| CARVIEW |
|
Tengda Han (韩腾达)
/tən-daː/
tengda@google.com / Google Scholar / GitHub / Twitter I am a research scientist at Google DeepMind. I was a post-doctoral research fellow in VGG at the University of Oxford. I obtained my PhD degree in the same group, supervised by Andrew Zisserman. I train neural networks to understand videos. |
|
I obtained a Bachelor of Engineering degree in
Australian National University,
with a major in mechanical & material engineering.
I had chance to work on deep learning projects with
Basura Fernando,
Anoop Cherian,
Mehrtash Harandi,
Stephen Gould and
Richard Hartley.
Before that, I studied business administration and law in
Renmin University of China,
Beijing, for one year.
Originally, I am from the suburbs of Hangzhou, China.
- [Jul. 2025] 1 paper on "AD for animations" accepted by ACMMM! Congrats Zhongrui!
- [Jun. 2025] 1 paper "Shot-by-Shot" accepted by ICCV! Congrats Junyu!
- [Apr. 2025] Co-organizing the "SLoMo" workshop on Movie Understanding at ICCV25.
- [Feb. 2025] 1 paper "Orthogonal Optimizer" accepted by CVPR!
- [Dec. 2024] Co-organizing the Self-Supervised Learning workshop at NeurIPS!
- [Dec. 2024] Our paper on Unique Video Captioning gets the Best Paper Award at ACCV! Big congrats to Toby!
- [Sep. 2024] 2 papers "AutoAD-Zero" and "Unique Video Captioning" accepted by ACCV, 1 paper "CountGD" accepted by NeurIPS! Great work from Junyu, Toby and Niki!
- [Apr. 2024] Published Stale Diffusion: Hyper-realistic 5D Movie Generation Using Old-school Methods on April 1st. Enjoy reading!
- [Mar. 2024] Join Google DeepMind in London, as a research scientist.
- [Mar. 2024] 2 papers "AutoAD-III" and "Unsupervised Counting" accepted by CVPR2024.
- [Feb. 2024] Give a talk at École des Ponts ParisTech on Movie Audio Description.
- [Jan. 2024] Co-organizing the Learning from Procedural Videos and Language. See you in Seattle!
- [Jan. 2024] Co-organizing the BMVA Symposium on Vision and Language with Michael Wray and Davide Moltisanti. See you in London!
- [Dec. 2023] Give talks at Shanghai Jiao Tong University and The AI Talks on Long Video Understanding.
- [Nov. 2023] Our paper "Open-world Text-specified Object Counting" received the Best Poster Award at BMVC2023. Congrats to Niki!
- [Nov. 2023] Received the BMVA Sullivan Doctoral Thesis Prize Runner-up.
- [Aug. 2023] 1 paper "Open-world Text-specified Object Counting" accepted by BMVC2023.
- [Jul. 2023] 1 paper "AutoAD II" accepted by ICCV2023.
- [Jul. 2023] Our workshop "Self-Supervised Learning - Theory and Practice" jointly organised with Ishan Misra, Mathilde Caron, Hilde Kuehne and Pengtao Xie is accepted by NeurIPS'23! We are accepting paper submissions!
- [Jun. 2023] Give a talk on The First Edinburgh Workshop on Affordable Machine Learning. [Slides pdf]
- [Apr. 2023] Our workshop "BigMAC: Big Model Adaptation for Computer vision" jointly organised with Yuki Asano, Mathilde Caron, Phillip Isola and Serge Belongie, is accepted by ICCV'23. See you in Paris!
- [Mar. 2023] 1 paper "AutoAD: Movie Description in Context" accepted by CVPR2023 with Highlight.
- [Nov. 2022] Passed my PhD with "no corrections" (thesis pdf). My examiners are Carl Vondrick and Andrea Vedaldi.
- [Sept. 2022] 1 paper "Turbo Training with Token Dropout" accepted by BMVC2022.
- [Jul. 2022] 1 paper "Prompting Visual-Language Models for Efficient Video Understanding" accepted by ECCV2022.
- [Mar. 2022] 1 paper "Temporal Alignment Networks" accepted by CVPR2022 as an oral.
- [Aug. 2021 - Dec. 2021] Research Scientist Intern at DeepMind hosted by Jean-Baptiste Alayrac. I am in the Flamingo project.
- [Sep. 2020] 1 paper "Self-supervised Co-Training (CoCLR)" accepted by NeurIPS2020.
More
- A template to produce your PhD thesis from arXiv zip files, liked by a few Oxford PhDs.
- A web server to monitor SLURM resource, internally used in VGG.