A Weighted Variational Model for Simultaneous Reflectance and Illumination Estimation

Fu, Xueyang; Zeng, Delu; Huang, Yue; Zhang, Xiaoping; Ding, Xinghao

doi:10.1109/cvpr.2016.304

Cited by 788 publications

(663 citation statements)

References 35 publications

Supporting

Mentioning

662

Contrasting

Order By: Relevance

“…The comparison results are shown in Figure . From Figure , it is clear that compared to the low‐light images, semantic filtering (SFT) does not show obvious correction effect, LIME and SRIE illy shift the color distribution. Our DIMNet achieves the most appealing visual effect with learning the combination of both high‐level semantics and low‐level features.…”

Section: Experiments and Resultsmentioning

confidence: 99%

“…(c) Low‐light image enhancement via illumination map estimation (LIME) . (d) Simultaneous reflectance and illumination estimation (SRIE) . (e) Deep Intensity Manipulation Network (DIMNet)…”

Section: Experiments and Resultsmentioning

confidence: 99%

“…MIoU comparison of correction methods attached with segmentation algorithms. There are six groups of inputs: low‐light images, semantic filtering (SFT), low‐light image enhancement via illumination map estimation (LIME), simultaneous reflectance and illumination estimation (SRIE), proposed cascaded network, and ground‐truth (GT) normal image. For each group, we attach a SegNet and an ICNet to perform a semantic segmentation on CamVid and Cityscapes, respectively…”

Section: Experiments and Resultsmentioning

confidence: 99%

“…We employ the detection network as a detector. Similarly, we compare with low‐light images directly and enhancement methods LIME and SRIE on two randomly selected examples from our testing data set. The visual comparisons are shown in Figure .…”

Section: Experiments and Resultsmentioning

confidence: 99%

“…We select two sample low‐light images from our testing data set. Low‐light image enhancement via illumination map estimation (LIME), simultaneous reflectance and illumination estimation (SRIE), and ours are processed on these samples. The original low‐light images, two groups of processed images, are fed into the detection algorithm, respectively.…”

Section: Experiments and Resultsmentioning

confidence: 99%

See 4 more Smart Citations

Cascaded network with deep intensity manipulation for scene understanding

Yang

Wang

Chen

et al. 2019

Computer Animation & Virtual

View full text Add to dashboard Cite

Scene understanding is essential to robotic navigation and autonomous driving as it provides semantic information to their controlling system. However, it will fail when processing low‐light images/videos captured under adverse weather or at night use state‐of‐the‐art scene understanding methods. A naive way to directly infer semantics from low‐light images is ill posed because the low‐light condition distorts pixel intensities and buries details. In order to address this problem, we propose the Deep Intensity Manipulation Network (DIMNet), which could relight the input images and recover the details, and combine the DIMNet with a scene understanding network to get a cascaded network to learn the semantics from low‐light images. Through learning pixel intensity manipulation, our method can generate images not only visually pleasing but also practical for scene understanding. Qualitative and quantitative experiments demonstrate that the proposed method is effective and robust for both synthetic and real‐world images.

show abstract

Section: Experiments and Resultsmentioning

confidence: 99%

Section: Experiments and Resultsmentioning

confidence: 99%

Section: Experiments and Resultsmentioning

confidence: 99%

Section: Experiments and Resultsmentioning

confidence: 99%

Section: Experiments and Resultsmentioning

confidence: 99%

See 3 more Smart Citations

Cascaded network with deep intensity manipulation for scene understanding

Yang

Wang

Chen

et al. 2019

Computer Animation & Virtual

View full text Add to dashboard Cite

show abstract

AFF‐Dehazing: Attention‐based feature fusion network for low‐light image Dehazing

Chen

Sheng

et al. 2021

Computer Animation & Virtual

View full text Add to dashboard Cite

Images captured in haze conditions, especially at nighttime with low light, often suffer from degraded visibility, contrasts, and vividness, which makes it difficult to carry out the following vision tasks. In this article, we propose an attention-based feature fusion network (AFF-Dehazing) for low-light image dehazing. Our method decomposes the low-light image dehazing into two task-independent streams containing four modules: image dehazing module, low-light feature extractor module, feature fusion module, and image restoration module. The basic block of these modules is the proposed attention-based residual dense block. Since the dual-branch are used, AFF-Dehazing can avoid learning the mixed degradation all-in-one and enhance the details of low-light haze images. Extensive experiments show that our method surpasses previous state-of-the-art image dehazing methods and low-light enhancement methods by a very large margin both quantitatively and qualitatively. K E Y W O R D S attention mechanism, image dehazing, low-light enhancement INTRODUCTIONHaze is common atmospheric phenomenon and often plagues the qualities of the images acquired by the cameras. While, at night with low light, the images further face severe degradations, including low visibility and intensity blurs. Recently, many efforts have been made to solve the problems of image dehazing and low-light image enhancement. However, existing methods dealt with the problems of image dehazing and low-light enhancement separately. These methods can achieve relatively good performance on corrupted images when removing a certain target type of distortion (i.e., haze or low-light), but they are not effective in removing both types of distortions.

show abstract

Denoising smooth‐detail representation of MRI‐SPECT fusion

Du,

Cheng,

Wang

et al. 2024

Int J Imaging Syst Tech

View full text Add to dashboard Cite

Fusing of brain imaging data is a hot task that has been previously approached by preserving saliency information from inputs to the output enhanced fused image. In this article, we present a method for calculating fused image from an anatomical image and a functional image, in which detail saliency information is preserved while suppressing noise. Our method is based on smooth‐detail representation used as a decomposition model scheme and based on denoising fusion rules to predict a fused version of the desired output. Unlike our previous approach, the method proposed in this article presents a smooth‐detail model for greyscale image with the extension a single‐channel to a three‐channel. We also adopt denoising method for map to construct the estimated fused image. The effectiveness of our method is proved in terms of analyzing noisy image, making comparison between proposed method and deep learning methods, and computing complexity of running time, showing better performance on denoising and preserving luminance saliency information.

show abstract

A Weighted Variational Model for Simultaneous Reflectance and Illumination Estimation

Cited by 788 publications

References 35 publications

Cascaded network with deep intensity manipulation for scene understanding

Cascaded network with deep intensity manipulation for scene understanding

AFF‐Dehazing: Attention‐based feature fusion network for low‐light image Dehazing

Denoising smooth‐detail representation of MRI‐SPECT fusion

Contact Info

Product

Resources

About