1. [Publications](/publications)
2. RGB-D Local Implicit Function for Depth Completion of Transparent Objects
 
 # RGB-D Local Implicit Function for Depth Completion of Transparent Objects

  ![](/sites/default/files/styles/wide/public/publications/teaser_6.png?itok=SoiSRwov)

 Majority of the perception methods in robotics require depth information provided by RGB-D cameras. However, standard 3D sensors fail to capture depth of transparent objects due to refraction and absorption of light. In this paper, we introduce a new approach for depth completion of transparent objects from a single RGB-D image. Key to our approach is a local implicit neural representation built on ray-voxel pairs that allows our method to generalize to unseen objects and achieve fast inference speed. Based on this representation, we present a novel framework that can complete missing depth given noisy RGB-D input. We further improve the depth estimation iteratively using a self-correcting refinement model. To train the whole pipeline, we build a large scale synthetic dataset with transparent objects. Experiments demonstrate that our method performs significantly better than the current state-of-the-art methods on both synthetic and real world data. In addition, our approach improves the inference speed by a factor of 20 compared to the previous best method, ClearGrasp.



 ## Authors



Luyang Zhu (University of Washington)

Arsalan Mousavian (NVIDIA)

Yu Xiang (NVIDIA)

Hammad Mazhar (NVIDIA)

Jozef van Eenbergen (NVIDIA)

Shoubhik Debnath (NVIDIA)

Dieter Fox (NVIDIA)

 

 

 ## Publication Date



Monday, March 29, 2021

 

 ## Published in



[CVPR 2021](http://cvpr2021.thecvf.com/)

 

 ## Research Area



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

[Robotics](/research-area/robotics)

 

 

 ## External Links



[Paper](https://arxiv.org/pdf/2104.00622.pdf)