• Corpus ID: 238744427

THOMAS: Trajectory Heatmap Output with learned Multi-Agent Sampling

  title={THOMAS: Trajectory Heatmap Output with learned Multi-Agent Sampling},
  author={Thomas Gilles and Stefano Sabatini and Dzmitry V. Tsishkou and Bogdan Stanciulescu and Fabien Moutarde},
In this paper, we propose THOMAS, a joint multi-agent trajectory prediction framework allowing for efficient and consistent prediction of multi-agent multimodal trajectories. We present a unified model architecture for fast and simultaneous agent future heatmap estimation leveraging hierarchical and sparse image generation. We demonstrate that heatmap output enables a higher level of control on the predicted trajectories compared to vanilla multi-modal trajectory regression, allowing to… 

Figures and Tables from this paper

DenseTNT: End-to-end Trajectory Prediction from Dense Goal Sets
This work proposes an anchor-free and end-to-end trajectory prediction model, named DenseTNT, that directly outputs a set of trajectories from dense goal candidates and introduces an offline optimization-based technique to provide multi-future pseudo-labels for the final online model.


Large Scale Interactive Motion Forecasting for Autonomous Driving : The Waymo Open Motion Dataset
This work introduces the most diverse interactive motion dataset to their knowledge, and provides specific labels for interacting objects suitable for developing joint prediction models, and introduces a new set of metrics that provides a comprehensive evaluation of both single agent and joint agent interaction motion forecasting models.
Scene Transformer: A unified multi-task model for behavior prediction and planning
This work demonstrates that formulating the problem of behavior prediction in a unified architecture with a masking strategy may allow us to have a single model that can perform multiple motion prediction and planning related tasks effectively.
Implicit Latent Variable Model for Scene-Consistent Motion Forecasting
This paper proposes to characterize the joint distribution over future trajectories via an implicit latent variable model and model the scene as an interaction graph and employs powerful graph neural networks to learn a distributed latent representation of the scene.
TNT: Target-driveN Trajectory Prediction
The key insight is that for prediction within a moderate time horizon, the future modes can be effectively captured by a set of target states, which leads to the target-driven trajectory prediction (TNT) framework.
MultiPath: Multiple Probabilistic Anchor Trajectory Hypotheses for Behavior Prediction
This work presents MultiPath, which leverages a fixed set of future state-sequence anchors that correspond to modes of the trajectory distribution, which is efficient, requiring only one forward inference pass to obtain multi-modal future distributions, and the output is parametric, allowing compact communication and analytical probabilistic queries.
INTERACTION Dataset: An INTERnational, Adversarial and Cooperative moTION Dataset in Interactive Driving Scenarios with Semantic Maps
An INTERnational, Adversarial and Cooperative moTION dataset (INTERACTION dataset) in interactive driving scenarios with semantic maps for highly complex behavior such as negotiations, aggressive/irrational decisions and traffic rule violations is presented.
DESIRE: Distant Future Prediction in Dynamic Scenes with Interacting Agents
The proposed Deep Stochastic IOC RNN Encoder-decoder framework, DESIRE, for the task of future predictions of multiple interacting agents in dynamic scenes significantly improves the prediction accuracy compared to other baseline methods.
GOHOME: Graph-Oriented Heatmap Output for future Motion Estimation
GOHOME, a method leveraging graph representations of the High Definition Map and sparse projections to generate a heatmap output representing the future position probability distribution for a given agent in a traffic scene, yields an unconstrained 2D grid representation of agent future possible locations, allowing inherent multimodality and a measure of the uncertainty of the prediction.
HOME: Heatmap Output for future Motion Estimation
Home, a framework tackling the motion forecasting problem with an image output representing the probability distribution of the agent's future location, allows for a simple architecture with classic convolution networks coupled with attention mechanism for agent interactions, and outputs an unconstrained 2D top-view representation of theAgent's possible future.
$AIR^2$ for Interaction Prediction
This work developed a solution that takes an anchored marginal motion prediction model with rasterization and augments it to model agent interaction and predicts the joint confidences using a rasterized image that highlights the ego agent and the interacting agent.