CNN-Based Luminance And Color Correction For ILL-Exposed Images

Steffens, Cristiano Rafael; Huttner, Valquiria; Messias, Lucas Ricardo Vieira; Drews, Paul; Botelho, Silvia S. C.; Guerra, Rodrigo da Silva

doi:10.1109/icip.2019.8803546

Cited by 6 publications

(7 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…All experiments are carried out on an Intel(R) Core (TM) i7-8500U CPU @1.80GHz 4-Core 8GB RAM machine using MAT-LAB R2019b. The default settings for algorithms of Masood [8] and Steffens [13] are used for comprehensive comparisons. The average execution time of the proposed method is 11.56sec for the whole dataset, while it is 7.90sec and 24.03sec for Masood and Steffens, respectively.…”

Section: Experimental Results and Conclusionmentioning

confidence: 99%

“…DeclipNet is able to recover information in clipped regions successfully. In a recent study of Steffens et al [13], a deep convolutional neural network model is proposed for contrast enhancement and restoration of images. The designed model is able to recover clipped pixels in both under-and over-saturated areas and it produces both visually and statistically successful results.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Saturated Region Recovery in Tone-Mapped HDR Images

Ulucan

Karakaya

Türkan

2021

2021 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

Tone-mapping is one of the prevailing methods to overcome high dynamic range imaging limitations over low dynamic range display devices, but the tone-mapped output image may suffer from saturated regions with texture and color information loss. In this paper, a novel approach is proposed to solve the so-called clipping problem in tone-mapped high dynamic range images. A successful saturation correction framework, which relies on linear embeddings, difference of pixel intensities and gradient-guided block-search, is developed as a post-processing technique to tone-mapping. Experimental results demonstrate that the proposed method successfully recovers clipped regions for the saturation problem in tone-mapped output images while avoiding artifacts.

show abstract

Section: Experimental Results and Conclusionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Saturated Region Recovery in Tone-Mapped HDR Images

Ulucan

Karakaya

Türkan

2021

2021 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

show abstract

“…[15] recently proposed an overexposure correction method based on a concatenated encoderdecoder UNet networks [21], processing different levels of the Laplacian pyramid outputs of the input and reference images. Other earlier methods again utilized auto-encoder and GAN based approaches [15,22,14]. The proposed models are mainly trained on paired data sets generated from raw-RGB (such as MIT-Adobe FiveK data set [23,15]) and multiple exposure based images [24,17,4].…”

Section: Deep Learning Based Over-exposure Enhancementmentioning

confidence: 99%

“…The proposed models are mainly trained on paired data sets generated from raw-RGB (such as MIT-Adobe FiveK data set [23,15]) and multiple exposure based images [24,17,4]. This work evaluates the two CNN models of Steffens et al [22,14] and my adaptations of two GAN architectures [20,19] for over-exposure enhancement, for their content and color fidelity performances. Brief descriptions of the models are given as follows.…”

Section: Deep Learning Based Over-exposure Enhancementmentioning

confidence: 99%

See 1 more Smart Citation

Content Fidelity of Deep Learning Methods for Clipping and Over-exposure Correction

Abebe¹

2021

lim

View full text Add to dashboard Cite

Exposure problems, due to standard camera sensor limitations, often lead to image quality degradations such as loss of details and change in color appearance. The quality degradations further hiders the performances of imaging and computer vision applications. Therefore, the reconstruction and enhancement of uderand over-exposed images is essential for various applications. Accordingly, an increasing number of conventional and deep learning reconstruction approaches have been introduced in recent years. Most conventional methods follow color imaging pipeline, which strongly emphasize on the reconstructed color and content accuracy. The deep learning (DL) approaches have conversely shown stronger capability on recovering lost details. However, the design of most DL architectures and objective functions don’t take color fidelity into consideration and, hence, the analysis of existing DL methods with respect to color and content fidelity will be pertinent. Accordingly, this work presents performance evaluation and results of recent DL based overexposure reconstruction solutions. For the evaluation, various datasets from related research domains were merged and two generative adversarial networks (GAN) based models were additionally adopted for tone mapping application scenario. Overall results show various limitations, mainly for severely over-exposed contents, and a promising potential for DL approaches, GAN, to reconstruct details and appearance.

show abstract