The Garden of Forking Paths: Towards Multi-Future Trajectory Prediction
arXiv:1912.06445
Abstract
This paper studies the problem of predicting the distribution over multiple possible future paths of people as they move through various visual scenes. We make two main contributions. The first contribution is a new dataset, created in a realistic 3D simulator, which is based on real world trajectory data, and then extrapolated by human annotators to achieve different latent goals. This provides the first benchmark for quantitative evaluation of the models to predict multi-future trajectories. The second contribution is a new model to generate multiple plausible future trajectories, which contains novel designs of using multi-scale location encodings and convolutional RNNs over graphs. We refer to our model as Multiverse. We show that our model achieves the best results on our dataset, as well as on the real-world VIRAT/ActEV dataset (which just contains one possible future).
arXiv commentCVPR 2020. Code, models and dataset are available at: https://next.cs.cmu.edu/multiverse/index.html
References in corpus (12)
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
- Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting
- ADADELTA: An Adaptive Learning Rate Method
- Sequence Level Training with Recurrent Neural Networks
- Emergence of Locomotion Behaviours in Rich Environments
- CARLA: An Open Urban Driving Simulator
- A Simple, Fast Diverse Decoding Algorithm for Neural Generation
- Overcoming Limitations of Mixture Density Networks: A Sampling and Fitting Framework for Multimodal Future Prediction
- Argoverse: 3D Tracking and Forecasting with Rich Maps
- MultiPath: Multiple Probabilistic Anchor Trajectory Hypotheses for Behavior Prediction
- Scene-LSTM: A Model for Human Trajectory Prediction
- Focal Visual-Text Attention for Visual Question Answering