Summary of Hierafashdiff: Hierarchical Fashion Design with Multi-stage Diffusion Models, by Zhifeng Xie et al.
HieraFashDiff: Hierarchical Fashion Design with Multi-stage Diffusion Models
by Zhifeng Xie, Hao Li, Huiming Ding, Mengtian Li, Xinhan Di, Ying Cao
First submitted to arxiv on: 15 Jan 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper proposes a novel hierarchical diffusion-based framework for fashion design, called HieraFashDiff. The model mimics the practical fashion design workflow by generating design proposals and refining them using low-level attributes. It supports both fashion design generation and fine-grained local editing in a single framework. To train the model, the authors contribute a new dataset of full-body fashion images annotated with hierarchical text descriptions. Evaluations show that HieraFashDiff outperforms prior approaches in generating fashion designs and edited results with higher fidelity and better prompt adherence. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Fashion designers use computers to create and edit outfits, but these tools don’t always produce realistic or desirable results. This paper presents a new way for AI to assist in the fashion design process. The model is designed to mimic how humans approach fashion design, generating ideas and refining them based on details like fabric and color. It can also make small changes to existing designs. The authors created a dataset of images with detailed descriptions to train their model, which outperforms previous approaches in creating realistic and on-trend outfits. |
Keywords
» Artificial intelligence » Diffusion » Prompt
