Research map / Detection and segmentation / CNN features
Cooking beyond Frames: A Stereo Event Camera Dataset in the Kitchen
Chengming Feng, Hesam Araghi, Liming Zheng, Julien Dupeyroux, Xucong Zhang, Jan van Gemert, Nergis Tomen
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 Swin-B instead of TSM | action recognition baseline · EventKitchen 69 actions | top-1 accuracy: 19.23 -> 24.69 | Improved |
| use Swin-B instead of TSM | action recognition baseline · EventKitchen 69 actions | top-5 accuracy: 42.26 -> 56.48 | Improved |
| use Swin-B instead of TSM | action recognition baseline · EventKitchen 18 verbs | top-1 accuracy: 38.17 -> 46.08 | Improved |
| predict verbs only instead of full actions | Swin · EventKitchen | top-1 accuracy: 24.69 -> 46.08 | Improved |
| evaluate YOLOv10 on events instead of COCO RGB | YOLOv10 · EventKitchen vs COCO | AP: 54.4% -> 16.2% | Hurt |
| evaluate RVT on EventKitchen instead of 1 Mpx | RVT · EventKitchen vs 1 Mpx | AP: 47.4% -> 7.6% | Hurt |
| evaluate EvRT-DETR on EventKitchen instead of 1 Mpx | EvRT-DETR · EventKitchen vs 1 Mpx | AP: 50.1% -> 8.5% | Hurt |
| use event-specialized detectors (RVT) instead of frame-based YOLOv10 | object detection · EventKitchen | AP: 16.2 -> 7.6 | Hurt |
| increase SE-CFF stack size from 5M to 15M events at 1 Hz (ablation) | SE-CFF · EventKitchen stereo depth | RMSE: 88.21 -> 84.91 | Improved |
| increase SE-CFF stack size from 5M to 15M events at 1 Hz (ablation) | SE-CFF · EventKitchen stereo depth | MAE: 57.38 -> 54.84 | Improved |
| increase sampling rate from 1 Hz to 3 Hz at 5M events (ablation) | SE-CFF · EventKitchen stereo depth | RMSE: 88.21 -> 88.19 | No clear effect |
| use SE-CFF trained on events instead of zero-shot FoundationStereo on E2VID images | stereo depth · EventKitchen | RMSE: 155.06 -> 84.91 | Improved |
Closest papers by meaning
- DailyDVS-200: A Comprehensive Benchmark Dataset for Event-Based Action Recognition (ECCV 2024 · similarity 0.89)
- EPFL-Smart-Kitchen: An Ego-Exo Multi-Modal Dataset for Challenging Action and Motion Understanding in Video-Language Models (NeurIPS 2025 · similarity 0.87)
- EventZoom: A Progressive Approach to Event-Based Data Augmentation for Enhanced Neuromorphic Vision (AAAI 2025 · similarity 0.86)
- HARDVS: Revisiting Human Activity Recognition with Dynamic Vision Sensors (AAAI 2024 · similarity 0.85)
- Ego-Exo4D: Understanding Skilled Human Activity from First- and Third-Person Perspectives (CVPR 2024 · similarity 0.84)
- PASS: Path-selective State Space Model for Event-based Recognition (NeurIPS 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.
