HAIFIT: Human-to-AI Fashion Image Translation

Jianan Jiang,Xinglin Li,Weiren Yu,Di Wu

2024-08-14

Abstract:In the realm of fashion design, sketches serve as the canvas for expressing an artist's distinctive drawing style and creative vision, capturing intricate details like stroke variations and texture nuances. The advent of sketch-to-image cross-modal translation technology has notably aided designers. However, existing methods often compromise these sketch details during image generation, resulting in images that deviate from the designer's intended concept. This limitation hampers the ability to offer designers a precise preview of the final output. To overcome this challenge, we introduce HAIFIT, a novel approach that transforms sketches into high-fidelity, lifelike clothing images by integrating multi-scale features and capturing extensive feature map dependencies from diverse perspectives. Through extensive qualitative and quantitative evaluations conducted on our self-collected dataset, our method demonstrates superior performance compared to existing methods in generating photorealistic clothing images. Our method excels in preserving the distinctive style and intricate details essential for fashion design applications. In addition, our method also has obvious advantages in model training and inference speed, contributing to reducing designers' time costs and improving design efficiency.

Computer Vision and Pattern Recognition

What problem does this paper attempt to address?

### The Problem This Paper Attempts to Solve

The main objective of this paper is to address the issue of detail loss in the cross-modal translation process from sketch to image in the field of fashion design. Specifically:

1. **Sketch Detail Preservation**:
- Existing methods for sketch-to-image generation often lose details from the sketch during the generation process, resulting in images that do not match the designer's original concept.
- The paper proposes a new method called HAIFIT (Human-to-AI Fashion Image Translation), which aims to preserve sketch details through multi-scale feature fusion and extensive feature map dependencies.

2. **High-Fidelity Image Generation**:
- By integrating multi-scale features and capturing feature map dependencies from different perspectives, HAIFIT can transform sketches into high-fidelity, realistic fashion images.
- Experimental results show that HAIFIT outperforms existing methods in generating realistic fashion images and excels in maintaining unique styles and details.

3. **Model Training and Inference Speed**:
- In addition to improving the quality of image generation, HAIFIT also has significant advantages in model training and inference speed, helping to reduce designers' time costs and improve design efficiency.

In summary, this paper aims to solve the problem of detail loss in the sketch-to-image conversion process through the HAIFIT method and to enhance the fidelity and speed of the generated images.