README.md

metadata

title: Fashion Category Classifier
emoji: ⚡
colorFrom: pink
colorTo: blue
sdk: gradio
sdk_version: 5.45.0
app_file: app.py
pinned: false

👗 Fashion Category Classifier

A Hugging Face Space that uses specialized AI models to accurately classify clothing items from images. It first isolates the garment using segmentation to avoid background noise, then performs fine-grained classification using a model trained on fashion data.

How it Works

1. Segmentation: The image is processed by a SegFormer model fine-tuned on clothing segmentation (mattmdjaga/segformer_b2_clothes). This model identifies and creates a precise mask around the main clothing item.
2. Isolation: The image is cropped to this mask, removing any distracting background.
3. Classification: The isolated garment is analyzed by a FashionCLIP model (edwardjross/fashion-clip), which compares it to a detailed list of fashion categories to find the best match.

This two-step process significantly reduces "hallucinations" and improves accuracy.