sascha-kirch/rgb-d-fusion
Official implementation of the paper "RGB-D-Fusion: Image Conditioned Depth Diffusion of Humanoid Subjects"
This project helps computer vision practitioners generate detailed 3D depth maps from standard RGB images of people or enhance existing low-resolution depth data. You provide a color image (RGB) and, optionally, a lower-resolution depth map. The system then outputs a high-resolution, dense depth map, showing precise distances to every point in the scene, especially for human subjects. This is ideal for researchers, 3D artists, and developers working with human pose estimation, virtual try-on, or scene reconstruction.
Use this if you need to create high-quality 3D depth information from 2D images of people or improve the resolution of existing 3D depth scans.
Not ideal if your primary goal is general object depth estimation, as this tool is specifically optimized for humanoid subjects and requires a developer to set up a Docker environment.
Stars
9
Forks
3
Language
Python
License
Apache-2.0
Category
Last pushed
Oct 29, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/sascha-kirch/rgb-d-fusion"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
PRIS-CV/DemoFusion
Let us democratise high-resolution generation! (CVPR 2024)
mit-han-lab/distrifuser
[CVPR 2024 Highlight] DistriFusion: Distributed Parallel Inference for High-Resolution Diffusion Models
Tencent-Hunyuan/HunyuanPortrait
[CVPR-2025] The official code of HunyuanPortrait: Implicit Condition Control for Enhanced...
giuvecchio/matfuse
MatFuse: Controllable Material Generation with Diffusion Models (CVPR2024)
Shilin-LU/TF-ICON
[ICCV 2023] "TF-ICON: Diffusion-Based Training-Free Cross-Domain Image Composition" (Official...