Research map / Detection and segmentation / Tracking
Virtual Worlds as Proxy for Multi-Object Tracking Analysis
Check 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 |
|---|---|---|---|
| Evaluate on cloned virtual sequences instead of the corresponding real sequences | DP-MCF and MDP · KITTI seed sequences and Virtual KITTI clones | MOTA: 81.7% -> 82.2%; 85.9% -> 86.0% | No clear effect |
| Pre-train on virtual clones, then fine-tune on real seed videos | DP-MCF · Held-out KITTI videos | MOTA: 71.9% -> 76.7% | Helped |
| Pre-train on virtual clones, then fine-tune on real seed videos | MDP · Held-out KITTI videos | MOTA: 78.1% -> 78.7% | Helped |
| Train on virtual clones only instead of real seed videos | DP-MCF · Held-out KITTI videos | MOTA: 71.9% -> 64.3% | Hurt |
| Train on virtual clones only instead of real seed videos | MDP · Held-out KITTI videos | MOTA: 78.1% -> 63.7% | Hurt |
| Rotate the camera 15 degrees right | DP-MCF · Virtual KITTI clones | MOTA: -2.9% | Hurt |
| Rotate the camera 15 degrees left | DP-MCF · Virtual KITTI clones | MOTA: -8.1% | Hurt |
| Change lighting to morning | DP-MCF · Virtual KITTI clones | MOTA: -2.8% | Hurt |
| Change lighting to sunset | DP-MCF · Virtual KITTI clones | MOTA: -6.8% | Hurt |
| Simulate overcast weather | DP-MCF · Virtual KITTI clones | MOTA: -2.0% | Hurt |
| Simulate fog | DP-MCF · Virtual KITTI clones | MOTA: -45.2% | Hurt |
| Simulate rain | DP-MCF · Virtual KITTI clones | MOTA: -7.8% | Hurt |
Closest papers by meaning
- TrackingWorld: World-centric Monocular 3D Tracking of Almost All Pixels (NeurIPS 2025 · similarity 0.86)
- Instance Tracking in 3D Scenes from Egocentric Videos (CVPR 2024 · similarity 0.85)
- VastTrack: Vast Category Visual Object Tracking (NeurIPS 2024 · similarity 0.84)
- M3SOT: Multi-Frame, Multi-Field, Multi-Space 3D Single Object Tracking (AAAI 2024 · similarity 0.83)
- From an Image to a Scene: Learning to Imagine the World from a Million 360° Videos (NeurIPS 2024 · similarity 0.83)
- Leader360V: A Large-scale, Real-world 360 Video Dataset for Multi-task Learning in Diverse Environment (NeurIPS 2025 · similarity 0.83)
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.