1. [Publications](/publications)
2. GEN3C: 3D-Informed World-Consistent Video Generation with Precise Camera Control
 
 # GEN3C: 3D-Informed World-Consistent Video Generation with Precise Camera Control

  ![](/sites/default/files/styles/wide/public/publications/gen3c-teaser.png?itok=fqn-moUn)

 We present **GEN3C**, a generative video model with precise **C**amera **C**ontrol and temporal 3D **C**onsistency. We achieve this with a 3D cache: point clouds obtained by predicting the pixel-wise depth of seed images or previously generated frames. When generating the next frames, GEN3C is conditioned on the 2D renderings of the 3D cache with the new camera trajectory provided by the user. Our results demonstrate more precise camera control than prior work, as well as state-of-the-art results in sparse-view novel view synthesis, even in challenging settings such as driving scenes and monocular dynamic video.



 ## Authors



Xuanchi Ren (NVIDIA)

Tianchang Shen (NVIDIA)

Jiahui Huang (NVIDIA)

Huan Ling (NVIDIA)

Yifan Lu (NVIDIA)

[Merlin Nimier-David](/person/merlin-nimier-david)

[Thomas Müller](/person/thomas-muller)

[Alex Keller](/person/alex-keller)

[Sanja Fidler](/person/sanja-fidler)

Jun Gao (NVIDIA)

 

 

 ## Publication Date



Wednesday, August 13, 2025

 

 ## Published in



[CVPR 2025](https://ieeexplore.ieee.org/document/11092782)

 

 ## Research Area



[Artificial Intelligence and Machine Learning ](/research-area/machine-learning-artificial-intelligence)

[Autonomous Vehicles](/research-area/autonomous-vehicles)

[Computational Photography and Imaging](/research-area/computational-photography-imaging)

[Computer Vision](/research-area/computer-vision)

 

 

 ## External Links



[Project page](https://research.nvidia.com/labs/toronto-ai/GEN3C/)

 

 

 ## Uploaded Files



[Paper](https://d1qx31qr3h6wln.cloudfront.net/publications/ren25gen3c.pdf "Open file in new window")26.72 MB

[Supplementary material](https://d1qx31qr3h6wln.cloudfront.net/publications/ren25gen3c-supp.pdf "Open file in new window")3.74 MB