Stargazer: A Transformer-based Driver Action Detection System for Intelligent Transportation
Authors:
- 1 June 2022
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 3159-3166
Abstract
Distracted driver actions can be dangerous and cause severe accidents. Thus, it is important to detect and eliminate distracted driving behaviors on the road to save lives. To this end, we study driver action detection using videos captured inside the vehicle. We propose Stargazer, an efficient, transformer-based system exploiting rich temporal features about the human behavioral information, with a simple yet effective action temporal localization framework. The core of our system contains an improved version of the multi-scale vision transformer network, which learns a hierarchy of robust representations. We then use a sliding-window classification strategy to facilitate temporal localization of actions-of-interest. The proposed system wins the second place in the Naturalistic Driving Action Recognition of AI City Challenge 2022 (Track 3) 1 . The code and models are released 2 .Keywords
Funding Information
- Tencent
This publication has 24 references indexed in Scilit:
- Minding the Gaps in a Video Action Analysis Pipeline2019
- A Closer Look at Spatiotemporal Convolutions for Action Recognition2018
- Single Shot Temporal Action Detection2017
- Asynchronous Temporal Fields for Action Recognition2017
- Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset2017
- SST: Single-Stream Temporal Action Proposals2017
- Deep Residual Learning for Image Recognition2016
- People tracking with human motion predictions from social forces2010
- Convolutional Learning of Spatio-temporal Features2010
- A Spatio-Temporal Descriptor Based on 3D-Gradients2008