xrenaa/CS-DisMo
[ICCVW 2021] Rethinking Content and Style: Exploring Bias for Unsupervised Disentanglement
This project helps researchers and practitioners in computer vision to separate visual elements like identity and expression, or object shape and texture, from images without needing labeled data. It takes in a collection of images and allows you to independently manipulate these 'content' and 'style' aspects. This tool is ideal for scientists working on advanced image generation, manipulation, or analysis tasks.
No commits in the last 6 months.
Use this if you need to disentangle the content and style of images in a completely unsupervised manner for tasks like face manipulation, object attribute editing, or generating variations.
Not ideal if you require supervised disentanglement with precise control over specific, predefined attributes, or if your primary goal is not image synthesis or manipulation.
Stars
20
Forks
1
Language
—
License
—
Category
Last pushed
Aug 18, 2021
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/xrenaa/CS-DisMo"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
WinfredGe/T2S
[IJCAI 2025] Official implementation of "T2S: High-resolution Time Series Generation with...
monniert/dti-sprites
(ICCV 2021) Code for "Unsupervised Layered Image Decomposition into Object Prototypes" paper
Mukosame/AODA
Official implementation of "Adversarial Open Domain Adaptation for Sketch-to-Photo...
shashankvkt/AlignMixup_CVPR22
Official Implementation of AlignMixup - CVPR 2022
JinXins/Adversarial-AutoMixup
Official PyTorch(MMCV) implementation of “Adversarial AutoMixup” (ICLR 2024 spotlight)