VRUNet: Multi-Task Learning Model for Intent Prediction of Vulnerable Road Users
arXiv:2007.05397 · doi:10.2352/ISSN.2470-1173.2020.16.AVM-109
Abstract
Advanced perception and path planning are at the core for any self-driving vehicle. Autonomous vehicles need to understand the scene and intentions of other road users for safe motion planning. For urban use cases it is very important to perceive and predict the intentions of pedestrians, cyclists, scooters, etc., classified as vulnerable road users (VRU). Intent is a combination of pedestrian activities and long term trajectories defining their future motion. In this paper we propose a multi-task learning model to predict pedestrian actions, crossing intent and forecast their future path from video sequences. We have trained the model on naturalistic driving open-source JAAD dataset, which is rich in behavioral annotations and real world scenarios. Experimental results show state-of-the-art performance on JAAD dataset and how we can benefit from jointly learning and predicting actions and trajectories using 2D human pose features and scene context.
This paper is reprinted from, "VRUNet: Multi-Task Learning Model for Intent Prediction of Vulnerable Road Users, IS&T Electronic Imaging: Autonomous Vehicles and Machines 2020 Proceedings, (IS&T, Springfield, VA, 2020) page 109-1-10. DOI: 10.2352/ISSN.2470-1173.2020.16.AVM-109." Reprinted with permission of The Society for Imaging Science and Technology, holders of the 2020 copyright
References in corpus (9)
- The Cityscapes Dataset for Semantic Urban Scene Understanding
- Deep Cosine Metric Learning for Person Re-Identification
- Differential Recurrent Neural Networks for Action Recognition
- Scene-LSTM: A Model for Human Trajectory Prediction
- Joint Attention in Autonomous Driving (JAAD)
- Peeking into the Future: Predicting Future Person Activities and Locations in Videos
- SR-LSTM: State Refinement for LSTM towards Pedestrian Trajectory Prediction
- Structure Inference Machines: Recurrent Neural Networks for Analyzing Relations in Group Activity Recognition
- Intention Recognition of Pedestrians and Cyclists by 2D Pose Estimation