1. [Publications](/publications)
2. NeRFDeformer: NeRF Transformation from a Single View via 3D Scene Flows
 
 # NeRFDeformer: NeRF Transformation from a Single View via 3D Scene Flows

  ![](/sites/default/files/styles/wide/public/publications/truck.png?itok=FHt0L7D2)

 We present a method for automatically modifying a NeRF representation based on a single observation of a non-rigid transformed version of the original scene. Our method defines the transformation as a 3D flow, specifically as a weighted linear blending of rigid transformations of 3D anchor points that are defined on the surface of the scene. In order to identify anchor points, we introduce a novel correspondence algorithm that first matches RGB-based pairs, then leverages multi-view information and 3D reprojection to robustly filter false positives in two steps. We also introduce a new dataset for exploring the problem of modifying a NeRF scene through a single observation. Our dataset contains 113 synthetic scenes leveraging 47 3D assets. We show that our proposed method outperforms NeRF editing methods as well as diffusion-based methods, and we also explore different methods for filtering correspondences.



 ## Authors



Zhenggang Tang (UIUC)

Zhongzheng Ren (UIUC)

Xiaoming Zhao (UIUC)

[Bowen Wen](/person/bowen-wen)

[Jonathan Tremblay](/person/jonathan-tremblay)

[Stan Birchfield](/person/stan-birchfield)

Alexander Schwing (UIUC)

 

 

 ## Publication Date



Saturday, June 1, 2024

 

 ## Published in



[CVPR 2024](https://cvpr.thecvf.com/Conferences/2024)

 

 ## Research Area



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

 

 

 ## External Links



[paper](https://arxiv.org/abs/2406.10543)

[code](https://github.com/nerfdeformer/nerfdeformer)

[video](https://www.youtube.com/watch?v=oZsA6i9g_yM&t=1s)

[website](https://nerfdeformer.github.io/)