If you’ve never heard the story of how the #DUSt3R or *St3R family was born & what’s going on now - don’t miss @vincentleroy.bsky.social & his new talk at the #E2E3D workshop, also Oct 19th
Workshop E2E3D (afternoon): e2e3d.github.io 🧵5/9

CVPR 2025-Fast3R: Towards 3D Reconstruction of 1000+ Images in One Forward Pass Fast3R breaks the pairwise bottleneck in multi-view 3D reconstruction. Building on DUSt3R, it introduces a transformer-based architecture that directly regresses dense 3D pointmaps from unposed.

CVPR 2025-Fast3R: Towards 3D Reconstruction of 1000+ Images in One Forward Pass Fast3R breaks the...

https://opencv.org/blog/fast3r/

#Research #Papers #3D #Dust3R #Fast3R

Event Attributes

Just "dissect" the cross-attention mechanism of #DUSt3R, making 4D reconstruction easier.

💡Humans naturally separate ego-motion from object-motion without dynamic labels. We observe that #DUSt3R has implicitly learned a similar mechanism, reflected in its attention layers.

🦣Easi3R: 4D Reconstruction Without Training!

Limited 4D datasets? Take it easy.

#Easi3R adapts #DUSt3R for 4D reconstruction by disentangling and repurposing its attention maps → make 4D reconstruction easier than ever!

🔗Page: easi3r.github.io

💻Code: github.com/fanegg/Feat2GS
🎥Video: youtu.be/4fT5lzcAJqo?...

Big thanks to the amazing team!
@fanegg.bsky.social, @xingyu-chen.bsky.social, Anpei Chen, Gerard Pons-Moll, Yuliang Xiu

#DUSt3R #MASt3R #MiDaS #DINOv2 #DINO #SAM #CLIP #RADIO #MAE #StableDiffusion #Zero123

Exciting research from NAVER LABS Europe
@cvpr.bsky.social #CVPR2025 🚀! A breakdown of the 7 papers covering visual navigation, distillation, 3D reconstruction (remember #DUSt3R ☺️), localization & motion segmentation. europe.naverlabs.com/updates/cvpr/ 🧵 1/9 ⬇️