Tengda Han

Organized by: Junyu Xie, Ridouane Ghermi, @tengda.bsky.social, Max Bain, Arsha Nagrani, @vickykalogeiton.bsky.social, @gulvarol.bsky.social, Weidi Xie, Ivan Laptev and Andrew Zisserman.

See you in Hawaii! 🌺

SF20KCompetition - a Hugging Face Space by SLoMO-Workshop This application allows users to view competition information, dataset details, leaderboards, and submission statuses. Users can fetch and manage their submissions, view rules, and check login stat...

As a part of the workshop, we have a MovieQA competition based on the SF20K dataset and hosted on HuggingFace @hf.co

Main Track: huggingface.co/spaces/SLoMO...

Plus, we have a special track for small models (< 8B)! huggingface.co/spaces/SLoMO...

We’re excited to have a fantastic lineup of speakers:
@amypavel.bsky.social, Anna Rohrbach, Mike Zheng Shou, Makarand Tapaswi. We’ll also host a panel discussion with the organizers!

Movies are more than just video clips, they are stories! 🎬

We’re hosting the 1st SLoMO Workshop at #ICCV2025 to discuss Story-Level Movie Understanding & Audio Descriptions!

Website: slomo-workshop.github.io
Competition: huggingface.co/spaces/SLoMO...

Thank @dimadamen.bsky.social for presenting our Orthogonal Optimizer! It’s a simple modification on standard optimizers for streaming video learning. We have code available at sites.google.com/view/orthogo...

Learning from Streaming Video with Orthogonal Gradients We address the challenge of representation learning from a continuous stream of video as input, in a self-supervised manner. This differs from the standard approaches to video learning where videos ar...

Check out our CVPR 2025 paper: arxiv.org/abs/2504.01961. Work with Dilara Gokay, Joseph Heyward, Chuhan Zhang, Daniel Zoran, Viorica Pătrăucean, João Carreira, Dima Damen and Andrew Zisserman, from Google DeepMind

Humans learn from one continuous visual stream, but large video models have to be trained on billions of web videos.
We found that learning from such sequential streams is challenging for video models—and we introduce a family of "orthogonal optimizers" to bridge the gap!

It's interesting to see that visual counting remains to be quite challenging for generalist AI models. But this specialist model counts very well. Nice work from @nikigoliai.bsky.social last year!

We are looking for a student researcher to work on video understanding plus 3D, in Google DeepMind London. DM/Email me or pass it to someone if you feel it may be a good fit!

How do you know he is not 🤔😆

From an award candidate... to best paper #ACCV2024
Glad to share that "It's Just Another Day" received the top award at the conference.
@bristoluni.bsky.social @ox.ac.uk

This paper is worth reading :-) based on the reviewers, AC and awards committee. We thank them for their time and effort.