check out my talk at EUVIP 2026, titled
"Is Computer Vision Ready for Autonomous Systems?"

I think it didn’t suck as much as I thought it would, so I’m happy to share it 😄

youtube.com/live/Dtf5YaxsT…

a student feedback after today’s lecture:
“I had to use my brain at max capacity.”

sınıfta spam filter’ın kişiye özel olması gerektiğini anlatırken “mesela sizin sevdiğiniz grubun konser duyurusu, benim için spam olabilir” dedikten sonra, favori grubunu sorduğum üç genZ’nin de ardarda “Mor ve Ötesi” demesi 🥹

hani generational gap vardı?

first, I felt attacked, but then I realized eval is not online, and Track-On is not even in the SOTA table. shouldn't a real-world benchmark test online capabilities? the title is a bit misleading IMHO.

I’m all for getting excited about the next amazing paper or model, but I’m also a bit tired of “amazing” discoveries we forget about within a couple of weeks.

I feel the same about our own work. great to have it on our CVs, but will it survive the next conf. cycle?

in our paper reading group channel, I send papers before sleep at 1:30 at night, and I send more papers when I wake up at 8:30 in the morning. and this was just Fri night 🥹
yeah my work-life balance is not in the room with us but at least, students tagged can check them whenever they want 😅

with NeurIPS results out and ICLR deadline passed..

the ERC’s pitch for participating in the “Science is Wonderful” fair, though:

“that warm, fuzzy feeling when you do something meaningful in life” 🥰

I do everything in life for that feeling, ERC.

a bit confused about the NeurIPS site assignment. we chose Paris but got Atlanta because Paris was full. all authors are based in Europe and don’t plan to travel to the US. if we can just be assigned elsewhere, what was the point of choosing a site? and what was the point of @EurIPSConf?

women give such specific feedback, I’m digging it. after my talk:

“I can tell you’ve been teaching. the way you explain things and go back to basics when needed..”

“you never lost me. slide after slide, I was listening carefully despite the early hour.”

these mean so much 🥹

my updated talk title 😜

next week, I’ll be in Luxembourg giving a keynote (my first!!) @Euvip2026, alongside @dimadamen, @eliricci_, and @silvia_zuffi!

many thx to @AouadaDjamila and the organizers for the invitation, what a lineup to be part of 🙂‍↕️

what a crowd 🤗🤗

I guess we know which chess piece Luxembourg would be

ppl quote RT’ing their arXiv papers to celebrate NeurIPS acceptances.

meanwhile, our borderline paper got in by some miracle after we’d already rewritten it for ICLR…

we were too chicken to put it on arXiv, ok? and double blind still means something to some of us 🥹

Astra got R2’ed!
great work by Georgia 👌🏾

Holger going through SOTA solutions in the AD space one by one, listing theirs tricks and showing how they do not generalize has been my favorite thing in this summer school 🤗

we need more of this pessimistic but more accurate view in self-driving 💪

he seems to think my followers won’t know who he is, do you know who Holger is?

🍯 Got exciting 3D vision work that deserves another spotlight -- or a bold idea that isn’t quite a paper yet?

Then the #3DV2027 Nectar Track is for you: showcase strong recent work, share early ideas and demos; all while enjoying bougatsa, meze, and rebetiko in beautiful Thessaloniki 🇬🇷

📅 Deadline: Feb 15, 2027 (more details coming soon)
👉 3dvconf.github.io/2027/call-for-…