Research map / Video / Video segmentation and tracking
NeMo-map: Neural Implicit Flow Fields for Spatio-Temporal Motion Mapping
Yufei Zhu, Shih-Min Yang, Andrey Rudenko, Tomasz Piotr Kucner, Achim J. Lilienthal, Martin Magnusson
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 |
|---|---|---|---|
| use continuous neural implicit MoD instead of CLiFF-map | MoD · ATC NLL | NLL: 2.945 -> 1.844 | Improved |
| use continuous neural implicit MoD instead of Online CLiFF-map | MoD · ATC NLL | NLL: 2.556 -> 1.844 | Improved |
| use continuous neural implicit MoD instead of STeF-map | MoD · ATC NLL | NLL: 5.371 -> 1.844 | Improved |
| use NeMo-map instead of CLiFF-map | MoD · ETH/UCY HOTEL NLL | NLL: 1.700 -> 0.096 | Improved |
| use NeMo-map instead of CLiFF-map | MoD · ETH/UCY UNIV NLL | NLL: 1.601 -> 1.436 | Improved |
| use NeMo-map instead of CLiFF-map for map building | MoD · ATC map building | train time (minutes): 1831 -> 19.26 | Improved |
| use NeMo-map instead of STeF-map for map building | MoD · ATC map building | train time (minutes): 0.815 -> 19.26 | Hurt |
| use NeMo-map prior instead of CLiFF-LHMP | long-term motion prediction · ATC LHMP 60 s | ADE: 5.45 -> 5.08 | Improved |
| use NeMo-map prior instead of CLiFF-LHMP | long-term motion prediction · ATC LHMP 60 s | FDE: 11.45 -> 10.97 | Improved |
| use MoD prior instead of TUTR transformer | long-term motion prediction · ATC LHMP 60 s | ADE: 12.10 -> 5.08 | Improved |
| use SIREN time encoding instead of temporal grid (ablation) | NeMo-map · ATC NLL | NLL: 1.946 -> 1.844 | Improved |
| use SIREN time encoding instead of Fourier features (ablation) | NeMo-map · ATC NLL | NLL: 1.928 -> 1.844 | Improved |
Closest papers by meaning
- SplatFlow: Self-Supervised Dynamic Gaussian Splatting in Neural Motion Flow Field for Autonomous Driving (CVPR 2025 · similarity 0.85)
- Homogeneous Dynamics Space for Heterogeneous Humans (CVPR 2025 · similarity 0.85)
- VidMan: Exploiting Implicit Dynamics from Video Diffusion Model for Effective Robot Manipulation (NeurIPS 2024 · similarity 0.84)
- SingularTrajectory: Universal Trajectory Predictor Using Diffusion Model (CVPR 2024 · similarity 0.84)
- Universal Humanoid Motion Representations for Physics-Based Control (ICLR 2024 · similarity 0.84)
- World Motion Models: Flexible Sequence Modeling of SE(3) Trajectories (NeurIPS 2026 · 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.
