Language Model Beats Diffusion - Tokenizer is Key to Visual Generation January 09, 2024Invited Talks / Job Talks; NYU, CalTech, HKUST, ICT CAS, ByteDance, Baidu, Kunlun Tech, AIsphere, PKU AANC / OpenAI, xAI, Nvidia; Online and Beijing
Towards Multi-Modal Foundation Models: A Multi-Task Generative Perspective September 20, 2023Thesis Proposal / Invited Talk; LTI, SCS, CMU / ByteDance; Pittsburgh, Pennsylvania
ArgusRoad: Road Activity Detection with Connectionist Spatiotemporal Proposals October 16, 2021Invited Talk; ROAD, ICCV 2021; Virtual
Argus++: Real-time Activity Detection in Unknown Facilities with Dense Spatio-temporal Proposals June 18, 2021Invited Talk / Paper; ActivityNet, CVPR 2021 / HADCV, WACV 2022; Virtual / Waikoloa, Hawaii
Real-time Activity Detection in Unknown Facilities with Dense Spatio-temporal Proposals January 05, 2021Invited Talk; HADCV, WACV 2021; Virtual
CMU Informedia at TRECVID 2020: Towards Real-time Activity Detection with Dense Spatio-temporal Proposals December 09, 2020Report; TRECVID 2020; Virtual
Zero-VIRUS: Zero-shot VehIcle Route Understanding System for Intelligent Transportation June 04, 2020Paper; AI City, CVPR 2020; Virtual
Argus: Efficient Activity Detection System for Extended Video Analysis March 05, 2020Paper; HADCV, WACV 2020; Aspen, Colorado
Training-free Monocular 3D Event Detection System for Traffic Surveillance December 09, 2019Paper; Big Data 2019; Los Angeles, California
Traffic Danger Recognition With Surveillance Cameras Without Training Data November 27, 2018Paper / Demo / Invited Talk; AVSS 2018 / ICCV 2019 / TRECVID 2019; Auckland, New Zealand / Seoul, South Korea / Gaithersburg, Maryland
MOBA-Slice: A Time Slice Based Evaluation Framework of Relative Advantage Between Teams in MOBA Games July 13, 2018Paper; WCG, IJCAI 2018; Stockholm, Sweden