Freezing an image foundation model and training only a recurrent temporal module yields strong temporal performance on video tasks without large-scale video pre-training.
mIoU (Waymo).Mean Intersection over Union be- tween predicted and ground-truth bounding boxes, averaged over all objects and frames
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Towards Data-Efficient Video Pre-training with Frozen Image Foundation Models
Freezing an image foundation model and training only a recurrent temporal module yields strong temporal performance on video tasks without large-scale video pre-training.