Published several seminal works, including but not limited to: Unifying Flow, Stereo and Depth Estimation (TPAMI 2023); GMFlow: Learning Optical Flow via Global Matching (CVPR 2022); JRDB: A Dataset and Benchmark of Egocentric Robot Visual Perception of Humans in Built Environments (TPAMI 2021); Generalized Intersection over Union: A Metric and A Loss for Bounding Box Regression (CVPR 2019); SoPhie: An Attentive GAN for Predicting Paths Compliant to Social and Physical Constraints (CVPR 2019); Social-BiGAT: Multimodal Trajectory Forecasting using Bicycle-GAN and Graph Attention Networks (NeurIPS 2019); Online multi-target tracking using recurrent neural networks (AAAI 2017); Joint probabilistic data association revisited (ICCV 2015).
Research Experience
Research areas cover fundamental vision problems (such as object detection and segmentation), multi-object tracking, 3D reconstruction of humans and objects in dynamic scenes, a multi-task 3D visual perception system for mobile robots in human environments, the JackRabbot dataset and benchmark, human trajectory/body motion forecasting, spatiotemporal action, social group, and activity detection, active visual navigation in unexplored environments, single or multi-UAV planning for discovering and tracking moving objects.
Background
Research interests include the theoretical foundations and practical applications of computer vision and machine learning, particularly for embodied AI's perception, prediction, and interaction. Focuses on object and scene understanding and reconstruction, as well as predicting and reasoning about human motion, activity, and behavior in the presence of physical and social interactions.
Miscellany
Recent news updates until April 2025, involving multiple research project outcomes published across various international conferences or journals.