ZhumingLian/SHINE
[ICLR 2026] "Does FLUX Already Know How to Perform Physically Plausible Image Composition?" (Official Implementation)
This project helps graphic designers and digital artists seamlessly insert an object into an existing image, ensuring it looks physically realistic. You provide a background image and a reference image of the object to be inserted, along with a mask. The output is a high-quality, composited image where the new object matches the lighting and reflections of the scene, even in complex conditions. This is for visual content creators who need convincing image compositions without manual editing for shadows and reflections.
125 stars.
Use this if you need to combine images, especially when the inserted object must realistically interact with the scene's lighting, shadows, or reflective surfaces.
Not ideal if you primarily need simple cut-and-paste image editing or are working with images that do not require complex physical plausibility.
Stars
125
Forks
2
Language
Python
License
—
Category
Last pushed
Mar 12, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/ZhumingLian/SHINE"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
bytedance/InfiniteYou
🔥 [ICCV 2025 Highlight] InfiniteYou: Flexible Photo Recrafting While Preserving Your Identity
River-Zhang/ICEdit
[NeurIPS 2025] Image editing is worth a single LoRA! 0.1% training data for fantastic image...
muzishen/IMAGHarmony
🧩 IMAGHarmony 🧩: Controllable image editing with consistent object quantity and layout. A...
AMAP-ML/FE2E
[CVPR 2026] Beyond Generation: Advancing Image Editing Priors for Depth and Normal Estimation
TencentARC/BrushNet
[ECCV 2024] The official implementation of paper "BrushNet: A Plug-and-Play Image Inpainting...