Abstract:Dunhuang murals are a collection of Chinese style and national style, forming an autonomous Chinese-style Buddhist art. It has a very great historical and cultural value and an important research value. Among them, the lines of Dunhuang murals are extremely general and expressive. It reflects the character's distinct character and intricate inner emotions. Hence, the outline drawing of the murals is of great importance in search of the Dunhuang culture. The generation of contours of Dunhuang murals belongs to the detection of image contours, which is an important branch of computer vision, aimed at extract salient contour information from the images. Although convolution-based deep learning networks have achieved good results in extracting image contours by exploring the contextual and semantic features of images. However, as the receptive field broadens, some detailed local information is lost. As a result, it is impossible for them to generate reasonable outline drawings of murals. In this paper, we propose a novel edge detector based on self-attention combined with convolution to generate line drawings of Dunhuang murals. Compared with existing edge detection methods, firstly, a new residual self-attention and convolution mixed module(Ramix) is proposed to merge local and global features into feature maps. Secondly, a novel densely connected backbone extraction network is designed to efficiently propagate information rich in edge features from shallow layers into deep layers. It is apparent from the experimental results that the method of this paper can generate richer and sharper edge maps on several standard edge detection datasets. In addition, tests on the Dunhuang mural dataset show that our method can achieve very competitive performance.

Mural Sketch Generation via Style-aware Convolutional Neural Network.

Sketch-R2CNN: an Attentive Network for Vector Sketch Recognition

<i>Sketch-R2CNN</i>: An RNN-Rasterization-CNN Architecture for Vector Sketch Recognition

Sketchformer++: A Hierarchical Transformer Architecture for Vector Sketch Representation

Intelligent Computer Technology-Driven Mural Pattern Recognition Method

SmartPaint: a Co-Creative Drawing System Based on Generative Adversarial Networks

Deep neural networks for Chinese traditional landscape painting creation

Convolutional Neural Network-Based Chinese Ink-Painting Artistic Style Extraction

Dunhuang murals contour generation network based on convolution and self-attention fusion

Color Hint-guided Ink Wash Painting Colorization with Ink Style Prediction Mechanism

Style Transfer for Anime Sketches with Enhanced Residual U-net and Auxiliary Classifier GAN

A New Method for Extracting Refined Sketches of Ancient Murals

Content-Conditioned Generation of Stylized Free hand Sketches

Sketch-to-Art: Synthesizing Stylized Art Images From Sketches

Art Style Transfer of Oil Painting Based on Parallel Convolutional Neural Network

SketchGAN: Joint Sketch Completion and Recognition with Generative Adversarial Network.

Sgrgan: sketch-guided restoration for traditional Chinese landscape paintings

Research on Image Generation and Style Transfer Algorithm Based on Deep Learning

Picture that Sketch: Photorealistic Image Generation from Abstract Sketches

Two-stage mural image restoration using an edge-constrained attention mechanism

GGD-GAN: Gradient-Guided Dual-Branch Adversarial Networks for Relic Sketch Generation