Bifurcated convolutional network for Specular Highlight Removal
- Article
- Figures
- Metrics
- Preview PDF
- Reference
- Related
- Cited by
- Materials
Abstract:
Specular highlight usually causes serious information degradation, which leads to the failure of many computer vision algorithms. We have proposed a novel bifurcated convolution neural network (Bifurcated-CNN) to tackle the problem of high reflectivity image information degradation. 1) The specular highlight features are extracted and removed in two stages from coarse to fine, to ensure the generation of diffuse images have no visual artifacts and information distortions. 2) A bifurcated feature selection strategy (BFSS) is designed to filter out the specular highlight features and enhance the detection capability of our network. The experiments on two types of challenging datasets demonstrate that our method outperforms state-of-the-art approaches for specular highlight detection and removal. The effectiveness of the proposed BFSS and Bifurcated-CNN are also verified.
Keywords:
Project Supported:
The National Key Technologies R&D Program of China
Clc Number:
