Leo Segre

[Or Hirschorn @Or_Hirsch]

Excited to share our new paper: SpheRoPE! 🌐
We introduce a zero-shot training-free framework that lets DiTs generate flawless 360° panorama images and videos. Prompt a world and instantly step inside it.
📄Paper: https://t.co/HrkEzm1vkY
🌎Project Page: https://t.co/5hfpEcWYCK

[Or Hirschorn @Or_Hirsch]

Accepted to #ECCV2026 🎉

We introduce a method to upgrade any foundation model to be multi-view consistent. Check out our paper 👇
@eccvconf

[Kwang Moo Yi @kwangmoo_yi]

Segre and Hirschorn et al., "Multi-View Foundation Models"

Train an adapter to 2D foundational models like DINO/SAM/CLIP that allows turning them into "multi-view" versions. More reliable results when doing multi-view tasks.

[Or Hirschorn @Or_Hirsch]

Excited to share our new paper:
Multi-View Foundation Models! 🌐

We introduce a general framework to upgrade 2D foundation models (DINO, SAM, CLIP) into multi-view consistent foundational models.

📄 Paper: https://t.co/c1gWvs27fG
⚡️ Project Page + Code: https://t.co/fgYzMD3L61

[Leo Segre @LeoSegre]

👻 Hate "floaters" ruining your NeRF renders?

I'm presenting "Optimize the Unseen" tomorrow at #NeurIPS2025! It’s a fast, post-hoc NeRF cleanup method.

Come say hi at the poster session👋

📍 Poster #4406 (Hall C-E) 🗓️ Tomorrow (Fri) 11am-2pm

🔗Project: https://t.co/WOqTRfZkPf