# VRUNet: Multi-Task Learning Model for Intent Prediction of Vulnerable Road Users

arXiv: https://arxiv.org/abs/2007.05397 · HTML: https://mlsanity.com/paper/2007.05397/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 10 Jul 2020.

## Built on by (most cited first, 7 of 7)

- [TrouSPI-Net: Spatio-temporal attention on parallel atrous convolutions and U-GRUs for skeletal pedestrian crossing prediction](https://mlsanity.com/paper/2109.00953/) (2021)
- [Assessing Cross-dataset Generalization of Pedestrian Crossing Predictors](https://mlsanity.com/paper/2201.12626/) (2022)
- [TrajFusionNet: Pedestrian Crossing Intention Prediction via Fusion of Sequential and Visual Trajectory Representations](https://mlsanity.com/paper/2508.19866/) (2025)
- [Single Shot Multitask Pedestrian Detection and Behavior Prediction](https://mlsanity.com/paper/2101.02232/) (2021)
- [Semantic and Articulated Pedestrian Sensing Onboard a Moving Vehicle](https://mlsanity.com/paper/2309.06313/) (2023)
- [From Camera-Based Sensing to Reasoning: A Comprehensive Review Toward Proactive Vulnerable Road User Safety](https://mlsanity.com/paper/2510.03314/) (2025)
- [Valeo Near-Field: a novel dataset for pedestrian intent detection](https://mlsanity.com/paper/2510.15673/) (2025)

## Abstract (arXiv)

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.