Coarse Temporal Attention Network (CTA-Net) for Driver's Activity Recognition

TL;DR

This paper introduces CTA-Net, a novel spatiotemporal attention framework that effectively captures subtle differences in driver activities from RGB videos, significantly improving recognition accuracy over existing methods.

Contribution

The paper proposes a new Coarse Temporal Attention Network with trainable branches and an attention mechanism to model subtle activity changes in driver videos, a novel approach for this task.

Findings

Outperforms state-of-the-art on four datasets

Uses only RGB videos for activity recognition

Effectively captures subtle temporal changes

Abstract

There is significant progress in recognizing traditional human activities from videos focusing on highly distinctive actions involving discriminative body movements, body-object and/or human-human interactions. Driver's activities are different since they are executed by the same subject with similar body parts movements, resulting in subtle changes. To address this, we propose a novel framework by exploiting the spatiotemporal attention to model the subtle changes. Our model is named Coarse Temporal Attention Network (CTA-Net), in which coarse temporal branches are introduced in a trainable glimpse network. The goal is to allow the glimpse to capture high-level temporal relationships, such as 'during', 'before' and 'after' by focusing on a specific part of a video. These branches also respect the topology of the temporal dynamics in the video, ensuring that different branches learn meaningful spatial and temporal changes. The model then uses an innovative attention mechanism to generate high-level action specific contextual information for activity recognition by exploring the hidden states of an LSTM. The attention mechanism helps in learning to decide the importance of each hidden state for the recognition task by weighing them when constructing the representation of the video. Our approach is evaluated on four publicly accessible datasets and significantly outperforms the state-of-the-art by a considerable margin with only RGB video as input.