LeBoardLeaderboard / nyu door opening surprising effectiveness
nyu door opening surprising effectiveness
Camera dropoutShare of frames from live episode cameras that are almost entirely one flat black or white value. A camera blank for a whole episode is counted as absent, shown underneath, not as a dropout. Method →
0
0 frames
BlurShare of camera frames inside an interval at least 0.5 s long where the image is much blurrier than that camera's typical frame. Method →
1.2%
250 frames
DarknessShare of camera frames inside an interval at least 0.5 s long where 60% or more of the pixels are near-black. Method →
0
0 frames
FreezeShare of camera frames inside an interval at least 1 s long where the video repeats the exact same image. Method →
0
0 frames
Duration mismatchShare of assessed episode-camera pairs whose metadata video length disagrees with the Parquet frame count by more than 1 frame. A hatched cell with * means only some pairs could be assessed (the small number shows how many); the rest are unknown, not clean. –* means none could be assessed. Method →
0
0 of 484 assessed
Flagged episodesEpisodes flagged by any episode check: actions that never change, a robot that barely moves, non-finite values, broken row indexes or timestamp gaps. The count is on top, its share of assessed episodes underneath. Method →
0
of 484 assessed episodes
All checks
| Check | Flagged | Share | Coverage |
|---|---|---|---|
| Files 1 | |||
| Video decoding | 0 | 0 | complete |
| Episodes 484 | |||
| Episode boundaries | 0 | 0 | complete |
| Finite action and state | 0 | 0 | complete |
| Timestamp gaps | 0 | 0 | complete |
| Static actions | 0 | 0 | complete |
| Idle episodes | – | – | not measured |
| Episode-camera pairs 484 | |||
| Duration mismatch | 0 | 0 | complete |
| Video/action sync | – | – | not measured |
| Frames 20,405 | |||
| Camera absent | 0 | 0 | complete |
| Camera dropout | 0 | 0 | complete |
| Blur | 25056 intervals | 1.2% | complete |
| Darkness | 0 | 0 | complete |
| Freeze | 0 | 0 | complete |
By camera
| Camera | Frames | Absent | Dropout | Blur | Darkness | Freeze |
|---|---|---|---|---|---|---|
| image | 20,405 | 0 | 0 | 1.2%56 intervals | 0 | 0 |
Evidence
Longest video intervals up to 4 per signal · times are within the video file
| Signal | Camera | Episode | File | Start | End | Duration | |
|---|---|---|---|---|---|---|---|
| Blur | image | 477 | file-000.mp4 | 111:54.7 | 111:58.3 | 3.7 s | Inspect → |
| Blur | image | 306 | file-000.mp4 | 70:11.3 | 70:13.7 | 2.3 s | Inspect → |
| Blur | image | 82 | file-000.mp4 | 18:09.3 | 18:11.3 | 2.0 s | Inspect → |
| Blur | image | 119 | file-000.mp4 | 26:43.0 | 26:45.0 | 2.0 s | Inspect → |
Provenance
- Revision
- 4ae8986f61485f5026fcff7269fff047149e3cba (main)
- Scanned
- 2026-10-10 01:08 UTC · NVIDIA RTX A6000
- Flag method
- rule-based-video-events-v1
- Episode method
- episode-motion-v1
- Code
- d54f0dee28b0d484 lock 282482d5a82a
Pinned dataset on Hugging Face ↗Result JSONMethodologyReport an issue with this result →