YAML Metadata
Warning:
empty or missing yaml metadata in repo card
(https://huggingface.co/docs/hub/model-cards#model-card-metadata)
Welcome to the official storage cite for AIpparel: A Multimodal Foundation Model for Digital Garments (CVPR 2025 Highlight).
This repository stores the data and the model weights of AIpparel.
For the project website, visit here.
For the official code repository, visit here.
GarmentCodeData-Multimodal (GCD-MM) Dataset.
Download the gcd-mm.parquet file to obtain different annotations for GarmentCodeData (GCD) used to train AIpparel. Specifically, the file contains the following columns representing different annotations for GCD:
- idx: unique id of the sewing pattern, used to associate the annotations with sewing patterns in GCD.
- descriptive_text: caption that describe the sewing pattern in detail.
- speculative_text: caption that speculates a situation suitable for the sewing pattern.
- keywords: short, keywords describing the design of sewing pattern.
- title: title of the sewing pattern.
- original_pattern: the sewing pattern JSON file.
- edited_pattern: the edited sewing pattern JSON file.
- edit_instruction: the editing instruction that converts the original pattern to the edited pattern.
AIpparel Pre-trained Model Weights
Download aipparel_pretrained.pth to obtain the pretrained weights to AIpparel. Note that this checkpoint do not contain weights for the vision encoder and mm-projector of LlAVA 1.5. Please download them separately.
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
๐
Ask for provider support
HF Inference deployability: The model has no library tag.