From Shadow Segmentation to Shadow Removal

Le, Hieu Van; Samaras, Dimitris

doi:10.1007/978-3-030-58621-8_16

Cited by 61 publications

(79 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based on SOBA dataset, we remove all the shadows to construct our DEshadowed Shadow-OBject Association (DESOBA) dataset, which can be used for shadow generation task. At the start, we tried to remove the shadows with the state-of-the-art deshadow methods [12,17,44,18,8]. However, their performance is far from satisfactory due to complex scenes.…”

Section: Introductionmentioning

confidence: 99%

Shadow Generation for Composite Image in Real-world Scenes

Yan¹,

Niu²,

Zhang³

et al. 2021

Preprint

View full text Add to dashboard Cite

Image composition targets at inserting a foreground object on a background image. Most previous image composition methods focus on adjusting the foreground to make it compatible with background while ignoring the shadow effect of foreground on the background. In this work, we focus on generating plausible shadow for the foreground object in the composite image. First, we contribute a real-world shadow generation dataset DESOBA by generating synthetic composite images based on paired real images and deshadowed images. Then, we propose a novel shadow generation network SGRNet, which consists of a shadow mask prediction stage and a shadow filling stage. In the shadow mask prediction stage, foreground and background information are thoroughly interacted to generate foreground shadow mask. In the shadow filling stage, shadow parameters are predicted to fill the shadow area. Extensive experiments on our DESOBA dataset and real composite images demonstrate the effectiveness of our proposed method.

show abstract

Section: Introductionmentioning

confidence: 99%

Shadow Generation for Composite Image in Real-world Scenes

Yan¹,

Niu²,

Zhang³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Recently, with the use of the convolutional neural networks (CNNs), many learning based shadow removal approaches [25,32,9,10,16,22,20,17] have been proposed, Figure 1. An illustration of our basic idea of incorporating shadow generation for learning shadow removal by using only shadow images.…”

Section: Introductionmentioning

confidence: 99%

“…In addition, in practice, it is still difficult to capture a large set of shadow-free images with good variety. To handle this problem, in their ECCV20 paper [17], Le and Samaras introduces a novel unsupervised shadow removal method using only shadow images. More specifically, they leverage the fact that a shadow image usually contains both shadow and non-shadow regions.…”

Section: Introductionmentioning

confidence: 99%

“…Although the patch cropping in [17] only requires the shadow masks that can be obtained by using an existing shadow detection method [43,18,41,9,33], it involves a careful design of cropping-window size, a set of strict physics-based constraints, and heavy computational load [17]. In this paper, we propose a new approach to address these problems by incorporating a shadow generation module, while keeping the desirable property of using only shadow images, as illustrated in Fig.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

From Shadow Generation to Shadow Removal

Liu

Yin

et al. 2021

Preprint

View full text Add to dashboard Cite

Shadow removal is a computer-vision task that aims to restore the image content in shadow regions. While almost all recent shadow-removal methods require shadowfree images for training, in ECCV 2020 Le and Samaras introduces an innovative approach without this requirement by cropping patches with and without shadows from shadow images as training samples. However, it is still laborious and time-consuming to construct a large amount of such unpaired patches. In this paper, we propose a new G2R-ShadowNet which leverages shadow generation for weaklysupervised shadow removal by only using a set of shadow images and their corresponding shadow masks for training. The proposed G2R-ShadowNet consists of three subnetworks for shadow generation, shadow removal and refinement, respectively and they are jointly trained in an endto-end fashion. In particular, the shadow generation subnet stylises non-shadow regions to be shadow ones, leading to paired data for training the shadow-removal sub-net. Extensive experiments on the ISTD dataset and the Video Shadow Removal dataset show that the proposed G2R-ShadowNet achieves competitive performances against the current state of the arts and outperforms Le and Samaras' patch-based shadow-removal method.

show abstract

“…Therefore, shadow detection has drawn a lot of interest in recent years especially with the rapid development of deep-learning-based methods [6,17,23,32,36]. However, recent shadow detection works mostly deal with shadows in single images while video shadow detection remains an open question despite many potential applications [14,15,16,33].…”

Section: Introductionmentioning

confidence: 99%

Temporal Feature Warping for Video Shadow Detection

Hu,

Le,

Samaras

2021

Preprint

Self Cite

View full text Add to dashboard Cite

While single image shadow detection has been improving rapidly in recent years, video shadow detection remains a challenging task due to data scarcity and the difficulty in modelling temporal consistency. The current video shadow detection method achieves this goal via co-attention, which mostly exploits information that is temporally coherent but is not robust in detecting moving shadows and small shadow regions. In this paper, we propose a simple but powerful method to better aggregate information temporally. We use an optical flow based warping module to align and then combine features between frames. We apply this warping module across multiple deep-network layers to retrieve information from neighboring frames including both local details and high-level semantic information. We train and test our framework on the ViSha dataset. Experimental results show that our model outperforms the state-of-the-art video shadow detection method by 28%, reducing BER from 16.7 to 12.0.

show abstract

From Shadow Segmentation to Shadow Removal

Cited by 61 publications

References 46 publications

Shadow Generation for Composite Image in Real-world Scenes

Shadow Generation for Composite Image in Real-world Scenes

From Shadow Generation to Shadow Removal

Temporal Feature Warping for Video Shadow Detection

Contact Info

Product

Resources

About