2020/04/04 by Junwei Liang, Liang, Junwei, Lu Jiang +4 · 15 citations
Computer Science · Engineering · Mathematics · #Adversarial system #Anomaly Detection Techniques and Applications #Artificial intelligence #Autonomous Vehicle Technology and Safety #Computer Vision and Pattern Recognition (cs.CV) #Computer science #Domain (mathematical analysis) #Drone #FOS: Computer and information sciences #Feature (linguistics) #Feature learning #Key (lock) #Machine learning #Mathematics #Representation (politics) #Test data #Training set #Trajectory #Variety (cybernetics) #Video Surveillance and Tracking Methods #cs.CV
paper · pdf · doi:10.48550/arxiv.2004.02022
published in arXiv (Cornell University) (Cornell University) · Accepted by ECCV 2020. Project website: https://next.cs.cmu.edu/simaug
openalex publication_date 2020/04/04 · arxiv created 2020/07/17 · arxiv updated 2020/07/21 · openalex created_date 2020/12/07 · openalex updated_date 2026/08/06
This paper studies the problem of predicting future trajectories of people in unseen cameras of novel scenarios and views. We approach this problem through the real-data-free setting in which the model is trained only on 3D simulation data and applied out-of-the-box to a wide variety of real cameras. We propose a novel approach to learn robust representation through augmenting the simulation training data such that the representation can better generalize to unseen real-world test data. The key idea is to mix the feature of the hardest camera view with the adversarial feature of the original view. We refer to our method as SimAug. We show that SimAug achieves promising results on three real-world benchmarks using zero real training data, and state-of-the-art performance in the Stanford Drone and the VIRAT/ActEV dataset when using in-domain training data.