Text-Enhanced Zero-Shot Action Recognition: A training-free approach

TL;DR

This paper introduces TEAR, a training-free method that leverages textual action descriptors to improve zero-shot video action recognition without extensive data or training, demonstrating effectiveness on multiple datasets.

Contribution

The paper presents a novel training-free approach for zero-shot video action recognition that uses textual descriptors to enhance recognition accuracy without dataset-specific training.

Findings

Effective on UCF101, HMDB51, and Kinetics-600 datasets.

Outperforms existing zero-shot methods in accuracy.

Requires no training data or extensive computation.

Abstract

Vision-language models (VLMs) have demonstrated remarkable performance across various visual tasks, leveraging joint learning of visual and textual representations. While these models excel in zero-shot image tasks, their application to zero-shot video action recognition (ZSVAR) remains challenging due to the dynamic and temporal nature of actions. Existing methods for ZS-VAR typically require extensive training on specific datasets, which can be resource-intensive and may introduce domain biases. In this work, we propose Text-Enhanced Action Recognition (TEAR), a simple approach to ZS-VAR that is training-free and does not require the availability of training data or extensive computational resources. Drawing inspiration from recent findings in vision and language literature, we utilize action descriptors for decomposition and contextual information to enhance zero-shot action recognition. Through experiments on UCF101, HMDB51, and Kinetics-600 datasets, we showcase the effectiveness and applicability of our proposed approach in addressing the challenges of ZS-VAR.