TNT: Target-driveN Trajectory Prediction

Predicting the future behavior of moving agents is essential for real world applications. It is challenging as the intent of the agent and the corresponding behavior is unknown and intrinsically multimodal. Our key insight is that for prediction within a moderate time horizon, the future modes can be effectively captured by a set of target states. This leads to our target-driven trajectory prediction (TNT) framework. TNT has three stages which are trained end-to-end. It first predicts an agent's potential target states $T$ steps into the future, by encoding its interactions with the environment and the other agents. TNT then generates trajectory state sequences conditioned on targets. A final stage estimates trajectory likelihoods and a final compact set of trajectory predictions is selected. This is in contrast to previous work which models agent intents as latent variables, and relies on test-time sampling to generate diverse trajectories. We benchmark TNT on trajectory prediction of vehicles and pedestrians, where we outperform state-of-the-art on Argoverse Forecasting, INTERACTION, Stanford Drone and an in-house Pedestrian-at-Intersection dataset.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Motion Forecasting Argoverse CVPR 2020 TNT - CoRL20 MR (K=6) 0.1656 # 173
minADE (K=1) 2.174 # 111
minFDE (K=1) 4.9593 # 105
MR (K=1) 0.7097 # 107
minADE (K=6) 0.9097 # 164
minFDE (K=6) 1.4457 # 159
DAC (K=6) 0.9889 # 33
brier-minFDE (K=6) 2.1401 # 147
Trajectory Prediction INTERACTION Dataset - Validation TNT minADE6 0.21 # 2
minFDE6 0.67 # 3
Trajectory Prediction PAID TNT minFDE3 0.32 # 3
minADE3 0.18 # 3
Trajectory Prediction Stanford Drone TNT ADE (8/12) @K=5 12.23 # 2
FDE(8/12) @K=5 21.16 # 6


No methods listed for this paper. Add relevant methods here