Research map / Video / Video segmentation and tracking
Asynchronous, Photometric Feature Tracking using Events and Frames
Daniel Gehrig, Henri Rebecq, Guillermo Gallego, Davide Scaramuzza
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 raw frame gradients with a generative event model instead of Canny-edge ICP | Kueng et al. · Event Camera Dataset, shapes 6dof | average pixel error (px): 1.75 -> 0.64 | Improved |
| use photometric registration instead of ICP on Canny edges | Kueng et al. · Event Camera Dataset, checkerboard | average feature age (s): 2.76 -> 8.23 | Improved |
| use photometric registration instead of ICP on Canny edges | Kueng et al. · Event Camera Dataset, boxes 6dof (high texture) | average pixel error (px): 3.10 -> 0.90 | Improved |
| use photometric registration instead of ICP on Canny edges | Kueng et al. · Event Camera Dataset, bicycles (natural) | average pixel error (px): 3.65 -> 0.75 | Improved |
| use frame-based generative template instead of event-based EM-ICP template | Zhu et al. · Event Camera Dataset, rocks (natural) | average pixel error (px): 3.24 -> 0.80 | Improved |
| use frame-based generative template instead of event-based EM-ICP template | Zhu et al. · Event Camera Dataset, rocks (natural) | average feature age (s): 1.13 -> 0.78 | Hurt |
| use photometric registration instead of EM-ICP | Zhu et al. · Event Camera Dataset, boxes 6dof | average feature age (s): 1.56 -> 1.56 | No clear effect |
| use photometric registration instead of ICP on Canny edges | Kueng et al. · Event Camera Dataset, bicycles (natural) | average feature age (s): 0.49 -> 1.15 | Improved |
| use photometric registration instead of ICP on Canny edges | Kueng et al. · Event Camera Dataset, rocks (natural) | average feature age (s): 0.85 -> 0.78 | Hurt |
| evaluate on simulated rather than real data | Ours · Simulated vs real Event Camera Dataset | average pixel error (px): 0.4 pixel on average | Improved |
Closest papers by meaning
- EKLT: Asynchronous Photometric Feature Tracking Using Events and Frames (IJCV 2019 · similarity 0.97)
- Data-Driven Feature Tracking for Event Cameras With and Without Frames (TPAMI 2025 · similarity 0.92)
- Event-Intensity Stereo: Estimating Depth by the Best of Both Worlds (ICCV 2021 · similarity 0.90)
- Data-Driven Feature Tracking for Event Cameras (CVPR 2023 · similarity 0.90)
- Event-Based Motion Segmentation by Motion Compensation (ICCV 2019 · similarity 0.90)
- ETAP: Event-based Tracking of Any Point (CVPR 2025 · similarity 0.88)
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.
