adobe-research/DiffusionHandles
Diffusion Handles is a training-free method that enables 3D-aware image edits using a pre-trained Diffusion Model.
This tool helps graphic designers and artists easily make 3D-aware edits to existing images using advanced diffusion models. You provide an image and a simple 3D transformation (like rotating or repositioning an object), and it outputs a new image with that object realistically altered in 3D space. It's ideal for anyone who wants to manipulate objects within images with a sense of depth without complex 3D software.
No commits in the last 6 months.
Use this if you need to realistically adjust the orientation or position of objects within a 2D image as if they were 3D models, without needing to recreate them from scratch.
Not ideal if you're looking to generate entirely new images from text prompts or perform simple 2D image adjustments like cropping or color correction.
Stars
41
Forks
6
Language
Python
License
—
Category
Last pushed
Feb 10, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/adobe-research/DiffusionHandles"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
OPTML-Group/Unlearn-Saliency
[ICLR24 (Spotlight)] "SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in...
Shilin-LU/VINE
[ICLR 2025] "Robust Watermarking Using Generative Priors Against Image Editing: From...
WindVChen/DiffAttack
An unrestricted attack based on diffusion models that can achieve both good transferability and...
koninik/DiffusionPen
Official PyTorch Implementation of "DiffusionPen: Towards Controlling the Style of Handwritten...
Wuyxin/DISC
(ICML 2023) Discover and Cure: Concept-aware Mitigation of Spurious Correlation