192 citations · 197 across the 8 of their papers we have counts for
8 papers · 1 filter
Exploring Spectral Characteristics for Single Image Reflection Removal
Pengbo Guo, Chengxu Liu, Guoshuai Zhao +3
Eliminating reflections caused by incident light interacting with reflective medium remains an ill-posed problem in the image restoration area. The primary challenge arises from th…
Learning Attribute-aware Representations for Few-shot Scene Text Segmentation
Yifan Tang, Chenming Li, Chengxu Liu +7
Supervised scene text segmentation has achieved notable progress in recent years. However, its development is largely constrained by the scarcity of high-quality datasets and the h…
QueryCDR: Query-Based Controllable Distortion Rectification Network for Fisheye Images
Pengbo Guo, Chengxu Liu, Xingsong Hou +1
Fisheye image rectification aims to correct distortions in images taken with fisheye cameras. Although current models show promising results on images with a similar degree of dist…
Dual Degradation-Inspired Deep Unfolding Network for Low-Light Image Enhancement
Huake Wang, Xingsong Hou, Chengcu Liu +3
Although low-light image enhancement has achieved great stride based on deep enhancement models, most of them mainly stress on enhancement performance via an elaborated black-box n…
Division Gets Better: Learning Brightness-Aware and Detail-Sensitive Representations for Low-Light Image Enhancement
Huake Wang, Xiaoyang Yan, Xingsong Hou +3
Low-light image enhancement strives to improve the contrast, adjust the visibility, and restore the distortion in color and texture. Existing methods usually pay more attention to…
Sparsity and Coefficient Permutation Based Two-Domain AMP for Image Block Compressed Sensing
Junhui Li, Xingsong Hou, Huake Wang +1
The learned denoising-based approximate message passing (LDAMP) algorithm has attracted great attention for image compressed sensing (CS) tasks. However, it has two issues: first,…
Absolute Zero-Shot Learning
Rui Gao, Fan Wan, Daniel Organisciak +6
Considering the increasing concerns about data copyright and privacy issues, we present a novel Absolute Zero-Shot Learning (AZSL) paradigm, i.e., training a classifier with zero r…
Noisy-As-Clean: Learning Self-supervised Denoising from the Corrupted Image
Jun Xu, Yuan Huang, Ming-Ming Cheng +4
Supervised deep networks have achieved promisingperformance on image denoising, by learning image priors andnoise statistics on plenty pairs of noisy and clean images. Unsupervised…