1. [Publications](/publications)
2. Neuralangelo: High-Fidelity Neural Surface Reconstruction
 
 # Neuralangelo: High-Fidelity Neural Surface Reconstruction

  ![](/sites/default/files/styles/wide/public/publications/best-inventions-2023-nvidia-neuralangelo.jpg?itok=bnv83STr)

 Neural surface reconstruction has been shown to be powerful for recovering dense 3D surfaces via image-based neural rendering. However, current methods struggle to recover detailed structures of real-world scenes. To address the issue, we present Neuralangelo, which combines the representation power of multi-resolution 3D hash grids with neural surface rendering. Two key ingredients enable our approach: (1) numerical gradients for computing higher-order derivatives as a smoothing operation and (2) coarse-to-fine optimization on the hash grids controlling different levels of details. Even without auxiliary inputs such as depth, Neuralangelo can effectively recover dense 3D surface structures from multi-view images with fidelity significantly surpassing previous methods, enabling detailed large-scale scene reconstruction from RGB video captures.



 ## Authors



[Max Zhaoshuo Li](/person/max-zhaoshuo-li)

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

Alex Evans (NVIDIA)

Russell H. Taylor (Johns Hopkins University)

Mathias Unberath (Johns Hopkins University)

[Ming-Yu Liu](/person/ming-yu-liu)

[Chen-Hsuan Lin](/person/chen-hsuan-lin)

 

 

 ## Publication Date



Thursday, June 1, 2023

 

 ## Published in



[CVPR 2023](https://cvpr2023.thecvf.com/)

 

 ## Research Area



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

[Computer Graphics](/research-area/computer-graphics)

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

[High Performance Computing](/research-area/high-performance-computing)

[Real-Time Rendering](/research-area/real-time-rendering)

 

 

 ## External Links



[Project Page](https://research.nvidia.com/labs/dir/neuralangelo)

 

 

 ## Uploaded Files



[Li\_Neuralangelo\_High-Fidelity\_Neural\_Surface\_Reconstruction\_CVPR\_2023\_paper.pdf](https://d1qx31qr3h6wln.cloudfront.net/publications/Li_Neuralangelo_High-Fidelity_Neural_Surface_Reconstruction_CVPR_2023_paper.pdf "Open file in new window")3.08 MB

 

 

 ## Award



The Best Inventions of 2023, TIME Magazine