computer-vision-3

Image Shadow Removal

3260 papers • 126 benchmarks • 313 datasets

This task has no description! Would you like to contribute one?

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in image-shadow-removal-3

Trend

Dataset

Best Model

Actions

No benchmarks available.

Libraries

i

Use these libraries to find image-shadow-removal-3 models and implementations

Datasets

SD7K

Subtasks

No subtasks available.

Most implemented papers

Robust Graph Learning From Noisy Data

S. Hoi, Zenglin Xu, Zhao Kang, Haiqi Pan•Sun Dec 16 2018

This paper proposes a novel robust graph learning scheme to learn reliable graphs from the real-world noisy data by adaptively removing noise and errors in the raw data and shows that the proposed model outperforms the previous state-of-the-art methods.

281

Content

0

Paper Graph

BEDSR-Net: A Deep Shadow Removal Network From a Single Document Image

Yung-Yu Chuang, Yun-Hsuan Lin, Wen-Chin Chen•Sun May 31 2020

Removing shadows in document images enhances both the visual quality and readability of digital copies of documents. Most existing shadow removal algorithms for document images use hand-crafted heuristics and are often not robust to documents with different characteristics. This paper proposes the Background Estimation Document Shadow Removal Network (BEDSR-Net), the first deep network specifically designed for document image shadow removal. For taking advantage of specific properties of document images, a background estimation module is designed for extracting the global background color of the document. During the process of estimating the background color, the module also learns information about the spatial distribution of background and non-background pixels. We encode such information into an attention map. With the estimated global background color and attention map, the shadow removal network can better recover the shadow-free image. We also show that the model trained on synthetic images remains effective for real photos, and provide a large set of synthetic shadow images of documents along with their corresponding shadow-free images and shadow masks. Extensive quantitative and qualitative experiments on several benchmarks show that the BEDSR-Net outperforms existing methods in enhancing both the visual quality and readability of document images.

78 0

Paper Graph

Auto-Exposure Fusion for Single-Image Shadow Removal

Qing Guo, Wei Feng, Hongkai Yu, Yang Liu, Felix Juefei-Xu, Lan Fu, Changqing Zhou, Song Wang•Sun Feb 28 2021

This paper proposes the shadow-aware FusionNet that takes the shadow image as input to generate fusion weight maps across all the over-exposure images, and proposes the boundary-aware RefineNet to eliminate the remaining shadow trace further.

164 0

Paper Graph

Efficient Model-Driven Network for Shadow Removal

Zhiwei Xiong, Zhengjun Zha, Xueyang Fu, Yurui Zhu, Zeyu Xiao, Yanchi Fang•Mon Jun 27 2022

This work proposes a new shadow illumination model, which ensures the identity mapping among unshaded regions, and adaptively performs fine grained spatial mapping between shadow regions and their references, and reformulates the shadow removal task as a variational optimization problem.

77 0

Paper Graph

High-Resolution Document Shadow Removal via A Large-Scale Real-World Dataset and A Frequency-Aware Shadow Erasing Net

Xuhang Chen, Chi-Man Pun, Zinuo Li, Xiaodong Cun•Sat Aug 26 2023

This work handles high-resolution document shadow removal directly via a larger-scale real-world dataset and a carefully-designed frequency-aware network, which shows a clearly better performance than previous methods in terms of visual quality and numerical results.

65 0

Paper Graph

DC-ShadowNet: Single-Image Hard and Soft Shadow Removal Using Unsupervised Domain-Classifier Guided Network

R. Tan, Yeying Jin, Aashish Sharma•Thu Sep 30 2021

This paper proposes an unsupervised domain-classifier guided shadow removal network, DC-ShadowNet, designed to integrate a shadow/shadow-free domain classifier into a generator and its discriminator, enabling them to focus on shadow regions.

152 0

Paper Graph

DeS3: Adaptive Attention-Driven Self and Soft Shadow Removal Using ViT Similarity

R. Tan, Wenhan Yang, Yeying Jin, Yuan Yuan, W. Ye•Mon Nov 14 2022

This paper presents DeS3, a method that removes hard, soft and self shadows based on adaptive attention and ViT similarity, and outperforms state-of-the-art methods on the SRD, AISTD, LRSS, USR and UIUC datasets.

47 0

Paper Graph

Shadocnet: Learning Spatial-Aware Tokens in Transformer for Document Shadow Removal

Xuhang Chen, Chi-Man Pun, Shuqiang Wang, Xiaodong Cun•Tue Nov 29 2022

A Transformer-based model for document shadow removal is proposed that utilizes shadow context encoding and decoding in both shadow and shadow-free regions and is competitive with state-of-the-art methods.

32 0

Paper Graph

ShadowDiffusion: When Degradation Prior Meets Diffusion Model for Shadow Removal

H. Pfister, B. Wen, Wenhan Yang, Lanqing Guo, Siyu Huang, Yufei Wang•Thu Dec 08 2022

A unified diffusion framework that integrates both the image and degradation priors for highly effective shadow removal and progressively refines the estimated shadow mask as an auxiliary task of the diffusion generator, which leads to more accurate and robust shadow-free image generation.

149 0

Paper Graph

Adding a benchmark result helps the community track progress.