A 128,892-clip, 2,826-hour real-world video dataset with camera trajectories, temporally grounded captions, and 982 revisit-rich panoramic sequences for interactive world-model training.
Title resolution pending
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
-
Sekai2: From World Exploration to Interactive World Modeling
A 128,892-clip, 2,826-hour real-world video dataset with camera trajectories, temporally grounded captions, and 982 revisit-rich panoramic sequences for interactive world-model training.