shadow2496/VITON-HD
Official PyTorch implementation of "VITON-HD: High-Resolution Virtual Try-On via Misalignment-Aware Normalization" (CVPR 2021)
Introduces ALIAS (ALIgnment-Aware Segment) normalization to handle misaligned regions between warped garments and body parts, enabling high-resolution 1024×768 synthesis compared to prior 256×192 limitations. The approach combines coarse clothing fitting with a specialized generator that preserves texture detail and minimizes artifacts in misaligned boundary areas. Built on PyTorch with a curated dataset of 11,647 frontal woman-clothing pairs for research purposes.
1,128 stars. No commits in the last 6 months.
Stars
1,128
Forks
247
Language
Python
License
—
Category
Last pushed
Apr 27, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/shadow2496/VITON-HD"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related models
theFoxofSky/ddfnet
The official implementation of the CVPR2021 paper: Decoupled Dynamic Filter Networks
ChristophReich1996/Involution
PyTorch reimplementation of the paper "Involution: Inverting the Inherence of Convolution for...
maryam089/SDViT
Official repository for "Self-Distilled Vision Transformer for Domain Generalization" (ACCV-2022 ORAL)
thomasverelst/dynconv
Code for Dynamic Convolutions: Exploiting Spatial Sparsity for Faster Inference (CVPR2020)
NikosEfth/freedom
Official PyTorch implementation of the WACV 2025 Oral paper "Composed Image Retrieval for...