ViT-Inception-GAN for Image Colourising
This work addresses image colorization, a task with high degrees of freedom, for researchers and practitioners in computer vision, but it appears incremental as it builds on existing GAN and transformer architectures.
The paper tackled the problem of image colorization by proposing a ViT-Inception-GAN method with Inception-v3 fusion embedding in the generator and Vision Transformer as the discriminator, resulting in demonstrated improvements on the Unsplash and COCO datasets compared to ViT-GANs without the embedding.
Studies involving colourising images has been garnering researchers' keen attention over time, assisted by significant advances in various Machine Learning techniques and compute power availability. Traditionally, colourising images have been an intricate task that gave a substantial degree of freedom during the assignment of chromatic information. In our proposed method, we attempt to colourise images using Vision Transformer - Inception - Generative Adversarial Network (ViT-I-GAN), which has an Inception-v3 fusion embedding in the generator. For a stable and robust network, we have used Vision Transformer (ViT) as the discriminator. We trained the model on the Unsplash and the COCO dataset for demonstrating the improvement made by the Inception-v3 embedding. We have compared the results between ViT-GANs with and without Inception-v3 embedding.