ChromaGAN: Adversarial Picture Colorization with Semantic Class Distribution

@article{Vitoria2020ChromaGANAP,
  title={ChromaGAN: Adversarial Picture Colorization with Semantic Class Distribution},
  author={Patricia Vitoria and Lara Raad and Coloma Ballester},
  journal={2020 IEEE Winter Conference on Applications of Computer Vision (WACV)},
  year={2020},
  pages={2434-2443}
}
The colorization of grayscale images is an ill-posed problem, with multiple correct solutions. In this paper, we propose an adversarial learning colorization approach coupled with semantic information. A generative network is used to infer the chromaticity of a given grayscale image conditioned to semantic clues. This network is framed in an adversarial model that learns to colorize by incorporating perceptual and semantic understanding of color and class distributions. The model is trained via… 

Figures and Tables from this paper

Human vs Objective Evaluation of Colourisation Performance
Automatic colourisation of grey-scale images is the process of creating a full-colour image from the grey-scale prior. It is an ill-posed problem, as there are many plausible colourisations for a
Analysis of Different Losses for Deep Learning Image Colorization
TLDR
This chapter analyzes the impact of the loss function on the estimated colorization results by reviewing the losses and evaluation metrics used in the literature and results show that the models trained with VGG-based LPIPS provide overall slightly better results for most evaluation metrics.
Influence of Color Spaces for Deep Learning Image Colorization
TLDR
The importance of carefully designing the architecture and evaluation protocols depending on the types of images that are being processed and their specificities is shown: strong/small contours, few/many objects, recent/archive images.
Image colorization using Scaled-YOLOv4 detector
TLDR
The effect of preceding the image colorization with an object detection phase, such that the colorization will be made for each object individually as well as the full image, is studied.
Fine‐grained semantic ethnic costume high‐resolution image colorization with conditional GAN
TLDR
This study introduces fine‐grained level semantic information and proposes a high‐resolution image colorization model for ethnic costumes targeting enhancement, inspired by semantic‐level colorization.
Two-stage pyramidal convolutional neural networks for image colorization
TLDR
This research addresses the issue of existing algorithms by presenting a two-stage convolutional neural network structure with the first and second stages being a chroma map generation network and a refinement network, respectively.
Focusing on Persons: Colorizing Old Images Learning from Modern Historical Movies
TLDR
Various quantitative and qualitative comparisons demonstrate that the HistoryNet focusing on historical person's diverse high fidelity clothing colorization outperforms the state-of-the-art colorization methods, especially on military uniforms, which has correct colors according to the historical literatures.
JPEG Robust Invertible Grayscale.
TLDR
Extensive experiments demonstrate that the proposed method is superior to the original invertible grayscale work both qualitatively and quantitatively while ensuring the JPEG robustness.

References

SHOWING 1-10 OF 54 REFERENCES
ImageNet: A large-scale hierarchical image database
TLDR
A new database called “ImageNet” is introduced, a large-scale ontology of images built upon the backbone of the WordNet structure, much larger in scale and diversity and much more accurate than the current image datasets.
Image-to-Image Translation with Conditional Adversarial Networks
TLDR
Conditional adversarial networks are investigated as a general-purpose solution to image-to-image translation problems and it is demonstrated that this approach is effective at synthesizing photos from label maps, reconstructing objects from edge maps, and colorizing images, among other tasks.
Generative Adversarial Nets
We propose a new framework for estimating generative models via an adversarial process, in which we simultaneously train two models: a generative model G that captures the data distribution, and a
Let there be Color!: Joint End-to-end Learning of Global and Local Image Priors for Automatic Image Colorization with Simultaneous Classification
TLDR
A novel technique to automatically colorize grayscale images that combines both global priors and local image features and can process images of any resolution, unlike most existing approaches based on CNN.
Wasserstein Generative Adversarial Networks
TLDR
This work introduces a new algorithm named WGAN, an alternative to traditional GAN training that can improve the stability of learning, get rid of problems like mode collapse, and provide meaningful learning curves useful for debugging and hyperparameter searches.
Wasserstein GAN
TLDR
The problem this paper is concerned with is that of unsupervised learning, what does it mean to learn a probability distribution and how to define a parametric family of densities.
Colorful Image Colorization
TLDR
This paper proposes a fully automatic approach to colorization that produces vibrant and realistic colorizations and shows that colorization can be a powerful pretext task for self-supervised feature learning, acting as a cross-channel encoder.
Learning Representations for Automatic Colorization
TLDR
A fully automatic image colorization system that leverages recent advances in deep networks, exploiting both low-level and semantic representations, and explores colorization as a vehicle for self-supervised visual representation learning.
Very Deep Convolutional Networks for Large-Scale Image Recognition
TLDR
This work investigates the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting using an architecture with very small convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers.
Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks
The architecture introduced in this paper learns a mapping function G : X 7→ Y using an adversarial loss such thatG(X) cannot be distinguished from Y , whereX and Y are images belonging to two
...
...