federicoarenasl/sdg-engine
A simple data generation engine for computer vision, compatible with 🤗 datasets.
This tool helps computer vision practitioners create synthetic image datasets for training models. You start with a scene designed in Blender, specifying which objects to include. It then renders numerous variations of this scene, outputting images with corresponding annotations (like bounding boxes) that can be directly used for tasks like object detection or pushed to the Hugging Face dataset hub.
Use this if you need large, diverse datasets of images with precise annotations for computer vision tasks, but real-world data collection and manual labeling are too costly or time-consuming.
Not ideal if your computer vision task requires highly realistic imagery that cannot be credibly replicated through 3D rendering, or if you don't have access to Blender or similar 3D modeling skills.
Stars
76
Forks
14
Language
Jupyter Notebook
License
—
Category
Last pushed
Dec 26, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/federicoarenasl/sdg-engine"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
meta-llama/synthetic-data-kit
Tool for generating high quality Synthetic datasets
Diyago/Tabular-data-generation
We well know GANs for success in the realistic image generation. However, they can be applied in...
Data-Centric-AI-Community/ydata-synthetic
Synthetic data generators for tabular and time-series data
tdspora/syngen
Open-source version of the TDspora synthetic data generation algorithm.
vanderschaarlab/synthcity
A library for generating and evaluating synthetic tabular data for privacy, fairness and data...