IMoRe couples a MAC-style memory network with program-function embeddings and multi-level ViT motion features to reach state-of-the-art accuracy on Babel-QA and a new HuMMan-QA benchmark.
Dynamic motion representation for human action recogni- tion
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
unclear 1representative citing papers
citing papers explorer
-
IMoRe: Implicit Program-Guided Reasoning for Human Motion Q&A
IMoRe couples a MAC-style memory network with program-function embeddings and multi-level ViT motion features to reach state-of-the-art accuracy on Babel-QA and a new HuMMan-QA benchmark.