Object Detection: Difference between revisions

From Open Source Ecology
Jump to navigation Jump to search
(Added some more information)
(Added some more links under the "Useful Links" section)
 
(3 intermediate revisions by one other user not shown)
Line 2: Line 2:
*The usage of [[Machine Vision]] to recognise objects
*The usage of [[Machine Vision]] to recognise objects
*Typically uses standard cameras, but some systems use other methods in conjunction such as laser dot grids (like the xbox kinect etc)
*Typically uses standard cameras, but some systems use other methods in conjunction such as laser dot grids (like the xbox kinect etc)
=Example=
*Detecting a toy airplane with Tensorflow - [[https://towardsdatascience.com/building-a-toy-detector-with-tensorflow-object-detection-api-63c0fdf2ac95]]
*From there, how to use with AR? Need integration - AR SDKs.
*AR Software Development Kits - https://opensource.com/article/18/6/open-source-augmented-reality-sdks
**ARToolKit - uses Python


=Tensorflow - OSS by Google=
=Tensorflow - OSS by Google=
Line 21: Line 27:


=Useful Links=
=Useful Links=
*
*[https://www.dfrobot.com/product-1922.html?search=huskylens&description=true A Dedicated Device with an Arduino camera module level Camera, and an AI Dedicated Chip + IO ]

Latest revision as of 22:18, 10 April 2020

Basics

  • The usage of Machine Vision to recognise objects
  • Typically uses standard cameras, but some systems use other methods in conjunction such as laser dot grids (like the xbox kinect etc)

Example

Tensorflow - OSS by Google

  • An open source object detection and recognition/naming software
  • Has two versions, the more intense Tensorflow , for pc's and render farms, and Tensorflow Lite

https://towardsdatascience.com/is-google-tensorflow-object-detection-api-the-easiest-way-to-implement-image-recognition-a8bd1f500ea0

COCO

COCO dataset - common objects in context.

Luminoth - OSS

See Also

Useful Links