Journal
2020 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
Volume -, Issue -, Pages 21-30Publisher
IEEE
DOI: 10.1109/CVPR42600.2020.00010
Keywords
-
Categories
Funding
- Louisiana Board of Regents [LEQSF(2018-21)-RD-A-10]
Ask authors/readers for more resources
Recovering the dynamic fluid surface is a long-standing challenging problem in computer vision. Most existing image-based methods require multiple views or a dedicated imaging system. Here we present a learning-based singleimage approach for 3D fluid surface reconstruction. Specifically, we design a deep neural network that estimates the depth and normal maps of a fluid surface by analyzing the refractive distortion of a reference background pattern. Due to the dynamic nature of fluid surfaces, our network uses recurrent layers that carry temporal information from previous frames to achieve spatio-temporally consistent reconstruction given a video input. Due to the lack of fluid data, we synthesize a large fluid dataset using physics-based fluid modeling and rendering techniques for network training and validation. Through experiments on simulated and real captured fluid images, we demonstrate that our proposed deep neural network trained on our fluid dataset can recover dynamic 3D fluid surfaces with high accuracy.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available