Frozen video diffusion models, probed at optimal depth and noise levels, produce representations competitive with discriminative encoders across semantic and geometric video tasks in a single forward pass.
A simple recipe for contrastively pre-training video-first encoders beyond 16 frames.2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 14386–14397,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Gen4U: Unifying Video Generation and Understanding via Diffusion
Frozen video diffusion models, probed at optimal depth and noise levels, produce representations competitive with discriminative encoders across semantic and geometric video tasks in a single forward pass.