Dr. Xing Di is a researcher in biometrics, video understanding, and agentic AI. He received his Ph.D. degree at Johns Hopkins University in 2021, advised by Prof. Vishal Patel. He previously worked as a Research Scientist at Protagolabs Inc. (2021–2026) with Yu Cheng, and worked/interned at PAII Inc. and Hikvision N.A. with Dr. Le Lu and Dr. Bo Wang respectively. He received his M.E. degree at Stevens Institute of Technology in 2015 and his B.E. degree at The University of Electronic Science and Technology of China (Chengdu, China) in 2013.

News

• 2026.10: Obtained the IBM RAG and Agentic AI Professional Certificate (IBM Skills Network, via Coursera) — View certificate
• 2024.09: Our paper Fine-Grained Rebalancing of Datasets for Correct Demographic Classification was presented in International Conference on Content-Based Multimedia Indexing (CBMI)
• 2023.11: Our paper Transform-Equivariant Consistency Learning for Temporal Sentence Grounding was accepted in ACM Transactions on Multimedia Computing, Communications, and Applications
• 2023.10: Our paper ProS: Facial Omni-Representation Learning via Prototype-based Self-Distillation was accepted in WACV 2024
• 2022.11: Our paper Hypotheses tree building for one-shot temporal sentence localization was accepted in AAAI 2023
• 2022.10: Our paper Rethinking the video sampling and reasoning strategies for temporal sentence grounding was accepted in Findings of EMNLP 2022
• 2022.07: Our paper Backdoor attacks on crowd counting accepted in ACM’MM 2022