Research map / Video / Video segmentation and tracking
Forecasting Animal Motion
Neerja Thakkar, Shiry Ginosar, Jacob Walker, Jitendra Malik, Joao Carreira, Carl Doersch
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 |
|---|---|---|---|
| train on full all-species data instead of Panthera subset only (ablation) | Ours (uncond) · MammalMotion high motion | FVMD: 84.8 -> 49.3 | Improved |
| train on full all-species data instead of Panthera subset only (ablation) | Ours (uncond) · MammalMotion high motion | PWT: 20.6 -> 26.0 | Improved |
| add optional ground-truth average displacement conditioning (ablation) | Ours · MammalMotion high motion | ADE: 0.119 -> 0.068 | Improved |
| use trajectory diffusion transformer with velocity parameterization vs Track2Act | Ours (uncond) · MammalMotion combined | ADE: 0.053 -> 0.046 | Improved |
| no-motion baseline vs learned models on low/medium motion | No motion · MammalMotion combined | ADE: 0.099 vs Ours 0.046 | No clear effect |
| forecast point tracks instead of pixels (compare with video diffusion) | Ours · qualitative vs Stable Video Diffusion | realism | Improved |
| DDIM sampling with 100 steps instead of 1000 (ablation) | Ours · inference | speed: 10x speedup | Improved |
Closest papers by meaning
- MolmoMotion: Forecasting Point Trajectories in 3D with Language Instruction (NeurIPS 2026 · similarity 0.87)
- BEAST3D: animal behavioral analysis and neural encoding from multi-view video via Gaussian splatting (NeurIPS 2026 · similarity 0.86)
- PlaySlot: Learning Inverse Latent Dynamics for Controllable Object-Centric Video Prediction and Planning (ICML 2025 · similarity 0.85)
- MOVEBENCH: A Benchmark for Global-Scale Wildlife Movement Forecasting (NeurIPS 2026 · similarity 0.85)
- DINO-WM: World Models on Pre-trained Visual Features enable Zero-shot Planning (ICML 2025 · similarity 0.84)
- Agent-to-Sim: Learning Interactive Behavior Models from Casual Longitudinal Videos (ICLR 2025 · similarity 0.84)
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.
