# SynthDet: An end-to-end object detection pipeline using synthetic data [](LICENSE.md) ## Overview SynthDet is an open source project that demonstrates an end-to-end object detection pipeline using synthetic image data. The project includes all the code and assets for generating a synthetic dataset in Unity. Using recent [research](#Citation), SynthDet utilizes Unity Perception package to generate highly randomized images of 64 common grocery products (example: cereal boxes and candy) and export them along with appropriate labels and annotations (2D bounding boxes). The synthetic dataset generated can then be used to train a deep learning based object detection model. This project is geared towards ML practitioners and enthusiasts who are actively exploring synthetic data or just looking to get started. [GTC 2020: Synthetic Data: An efficient mechanism to train Perception Systems](https://developer.nvidia.com/gtc/2020/video/s22700) ## Contents * SynthDet - Unity Perception sample project * 3D Assets - High quality models of 64 commonly found grocery products * Unity Perception package * Unity Dataset Insights Python package ## Release & Documentation #### [Getting started with SynthDet](docs/Readme.md) Version|Release Date |Source -------|-------------|------ V0.1 |May 26, 2020|[source](https://github.com/Unity-Technologies/SynthDet) ## Citation SynthDet was inspired by the following research paper from Google Cloud AI: Hinterstoisser, S., Pauly, O., Heibel, H., Marek, M., & Bokeloh, M. (2019). [*An Annotation Saved is an Annotation Earned: Using Fully Synthetic Training for Object Instance Detection.* ](https://arxiv.org/pdf/1902.09967.pdf) ## Support For general questions or concerns please contact the Perception team at perception@unity3d.com For feedback, bugs, or other issues please file a github issue and the Perception team will investigate the issue as soon as possible. ## License * [License](LICENSE.md)