Yup, ECCV'22 will not have a pre-registration deadline; only the paper and then supp deadlines. Hoping this simplifies the submission process.

Now that the ECCV'22 deadline is behind us, good time for a SHOUTOUT to the anonymous CVPR'22 AC who accepted our paper, despite (partially) negative reviews. Grateful for the experienced AC weighing in with their perspectives on a tough call, especially when choosing Accept :)

ALERT: Clarifications coming to #ECCV2022 Supplementary Materials tomorrow (Saturday). Info will be shared here and on the webpage. Stay tuned.

Please take note of this announcement. Some issues require immediate attention from ECCV'22 users and, if not addressed, can unfortunately cause papers to get desk rejected or people being relieved of their reviewer / AC roles (e.g., missing domain conflicts).

Junior just told me he's started using Overleaf for his assignments. Strange mixture of feeling very proud and very old, all at once.

Noisy labels are inevitable, but 40% mislabeled samples? 60%??? When would that ever happen? When learning from few samples For solutions, see our Noisy Few-Shot CVPR'22 paper: shorturl.at/gkGO2 (tl;dr: Transformer self-attention for weighing samples based on label noise)

Worried about others using your media for #deepfakes? Be proactive protecting it! Details+code in our #CVPR2022 paper: shorturl.at/oxyWZ Super grateful to my amazing collaborators: Vishal Asnani, Xi Yin, Sijia Liu, and Xiaoming Liu #deepfakedetection

Honored and grateful to the organisers for inviting me to speak at this European Conference on Computer Vision #ECCV2026 workshop.

Thank you for all the reviewers, ACs, and especially emergency reviewers for the amazing work, right up to the very last minute. European Conference on Computer Vision #ECCV2026

ECCV accepted IDs are finally out! A HUGE THANK YOU to all the authors, reviewers, and area chairs. CONGRATULATIONS to those in the list and sincere commiserations to authors of submissions that did not make it. More info on CMT in the coming days.

Had a lot of fun working on this: Very cool exercise trying to imagine what foreigners would need to know when visiting your hometown. Hoping it would help European Conference on Computer Vision #ECCV2026 participants make more of their visit. See you all in a little over a week!

Crazy to think 20y ago, my first paper at ECCV02 and I couldn't even attend the galla as a student. Now, at European Conference on Computer Vision #ECCV2026, a PC. Wonder if 2044 PCs are in this picture... A hard year but sooo worth it, now seeing you all.

Chuffed to have AK mentioned a paper I was fortunate to be involved with. Huge thank you to my wonderful coauthors!

One last #eccv '22 post, sharing a short video summary of the conference, for all those who'd like to reminisce and those who're wondering what they missed.

Looking for a postdoc position AND industry experience? Passionate about publishing cutting edge research in top venues? Is AI fairness and ethics important to you? Why not apply for a postdoc position with us, at Meta RAI? DM me for more info metacareers.com/jobs/510723541…

Signal boosting this amazing opportunity to work with the wonderful Dima Damen. Envious of the person that gets accepted!

**Only 2 days left to the deadline!** Submit your full papers to the Unlearning and Model Editing (U&Me) workshop at ECCV'24, Milan, Italy Submission Deadline: July 10, 2024 Accepted papers will appear as ECCV'24 workshop proceedings. sites.google.com/view/u-and-me-…

"The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery" (sakana.ai/ai-scientist/). Curious how conflict-free peer-review looks like when the same AI is used by authors to write a paper, and then by reviewers to review it...