A fully-automatic image colorization scheme using improved CycleGAN with skip connections

Huang, Shanglian; Jin, Xin; Jiang, Qian; Li, Jie; Lee, Shin Jye; Wang, Puming; Yao, Shaowen

doi:10.1007/s11042-021-10881-5

Cited by 14 publications

(4 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These collected images are cropped and preprocessed to obtain source domain X images and target domain Y images for the image-to-image translation model. ( 2 necessitating the reasonable design of loss functions and weights during the training process to allow CycleGAN to learn better image conversion mappings [37][38][39].…”

Section: Proposed Methodsmentioning

confidence: 99%

“…Therefore, careful parameter tuning and optimization of CycleGAN are required to obtain high-quality synthetic images. Secondly, there may be significant differences between CAD images and real images, necessitating the reasonable design of loss functions and weights during the training process to allow CycleGAN to learn better image conversion mappings [37][38][39].…”

Section: The Cyclegan-based Image Translation Networkmentioning

confidence: 99%

See 1 more Smart Citation

Synthetic Data Generation Based on RDB-CycleGAN for Industrial Object Detection

Hu,

Xiao,

Jin

et al. 2023

Mathematics

View full text Add to dashboard Cite

Deep learning-based methods have demonstrated remarkable success in object detection tasks when abundant training data are available. However, in the industrial domain, acquiring a sufficient amount of training data has been a challenge. Currently, many synthetic datasets are created using 3D modeling software, which can simulate real-world scenarios and objects but often cannot achieve complete accuracy and realism. In this paper, we propose a synthetic data generation framework for industrial object detection tasks based on image-to-image translation. To address the issue of low image quality that can arise during the image translation process, we have replaced the original feature extraction module with the Residual Dense Block (RDB) module. We employ the RDB-CycleGAN network to transform CAD models into realistic images. Additionally, we have introduced the SSIM loss function to strengthen the network constraints of the generator and conducted a quantitative analysis of the improved RDB-CycleGAN-generated synthetic data. To evaluate the effectiveness of our proposed method, the synthetic data we generate effectively enhance the performance of object detection algorithms on real images. Compared to using CAD models directly, synthetic data adapt better to real-world scenarios and improve the model’s generalization ability.

show abstract

Section: Proposed Methodsmentioning

confidence: 99%

Section: The Cyclegan-based Image Translation Networkmentioning

confidence: 99%

Synthetic Data Generation Based on RDB-CycleGAN for Industrial Object Detection

Hu,

Xiao,

Jin

et al. 2023

Mathematics

View full text Add to dashboard Cite

show abstract

“…Two U-NET-based colouring approaches were implemented by Qiao et al [22], with the following improvements over prior work: Moreover, the network will also supply users with a data-driven colour palette, proposing the optimal hue of the grey map at a specific place based on its training to directly predict the mapping from grayscale photos with coloured dots to colour images. This method has the potential to save time for its users and can use the global histogram of a colour reference map to colour the grey map [23]. This research shows that the aforementioned techniques have given much thought to the problem of automatically colouring grayscale images.…”

Section: Related Workmentioning

confidence: 97%

An approach to pseudocoloring of grey scale image using deep learning technique

Srivastava

Gogia²,

Rohith³

2023

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

Image pseudo colorization is the process of adding RGB colours to grayscale images to make them more appealing. Deep learning technology has made progress in the field of automatic colouring. In general, we divide automatic colouring methods into three groups based on where the colour information comes from: colouring based on what you already know and on reference pictures. The colouring method can meet the needs of most users, but there are some drawbacks. For example, users can’t colour different reference graphs for the different things in a picture. In order to solve this problem by recognising several objects and background regions in a picture and combine the final colouring results, the proposed method uses the deep learning approach that regional mixed colours be used more and the method be mastered by using deep learning. Qualitative results (visual perception) validate the effectiveness of pseudocolorisation which split into foreground colour based on a reference picture and background colour based on prior knowledge. Quantitative results such as Structural Similarity (SSIM), Peak Signal to Noise Ratio (PSNR), Image Matching Error and Entropy validates the effectiveness of strong edge strength, visually appealing quality and retention of maximum information without disturbing quality of image.

show abstract

“…However, some problem still exists. For example, no scholar has applied the CNN model to this field till now, so the research here is still a blank, which has great theoretical research and practical application value for logistics enterprises [ 34 ].…”

Section: Related Workmentioning

confidence: 99%

Automatic Gray Image Coloring Method Based on Convolutional Network

Fan

Xie

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

Image coloring is a time-consuming and laborious work. For a work, color collocation is an important factor to determine its quality. Therefore, automatic image coloring is a topic with great research significance and application value. With the development of computer hardware, deep learning technology has achieved satisfactory results in the field of automatic coloring. According to the source of color information, this paper can divide automatic coloring methods into three types: image coloring based on prior knowledge, image coloring based on reference pictures, and interactive coloring. The coloring method can meet the needs of most users, but there are disadvantages such as users cannot get the multiple objects in a picture of different reference graph coloring. Aiming at this problem, based on the instance of color image segmentation and image fusion technology, the use of deep learning is proposed to implement regional mixed color more and master the method. It can be divided into foreground color based on reference picture and background color based on prior knowledge. In order to identify multiple objects and background areas in the image and fuse the final coloring results together, a method of image coloring based on CNN is proposed in this paper. Firstly, CNN is used to extract their semantic information, respectively. According to the extractive semantic information, the color of the designated area of the reference image is transferred to the designated area of the grayscale image. During the transformation, images combined with semantic information are input into CNN model to obtain the content feature map of grayscale image and the style feature map of reference image. Then, a random noise map is iterated to make the noise map approach the content feature map as a whole and the specific target region approach the designated area of the style feature map. Experimental results show that the proposed method has good effect on image coloring and has great advantages in network volume and coloring effect.

show abstract

A fully-automatic image colorization scheme using improved CycleGAN with skip connections

Cited by 14 publications

References 53 publications

Synthetic Data Generation Based on RDB-CycleGAN for Industrial Object Detection

Synthetic Data Generation Based on RDB-CycleGAN for Industrial Object Detection

An approach to pseudocoloring of grey scale image using deep learning technique

Automatic Gray Image Coloring Method Based on Convolutional Network

Contact Info

Product

Resources

About