roeiherz/CanonicalSg2Im
Code for "Learning Canonical Representations for Scene Graph to Image Generation", Herzig & Bar et al., ECCV2020
This project helps researchers and developers create realistic images from structured descriptions of scenes. You input a 'scene graph' – a diagram showing objects and their relationships (e.g., 'a cat sitting on a mat') – and the system generates a corresponding image. It's particularly useful for those working on computer vision tasks like synthetic data generation or visual storytelling.
No commits in the last 6 months.
Use this if you need to generate high-quality, complex images from textual or graphical scene descriptions, especially when dealing with many objects or detailed spatial relationships.
Not ideal if you're looking for a user-friendly, out-of-the-box application for general image editing or generation without prior technical setup.
Stars
30
Forks
3
Language
Python
License
MIT
Category
Last pushed
Nov 22, 2022
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/roeiherz/CanonicalSg2Im"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
Vchitect/VBench
[CVPR2024 Highlight] VBench - We Evaluate Video Generation
VectorSpaceLab/OmniGen
OmniGen: Unified Image Generation. https://arxiv.org/pdf/2409.11340
EndlessSora/focal-frequency-loss
[ICCV 2021] Focal Frequency Loss for Image Reconstruction and Synthesis
JIA-Lab-research/DreamOmni2
This project is the official implementation of 'DreamOmni2: Multimodal Instruction-based Editing...
SkyworkAI/UniPic
Open-source SOTA multi-image editing model