(Part 1) How to use Object Detection for Complex Image Classification Scenarios
The Visual Object Tagging Tool (VOTT) provides end-to-end support for generating data sets as well as validating object detection models from video and image assets.
The tool supports the following features:
• The ability to tag and annotate Image Directories or Stand-alone videos.
• Computer-assisted tagging and tracking of objects in videos using the Camshift tracking algorithm.
• Exporting tags and assets to Custom Vision Service CNTK , Tensorflow (PascalVOC) or YOLO format for training an object detection model.
• Use Active Learning with trained object detection models (locally or remotely) on new videos to generate stronger models.
Join Jerry Nixon and Sabrina Smai as they welcome one of the authors of the VOTT, Ari Bornstein, for part 1 of this 2 part series where they give us an overview of what the VOTT is and how you can use it in your applications.
Additional Resources:
https://github.com/Microsoft/VoTT
https://github.com/Microsoft/CNTK/wik...
http://github.com/aribornstein/cvwork...
https://github.com/fizyr/keras-retinanet
https://www.microsoft.com/developerbl...
Watch video (Part 1) How to use Object Detection for Complex Image Classification Scenarios online without registration, duration hours minute second in high quality. This video was added by user Microsoft DevRadio 14 August 2018, don't forget to share it with your friends and acquaintances, it has been viewed on our site 14,302 once and liked it 30 people.