Scaling masked autoencoding video transformers from 20M to 22B parameters steadily improved camera pose, tracking, and depth estimation, while language-supervised and image-only models lagged on these tasks.
Revisiting feature prediction for learning visual rep- resentations from video
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Scaling 4D Representations
Scaling masked autoencoding video transformers from 20M to 22B parameters steadily improved camera pose, tracking, and depth estimation, while language-supervised and image-only models lagged on these tasks.