Modelling Spatio-Temporal Interactions for Compositional Action Recognition

no code implementations • 4 May 2023 • Ramanathan Rajendiran, Debaditya Roy, Basura Fernando

The final context-infused spatio-temporal interaction tokens are used for compositional action recognition.

Action Recognition Human-Object Interaction Detection +1

Interaction Region Visual Transformer for Egocentric Action Anticipation

no code implementations • 25 Nov 2022 • Debaditya Roy, Ramanathan Rajendiran, Basura Fernando

On the EK100 evaluation server, InAViT is the top-performing method on the public leaderboard (at the time of submission) where it outperforms the second-best model by 3. 3% on mean-top5 recall.

Action Anticipation Human-Object Interaction Detection +1