Skip to content

Open-source cross-platform data structures, IO, and algorithms for expressing, storing, and processing spatial data.

License

Notifications You must be signed in to change notification settings

StandardCyborg/scsdk

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

scsdk

Open-source cross-platform data structures, serialization, and algorithms for expressing, storing, and processing spatial data.

SCSDK is a cross-platform package that provides fast, flexible, and expressive data structures designed to make working with spatial data both easy and intuitive. It aims to be the fundamental high-level building block for solving computer vision and perception problems. SCSDK also provides non-library dependent data structures through our protobuf definitions. Lastly SCSDK aims to have a robust catalog of core processing algorithms for spatial data.

The core SDK is written in highly performant C++ with minimal dependencies.

Status: This library is early in its development but has a solid well tested foundation.

Supports:

Example

For example, you can capture a stream of images on iOS, open them in the browser to view, and then easily fetch them via python and load them into a pandas dataframe. You can do this across not just trivial data like images, but also meshes, point clouds, accelerometer data and more. more example scripts

Motivations and Goals

  • Have a consistent set of high-performant data structures and fast IO for working with spatial data across platforms
  • Have very few dependencies and reliable build instructions for easy breezy usage
  • Easy interoperability with other data structures: for example tfrecords, COCO, pascal VOC, etc.
  • Easy interoperability with other geometry processing libraries: for example libigl
  • Open format and intrinsically cross-platform data structures that do not require the library using protobuf messages

If you need an archive compressed format for at-rest storage or moving things around, we suggest checking out protobag

Use cases:

  • Static scenes of data from a single sensor
  • Static scenes of data from multiple inputs/sensors
  • Dynamic time-series data from one or more sensors

Data types

  • RGB-D: Images and depth frames
  • Primitives: Points, polylines, planes, bounding boxes,
  • Cameras
  • 3D assets: point clouds and meshes
  • Scene graph

Documentation

While documentation is early compared to our goals for the project, you can find example code as well as header comments throughout the project.

Getting help

We intend to use Github Discussions, Gitter and other tools, but please just use Issues for now.

Contributions

Work on scsdk started at Standard Cyborg in 2018 and has been under active development since then. We welcome contributions though we don't have a contribution guide yet.

License

Apache License 2.0

About

Open-source cross-platform data structures, IO, and algorithms for expressing, storing, and processing spatial data.

Topics

Resources

License

Stars

Watchers

Forks

Languages