GitHub repository
An open-source SDK for logging, storing, querying, and visualizing multimodal, time-series data like images, point clouds, and tensors.
What is GitHub repository?
Rerun is an open-source SDK and platform for logging, storing, querying, and visualizing multimodal and multi-rate data, such as images, tensors, point clouds, and text. It solves the problem of debugging and understanding complex, time-evolving systems—like robots or simulations—by providing a visual and temporal debugger that goes beyond traditional text logging.
Developers and researchers in robotics, spatial AI, computer vision, simulation, and industrial processing who need to visualize and analyze rich, temporal sensor data and system state.
Developers choose Rerun for its unified approach to multimodal data logging and visualization, powerful time-aware querying, and ability to extract datasets from recordings, offering a more comprehensive alternative to pure visualization tools like RViz.
Overview
An open source SDK for logging, storing, querying, and visualizing multimodal and multi-rate data
Use Cases
Best For
- Debugging autonomous robots or drones with multiple sensor feeds
- Visualizing and analyzing 3D perception data like lidar scans and point clouds
- Recording and replaying test runs for robotics or simulation systems
- Creating training datasets from logged multimodal recordings
- Live visual debugging of computer vision pipelines
- Understanding temporal evolution of 2D/3D state in embodied AI
Not Ideal For
- Projects requiring real-time visualization of extremely dense point clouds (millions of points) without performance lag
- Teams prioritizing long-term API stability over new features, given the active development and breaking changes
- Applications with only simple, static data visualization needs where lightweight tools like matplotlib suffice
- Environments with very limited memory where storing historical data in RAM for the time buffer is not feasible
Pros & Cons
Pros
Supports diverse data types like images, tensors, point clouds, and text via Python, Rust, and C++ SDKs, enabling comprehensive data capture for complex systems such as robotics and simulation.
The Rerun Viewer acts as a visual debugger with scrubbing capabilities for temporal data streams, allowing users to pinpoint issues over time, as demonstrated in the vacuum robot debugging example.
Logs can be streamed to file or over the network and queried through a dataframe API, facilitating dataset creation and analysis from recordings, which is highlighted in the docs for extracting training data.
Supports real-time visualization and post-hoc analysis of recordings in formats like MCAP, making it versatile for both development and testing phases without requiring an account.
Cons
The viewer slows down with too many entities and struggles with multi-million point clouds, as admitted in the README's status section, which can hinder real-time debugging.
The project is in active development with breaking changes expected, disrupting existing integrations and requiring frequent updates, as warned in the README.
C++ and Rust SDKs require separate installation of the Rerun Viewer binary, unlike Python, adding setup overhead and potential dependency issues, such as needing nasm for optimal performance.
As a specialized tool, it may lack the extensive community support, plugins, and third-party integrations found in more established visualization or logging libraries.
Open Source Alternative To
GitHub repository is an open-source alternative to the following products:
Frequently Asked Questions
Related Projects
GitHub repositoryThe open source mesh processing system
plottersA rust drawing library for high quality data plotting for both WASM and native, statically and realtimely 🦀 📈🚀
Semantic Segmentation EditorWeb labeling tool for bitmap images and point clouds
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.