Abstract:A novel Transformer‐based medical image colourization method, which consists of content and style Transformer encoder to capture domain‐specific telematic information for long‐term dependency in colourization task is proposed. Then, a Transformer decoder is designed to translate content sequences and refer to style sequences, which achieves better fusion of content and style features. Additionally, content‐aware positional encoding scheme and style‐aware positional encoding scheme are employed for scale‐invariant visual generation tasks. Medical image colouring techniques enable to colourize grey‐scale medical images for assisting doctors in diagnosis. Benefiting from the non‐linear fitting ability of deep neural network, deep medical image colouring techniques have achieved remarkable results. However, existing methods are still facing content and structure feature leakage, unrealistic colouring and poor scale invariability. Thus, this paper, proposes a Transformer‐based medical image colouring algorithm with long‐term dependency to avoid feature leakage of coloured images. To be specific, this method employs two different Transformer encoders to generate and encode feature sequences for grey‐scale medical images and real human colour slice images, respectively. Then, a novel multi‐layer Transformer decoder is used to stylize grey‐scale map image features based on the real physical colour feature sequences. For colouring images at different scales, we implement content‐ aware positional encoding with scale invariance and propose style‐aware positional encoding strategy to take realistic and physical colour prior into account. Extensive experimental results indicate our method has achieved better colourization effects than recent state‐of‐the‐art medical image colourization methods.

Colorizing Biomedical Images Based On Color Transfer

Improved method of image color transfer.

Color Transfer Between Images with Auto Selecting Swatches

Uniform Color Transfer

L0 Gradient-Preserving Color Transfer

Transferring Color Between Three-Dimensional Objects

Deep Exemplar-Based Color Transfer for 3D Model

Pseudo-colorization of medical images based on two-stage transfer model

Reference Based Color Transfer for Medical Volume Rendering

A Color Transfer Method for Colorization of Grayscale Image Based on Region Histogram Statistics

Colorizing Grayscale CT Images of Human Lungs Using Deep Learning Methods

Selective color transfer with multi-source images

Color Retinal Image Enhancement Based on Luminosity and Contrast Adjustment

Color transfer via local binary patterns mapping

Fast Color Transfer from Multiple Images.

A Comprehensive Method for Example-Based Color Transfer with Holistic–Local Balancing and Unit-Wise Riemannian Information Gradient Acceleration

Digital Makeup from Internet Images

Homogenous Color Transfer Using Texture Retrieval and Matching.

Color restoration based on digital pathology image

A learning-based approach for automatic image and video colorization

TanrsColour: Transformer‐based medical image colourization with content and structure preservation