[](LICENSE.md) > com.unity.perception is in active development. Its features and API are subject to significant change as development progresses. # Perception The Perception package provides a toolkit for generating large-scale datasets for perception-based machine learning training and validation. It is focused on a handful of camera-based use cases for now and will ultimately expand to other forms of sensors and machine learning tasks. **[Quick Installation Instructions](com.unity.perception/Documentation~/SetupSteps.md)** Get your local Perception workspace up and running quickly. Recommended for users with prior Unity experience. **[Perception Tutorial](com.unity.perception/Documentation~/Tutorial/TUTORIAL.md)** Detailed instructions covering all the important steps from installing Unity Editor, to creating your first Perception project, building a randomized Scene, and generating large-scale synthetic datasets by leveraging the power of Unity Simulation. No prior Unity experience required. **[Documentation](com.unity.perception/Documentation~/index.md)** Sample projects and in-depth documentation for various components of the SDK, including **Labeling, LabelConfig, Perception Camera, DatasetCapture, and Randomization**. ## Local development The repository includes two projects for local development in `TestProjects` folder, one set up for HDRP and the other for URP. ### Suggested IDE Setup For closest standards conformity and best experience overall, JetBrains Rider or Visual Studio w/ JetBrains Resharper are suggested. For optimal experience, perform the following additional steps: * To allow navigating to code in all packages included in your project, in your Unity Editor, navigate to `Edit -> Preferences... -> External Tools` and check `Generate all .csproj files.` ## License * [License](com.unity.perception/LICENSE.md) ## Citation If you find this package useful, consider citing it using: ``` @misc{com.unity.perception2020, title={Unity {P}erception Package}, author={{Unity Technologies}}, howpublished={\url{https://github.com/Unity-Technologies/com.unity.perception}}, year={2020} } ```