Social GAN: Socially Acceptable Trajectories with Generative Adversarial Networks

Understanding human motion behavior is critical for autonomous moving platforms (like self-driving cars and social robots) if they are to navigate human-centric environments. This is challenging because human motion is inherently multimodal: given a history of human motion paths, there are many socially plausible ways that people could move in the future. We tackle this problem by combining tools from sequence prediction and generative adversarial networks: a recurrent sequence-to-sequence model observes motion histories and predicts future behavior, using a novel pooling mechanism to aggregate information across people. We predict socially plausible futures by training adversarially against a recurrent discriminator, and encourage diverse predictions with a novel variety loss. Through experiments on several datasets we demonstrate that our approach outperforms prior work in terms of accuracy, variety, collision avoidance, and computational complexity.

PDF Abstract CVPR 2018 PDF CVPR 2018 Abstract


Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Trajectory Prediction ETH Social-GAN Avg AMD/AMV 8/12 1.42 # 4
Trajectory Prediction ETH/UCY Social GAN ADE-8/12 0.58 # 20
Avg AMD/AMV 8/12 1.42 # 4

Results from Other Papers

Task Dataset Model Metric Name Metric Value Rank Source Paper Compare
Trajectory Prediction Stanford Drone Social GAN ADE-8/12 @K = 20 27.23 # 16
FDE-8/12 @K= 20 41.44 # 16
ADE (8/12) @K=5 27.25 # 2
FDE(8/12) @K=5 41.44 # 2


No methods listed for this paper. Add relevant methods here