🚀How to preserve object consistency in NVS, ensuring correct position, orientation, plausible geometry, and appearance? This is especially critical for image/video generative models and world models.

🎉Check out our #CVPR2025 paper: MOVIS (jason-aplp.github.io/MOVIS) 👇 (1/6)

🚀 How to reconstruct 3D scenes with decomposed objects from sparse inputs?

Check out DPRecon (dp-recon.github.io) at #CVPR2025 — it recovers all objects, achieves photorealistic mesh rendering, and supports text-based geometry & appearance editing. More details👇 (1/n)

📢📢📢Excited to announce the 5th Workshop on 3D Scene Understanding for Vision, Graphics, and Robotics at #CVPR2025! Expect our awesome speakers and challenges on multi-modal 3D scene understanding and reasoning. 🎉🎉🎉

Learn more at scene-understanding.com.