Colorization Using Neural Network Ensemble

Cheng, Zezhou; Yang, Qingxiong; Sheng, Bin

doi:10.1109/tip.2017.2740620

Cited by 32 publications

(13 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The existence of dissimilar datasets increases the amount of training material, but it also creates setbacks. For example, if using semantic labels, inconsistency in categories between different datasets leads to incompatibility [44]. Also, ResNet architecture with skip connections is frequent in colorization neural network construction [32].…”

Section: ) Commonalities Of the Deep Learning Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Grayscale Image Colorization Methods: Overview and Evaluation

et al. 2021

View full text Add to dashboard Cite

Colorization is a process of transforming grayscale images to color images in a visually acceptable way. The main goal is to convince a viewer in the authenticity of the result. Grayscale images requiring colorization are in most cases images with natural scenes. Over the past 20 years a wide range of colorization methods has been developedfrom algorithmically simple, yet time-and energy-consuming because of unavoidable human intervention to more complicated, but simultaneously more automated methods. Automatic conversion has become a challenging area that links machine learning and deep learning with art. This paper presents an overview and evaluation of the grayscale image colorization methods and techniques applied to natural images. The paper provides classification of the existing colorization methods, explains the principles which they are based on and emphasizes their advantages and disadvantages. Special focus is put on deep learning methods. Relevant methods are compared regarding image quality and processing time. Different metrics for colored image quality assessment are used in this paper. Measuring the perceived quality of a colored image is extremely challenging because of the complexity of the human visual system. Multiple metrics used for colorization method evaluation provide results by determining the difference between the predicted color value and the ground-truth, which in several cases is not in coherence with image plausibility. The results indicate that user-guided colorization neural networks are the most promising colorization category because of the successful merge of human intervention and neural network automation. INDEX TERMSAutomatic methods, black-and-white image, colorfulness, colorization, deep learning methods, example-based methods, grayscale image, image quality assessment, scribble-based methods, userguided methods.

show abstract

Section: ) Commonalities Of the Deep Learning Methodsmentioning

confidence: 99%

“…Although producing a broad range of convincing results, simple neural networks might experience difficulties with accurately capturing the color characteristics for many different scenes with distinct color styles [44]. The results often contain improper colors and noticeable artifacts [9].…”

Section: ) Plain Colorization Neural Networkmentioning

confidence: 99%

Grayscale Image Colorization Methods: Overview and Evaluation

et al. 2021

View full text Add to dashboard Cite

show abstract

“…In addition, it has been used in colorization, that is, conversion of grayscale image to a colored one, which is supposed to be a challenging task since it requires labeling scribbles on the target images manually, or curation of a diverse set of colored reference images. 55 An electrocardiogram beat classification system was implemented by BIRCNN, that is, Bidirectional Recurrent Neural Network (BIRNN) combined with Convolutional Neural Network (CNN). The morphological features were drawn out using CNN, followed by considering them in the context using BIRNN.…”

Section: F X E X ( ) / (mentioning

confidence: 99%

Utilization of Time Series Tools in Life-sciences and Neuroscience

2020

View full text Add to dashboard Cite

Time series tools are part and parcel of modern day research. Their usage in the biomedical field; specifically, in neuroscience, has not been previously quantified. A quantification of trends can tell about lacunae in the current uses and point towards future uses. We evaluated the principles and applications of few classical time series tools, such as Principal Component Analysis, Neural Networks, common Auto-regression Models, Markov Models, Hidden Markov Models, Fourier Analysis, Spectral Analysis, in addition to diverse work, generically lumped under time series category. We quantified the usage from two perspectives, one, information technology professionals’, other, researchers utilizing these tools for biomedical and neuroscience research. For understanding trends from the information technology perspective, we evaluated two of the largest open source question and answer databases of Stack Overflow and Cross Validated. We quantified the trends in their application in the biomedical domain, and specifically neuroscience, by searching literature and application usage on PubMed. While the use of all the time series tools continues to gain popularity in general biomedical and life science research, and also neuroscience, and so have been the total number of questions asked on Stack overflow and Cross Validated, the total views to questions on these are on a decrease in recent years, indicating well established texts, algorithms, and libraries, resulting in engineers not looking for what used to be common questions a few years back. The use of these tools in neuroscience clearly leaves room for improvement.

show abstract

“…Deep learning has recently been applied to image colourisation [29]- [36]. Such methods are fully automatic, capable of colourising an input grayscale image without reference, although they need a very large training dataset and may fail to produce desired results when semantic ambiguities exist.…”

Section: Related Workmentioning

confidence: 99%