A hierarchical event memory with segment-tree proposals and a future prediction branch achieves state-of-the-art online video temporal grounding on TACoS, ActivityNet Captions, and MAD.
Miniroad: Minimal rnn framework for online action detection
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Hierarchical Event Memory for Accurate and Low-latency Online Video Temporal Grounding
A hierarchical event memory with segment-tree proposals and a future prediction branch achieves state-of-the-art online video temporal grounding on TACoS, ActivityNet Captions, and MAD.