Deep Koalarization: Image Colorization using CNNs and Inception-ResNet-v2

Abstract

We review some of the most recent approaches to colorize gray-scale images using deep learning methods. Inspired by these, we propose a model which combines a deep Convolutional Neural Network trained from scratch with high-level features extracted from the Inception-ResNet-v2 pre-trained model. Thanks to its fully convolutional architecture, our encoder-decoder model can process images of any size and aspect ratio. Other than presenting the training results, we assess the "public acceptance" of the generated images by means of a user study. Finally, we present a carousel of applications on different types of images, such as historical photographs.

12 pages

References in corpus (5)

• Deep Residual Learning for Image Recognition
• Striving for Simplicity: The All Convolutional Net
• Rethinking the Inception Architecture for Computer Vision
• Learning Representations for Automatic Colorization
• Real-Time User-Guided Image Colorization with Learned Deep Priors

Cited by in corpus (7)

• Vector database management systems: Fundamental concepts, use-cases, and current challenges
• Unsupervised machine learning of topological phase transitions from experimental data
• Deep Exemplar-based Video Colorization
• Learning Image Aesthetic Assessment from Object-level Visual Components
• FlowChroma -- A Deep Recurrent Neural Network for Video Colorization
• ViT-Inception-GAN for Image Colourising
• Perturbation Analysis of Learning Algorithms: A Unifying Perspective on Generation of Adversarial Examples