Adam Kortylewski 🚨 Hiring PhD students

EurIPS is coming! 📣 Mark your calendar for Dec. 2-7, 2025 in Copenhagen 📅 EurIPS is a community-organized conference where you can present accepted NeurIPS 2025 papers, endorsed by @NeurIPSConf and #NordicAIR and is co-developed by @ELLISforEurope eurips.cc

🎥 The recordings for our @CVPR Workshop on Generative Models for Computer Vision are live! Missed a keynote or want to revisit your favorites? Find them here: youtube.com/playlist?list=… #CVPR2025

In our @ICCVConference work, we enhance DINO features in terms of cross-view and cross-instance consistency. The new model is strong at semantic correspondence and is a plug-and-play replacement for any pipeline with DINO backbone. Try it out with just a single line of code.

🚨 Come see our poster at @CVPR ! 🖼️ Poster #104 🕓 4:00 PM – 6:00 PM We're presenting Common3D — a fully self-supervised method to learn 3D morphable models of everyday objects from videos, enabling zero-shot semantic correspondence, 3D shape and pose estimation.

Full House with Kaiming He @CVPR 🥳

Getting crowded, make sure to check out our next speaker Jiaotao Gu, talking about spatial world models 🔥

We are starting the next session with the legendary Bjorn Ommer @CVPR Grand A2

Starting soon

We are about to start our Workshop on Generative Models for Computer Vision in Grand A2 at 8:45am. Make sure to check it out! 🔥

Submit your extended abstract to our workshop on "Generative Models for Computer Vision" #CVPR2025 @CVPR Authors with accepted CVPR papers are welcome to present their poster as well! Deadline: April 25th We also have an incredible speaker line-up!

We had an incredible week at @dagstuhl for the seminar on Generative Models for 3D Vision! Great talks and lots of inspiration on the future of 3D vision and generative models 🚀

🚨Paper Alert 🚨 ➡️Paper Title: Evolutive Rendering Models 🌟Few pointers from the paper 🎯The landscape of computer graphics has undergone significant transformations with the recent advances of differentiable rendering models. 🎯These rendering models often rely on…

🚀Excited to share our ImageNet3D dataset for general-purpose object-level 3D understanding, which augments 200 categories from ImageNet21k with 3D annotations. 🔗Project page: imagenet3d.github.io 🥰Big thanks to my advisors: @AdamKortylewski @yaoyaoliu1 @YuilleAlan

Proud advisor moment - Yu Chi @YuChi__26 yesterday presented his internship project (done even before starting his MSc thesis!). Make sure to reach out when you see him at the conference 🥳 @eccvconf