ECCV 2026: EventKitchen Stereo Event Camera Dataset
rsasaki0109 · x · 2026-08-15
EventKitchen is a large-scale stereo event camera dataset focusing on human cooking activities in kitchen scenarios, collected by TU Delft and others, accepted as an ECCV 2026 Spotlight paper.
- Scale: 5.5 hours of synchronized stereo event recordings, supporting tasks like action recognition, object detection, and stereo depth estimation.
- Sensors: Utilizes Prophesee Gen4 event camera pairs, CMOS camera pairs, and RealSense depth cameras to provide high temporal resolution and dynamic range data.
- Application: Fills the gap for human-centric daily-life scenarios in neuromorphic vision, including 10,762 action segments and 13,482 bounding box annotations.
Related event: EventKitchen Stereo Event-Camera Dataset Named ECCV 2026 Spotlight(2 posts)→
More from Multimodal
- How meloform works: WASM pitch detection plus differentiable circuits learning 1,022 sine coefficients — henloitsjoyce · 2026-10-16
- Documentary filmmaker urges peers to watch 'You Can See Everything' — kyliebytes · 2026-10-16
- Redditor's first Cumfy/Seedance video nails a cinematic one-shot generation — Wave_N_Wind · 2026-10-16
- Kling 4 returns after 8 months: 10 hyper-realistic AI ads for under $5 — CurieuxExplorer · 2026-10-16
- Shanghai AI Lab unveils Intern Lumina U2, a diffusion LLM unifying image, video and 3D tasks — Shanghai-AI-Laboratory · 2026-10-16
- NAVER AI Lab proposes Speculative Forcing to speed up autoregressive video generation — naver-ai · 2026-10-16