Multimodal data streams such as images, text, sounds, etc. can be visualized. The goal is to enable users to more intuitively and effectively view, analyze and understand these data streams from different sources through a series of tools and functions. Doing so makes it easier to capture data structures and patterns for more accurate analysis and decisions.
Technical details and working principle:
Data logging and visualization: The Rerun SDK provides a set of APIs that allow users to record data from their programs and then visualize it in the Rerun Viewer.
Type-centric logging APIs: The new APIs focus more on data types, allowing users to more directly control the underlying ECS and more easily record their own objects.
Upgrade content for Rerun 0.9 version:
1. API changes and underlying control
An important feature of Rerun 0.9 is the refactoring of its API. The new API design is more flexible and allows users to more directly control the underlying Entity Component System (ECS). This is a hierarchical and time-sensitive system for managing multimodal data flows.
- From Python to C++: Cross-language consistency
The new version also includes the foundation of the C++ SDK, unifying APIs for Rust, Python, C++ and other languages. This means that Rerun will be able to better serve production systems developed using C++, especially in the fields of robotics, computer vision and games. - performance and ease of use
In addition to performance improvements, Rerun 0.9 also adds support for Markdown and a new viewer entry experience, which will make it easier for users to get started.
Details:https://rerun.io/blog/release-0.9
