Knowledge-guided Disentanglement with Atomic Actions for Action Recognition⭐9

Proposes KDA, using LLMs to decompose actions into atomic actions with spatial-temporal semantics for enhanced, disentangled action recognition, achieving state...