Tengda Han

Tengda Han is a research scientist at Google DeepMind, focusing on video understanding and visual-language models. He previously completed his PhD at the University of Oxford under Andrew Zisserman and earned a Bachelor of Engineering from Australian National University in mechanical & material engineering, with prior studies in business administration and law at Renmin University of China.

Research Interests: His work explores neural networks for video analysis, including self-supervised learning, video captioning, object counting, and prompt engineering. Key projects include AutoAD for movie description, Temporal Alignment Networks, and Dense Predictive Coding for video representation.

Scientific Awards:

• Best Paper Award at ACCV 2024
• Best Poster Award at BMVC 2023
• BMVA Sullivan Doctoral Thesis Prize Runner-up

Collaborations & Students: He has collaborated with researchers like Andrew Zisserman, Max Bain, and Arsha Nagrani, and mentored students Junyu Xie, Toby Perrett, and Niki Amini-Naieni on projects including Shot-by-Shot and Unique Video Captioning.