mlsanityData to 7 Oct 2026

M2DAR: Multi-View Multi-Scale Driver Action Recognition with Vision Transformer

arXiv 2305.08877 ↗First posted 13 May 2023cs.CVcited by 5

Not on the map. Why?

Abstract (arXiv)

Ensuring traffic safety and preventing accidents is a critical goal in daily driving, where the advancement of computer vision technologies can be leveraged to achieve this goal. In this paper, we present a multi-view, multi-scale framework for naturalistic driving action recognition and localization in untrimmed videos, namely M2DAR, with a particular focus on detecting distracted driving behaviors. Our system features a weight-sharing, multi-scale Transformer-based action recognition network that learns robust hierarchical representations. Furthermore, we propose a new election algorithm consisting of aggregation, filtering, merging, and selection processes to refine the preliminary results from the action recognition module across multiple views. Extensive experiments conducted on the 7th AI City Challenge Track 3 dataset demonstrate the effectiveness of our approach, where we achieved an overlap score of 0.5921 on the A2 test set. Our source code is available at urlhttps://github.com/PurdueDigitalTwin/M2DAR.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 19 of its 31 references that are in the map. First posted on arXiv 13 May 2023.

Builds on

19 of 31 references.

References · 19

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  4. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  5. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,617
  6. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 4,993
  7. 2017The Kinetics Human Action Video Datasetcited by 1,894
  8. 2021Multiscale Vision Transformerscited by 896
  9. 2021MViTv2: Improved Multiscale Vision Transformers for Classification and Detectioncited by 516
  10. 2018Rethinking the Faster R-CNN Architecture for Temporal Action Localizationcited by 194
  11. 2021Learning Salient Boundary Feature for Anchor-free Temporal Action Localizationcited by 156
  12. 2017CDC: Convolutional-De-Convolutional Networks for Precise Temporal Action Localization in Untrimmed Videoscited by 112
7 more
  1. 2017Temporal Context Network for Activity Localization in Videoscited by 37
  2. 2022Fine-grained Temporal Contrastive Learning for Weakly-supervised Temporal Action Localizationcited by 33
  3. 2022The 6th AI City Challengecited by 23
  4. 2023The 7th AI City Challengecited by 20
  5. 2022Synthetic Distracted Driving (SynDD2) dataset for analyzing distracted behaviors and various gaze zones of a drivercited by 15
  6. 2023Federated Transfer-Ordered-Personalized Learning for Driver Monitoring Applicationcited by 11
  7. 2022ViT-DD: Multi-Task Vision Transformer for Semi-Supervised Driver Distraction Detectioncited by 6

Built on by

Cited by year

Citing papers per year: 2023: 3; 2024: 0; 2025: 2.

5 papers.

  1. 2023A Survey on Multimodal Large Language Models for Autonomous Drivingcited by 288
  2. 2023Federated Learning for Connected and Automated Vehicles: A Survey of Existing Approaches and Challengescited by 39
  3. 2023Mitigating Transformer Overconfidence via Lipschitz Regularizationcited by 15
  4. 2025Dual-view Spatio-Temporal Feature Fusion with CNN-Transformer Hybrid Network for Chinese Isolated Sign Language Recognitioncited by 2
  5. 2025Spatial-Temporal Perception with Causal Inference for Naturalistic Driving Action Recognitioncited by 0