Research map / Video / Video segmentation and tracking
TETO: Tracking Events with Teacher Observation for Motion Estimation and Frame Interpolation
Jini Yang, Eunbeen Hong, Soowon Son, Hyunkoo Lee, Sunghwan Hong, Sunok Kim, Seungryong Kim
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 object-motion query sampling (M_obj) (ablation) | TETO · EVIMO2 | AJ: 60.3 -> 67.9 | Improved |
| add temporal augmentation (ablation) | TETO · EVIMO2 | AJ: 63.3 -> 67.9 | Improved |
| add flow loss L_flow (ablation) | TETO · EVIMO2 | AJ: 66.3 -> 67.9 | Improved |
| add feature loss encouraging RGB-like representations (ablation) | TETO · EVIMO2 | AJ: 67.9 -> 63.7 | Hurt |
| add concentration network before pretrained encoder (ablation) | TETO · EVIMO2 | AJ: 56.5 -> 67.9 | Improved |
| add flow warping F_warp (ablation) | TETO-VFI · BS-ERGB x3 | FID: 13.88 -> 10.11 | Improved |
| add trajectory attention loss L_attn (ablation) | TETO-VFI · BS-ERGB x3 | FID: 10.82 -> 10.11 | Improved |
| add event motion mask M_event (ablation) | TETO-VFI · BS-ERGB x3 | FID: 11.06 -> 10.11 | Improved |
| use TETO motion estimator instead of E2FAI (ablation) | TETO-VFI · BS-ERGB x3 | FID: 13.34 -> 10.11 | Improved |
| add in-domain pseudo-label fine-tuning on DSEC (ablation) | TETO · DSEC | EPE: 2.15 -> 1.39 | Improved |
| train on 25 minutes of real event data with RGB teacher instead of synthetic data (ETAP) (ablation) | TETO · EVIMO2 | AJ: 66.1 -> 67.9 | Improved |
| use TETO motion conditioning instead of CBMNet-Large for VFI (ablation) | TETO-VFI · HQ-EVFI x15 | FID: 26.19 -> 19.89 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- ETAP — ETAP: Event-based Tracking of Any Point (CVPR 2025) · better on 4 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- TimeTracker: Event-based Continuous Point Tracking for Video Frame Interpolation with Non-linear Motion (CVPR 2025 · similarity 0.87)
- Event Stream-based Visual Object Tracking: A High-Resolution Benchmark Dataset and A Novel Baseline (CVPR 2024 · similarity 0.87)
- Track-On: Transformer-based Online Point Tracking with Memory (ICLR 2025 · similarity 0.86)
- E-Motion: Future Motion Simulation via Event Sequence Diffusion (NeurIPS 2024 · similarity 0.86)
- Repurposing Pre-trained Video Diffusion Models for Event-based Video Interpolation (CVPR 2025 · similarity 0.86)
- TTA-EVF: Test-Time Adaptation for Event-based Video Frame Interpolation via Reliable Pixel and Sample Estimation (CVPR 2024 · 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.
