+FG instead of SIFT+FG | sequence alignment · horses | average precision: 0.235 -> 0.265 | Improved |
Closest papers by meaning
- Discovering the Physical Parts of an Articulated Object Class From Multiple Videos (CVPR 2016 · similarity 0.91)
- Ponymation: Learning Articulated 3D Animal Motions from Unlabeled Online Videos (ECCV 2024 · similarity 0.88)
- Pose Splatter: A 3D Gaussian Splatting Model for Quantifying Animal Pose and Appearance (NeurIPS 2025 · similarity 0.86)
- Emergence of a Shared Canonical Object Frame from In-the-Wild Videos (NeurIPS 2026 · similarity 0.85)
- BEAST3D: animal behavioral analysis and neural encoding from multi-view video via Gaussian splatting (NeurIPS 2026 · similarity 0.85)
- Articulation in Prime: Primitive-Based Articulated Object Understanding from a Single Casual Video (NeurIPS 2026 · 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.
