Back from #ICVSS2025 in Sicily where I presented “Learn Your Scales: Scale-Consistent Generative Novel View Synthesis.” 🚀
Huge thanks to the ICVSS organizers, speakers, and all the brilliant attendees who made the week unforgettable. 🤩
Our PhD students, Alina D. and @luciehuang.bsky.social, are participating in #ICVSS2025. During the poster session, Alina presented her research on parametric shape models of 3D vessels learned from segmentations using differentiable voxelization (arxiv.org/pdf/2507.02576).
Enjoyed a fantastic deep dive into generative models with Vittorio Ferrari on day 3 of #ICVSS2025! A great lecture covering the evolution from fundamental diffusion models to the latest in controllable image and video generation like ControlNet.
#ICVSS2025 getting intense!
Yesterday, two great lectures by Leonidas Guibas on foundation models for 3D/4D modeling and Daniel Cremers on 3D computer vision in the age of deep learning.
Then, nice aperitif and dinner 😄
Now an excing day 3 is ahead!
Another poster presented yesterday at ICVSS is our collaboration with Keio University in Japan. Masashi Hatano spent 1-year visiting @bristoluni.bsky.social and was presenting our ArXiv paper: The Invisible EgoHand at #ICVSS2025. Great work worth your attention
arxiv.org/abs/2504.08654
3/3
Also proud of the HD-EPIC team (Omar, Fahd, Kaiting) who are attending @ICVSS on behalf of @bristoluni.bsky.social and @unileiden.bsky.social. They had a busy poster session #ICVSS2025 yesterday detailing our @cvprconference.bsky.social work and public dataset
hd-epic.github.io
2/3
Giving a tutorial at #ICVSS2025 is special - you could inspire the brightest upcoming researchers in the field, eager to learn and discuss.
I talked about a new passion of mine - Video Understanding *out of the frame*.
If interested slides are at: tinyurl.com/Dima-ICVSS2025
🧵1/3
Kicking off the day at #ICVSS2025 with a thought-provoking talk by Andrea Vedaldi on Spatial Intelligence as new frontier of Computer Vision.
Great to see ML-compatible building 3D blocks such as VGGT appearing - then on to dynamic reconstruction and 3D priors in latent space!
How to make 3D fundamental to all aspects of computer vision?
A Vedaldi @oxford-vgg.bsky.social with a brilliant tutorial this morning #icvss2025
Great talk yesterday by @dimadamen.bsky.social on the past, present, and future of egocentric video!
Video understanding out of the frame was especially exciting, pushing us to consider the 3D world beyond the camera's view.
Now on to day 2! #ICVSS2025
Happening now: @fereshtehforghani.bsky.social presenting at #ICVSS2025 ❤️
Thanks @dimadamen.bsky.social for sharing!
Great reminder from Christoph Feichtenhofer at #ICVSS2025 that the "secret sauce" of models like CLIP is the strong supervision from training data, not just novel architectures.
Insightful to see a dissection of how CLIP pretraining works and what it can enable beyond classification.
Great talk by @ajdavison.bsky.social on the journey from SLAM to Spatial AI at #ICVSS2025!
A compelling argument for the power of dynamic, updatable 3D representations and their direct impact on the future of robotics and wearable devices.
And nice insights for spatial AI researchers!
🚀 And the tour goes on! I'll be at #ICVSS2025 all week!
Leaning back in full learning mode and looking forward to meeting old and new friends!
If you're here, feel free to ping me!
