Learning a Pose Lexicon for Semantic Action Recognition

TL;DR

This paper introduces a method to learn a semantic pose lexicon linking textual instructions with visual poses, enabling improved action recognition, including zero-shot scenarios, through probabilistic mapping and sequence matching.

Contribution

It proposes a novel approach to learn a pose lexicon that bridges semantic and visual representations for action recognition.

Findings

Effective in zero-shot action recognition

Validated on MSRC-12 and WorkoutSu-10 datasets

Improves semantic understanding of actions

Abstract

This paper presents a novel method for learning a pose lexicon comprising semantic poses defined by textual instructions and their associated visual poses defined by visual features. The proposed method simultaneously takes two input streams, semantic poses and visual pose candidates, and statistically learns a mapping between them to construct the lexicon. With the learned lexicon, action recognition can be cast as the problem of finding the maximum translation probability of a sequence of semantic poses given a stream of visual pose candidates. Experiments evaluating pre-trained and zero-shot action recognition conducted on MSRC-12 gesture and WorkoutSu-10 exercise datasets were used to verify the efficacy of the proposed method.