In this version, I have consolidated all changes made to Version 1.0 and added a number of enhancements:
- Changed the architecture to RESNET50 to improve training accuracy
- Enhanced the model with a couple of booster conv2 layers to increase the power of the model to recognize small objects
- Added prediction code at the end of the notebook to test external images with and without NMS
- Added a model export section that creates a .pkl file. This file is read by the external image prediction section, which can be in a separate computer
I am working on Version 3.0 that will include:
- Enhanced data augmentation. Implementation of the paper “Learning Data Augmentation Strategies for Object Detection” by Barret Zoph, Elkin Cubuk, et. al.
- Mechanism to calculate Mean Average Precision mPA
You can find Version 2.0 in my GitHub at https://github.com/jav0927/course-v3/blob/master/SSD_Object_Detection_RS50_V2_0_Fixed.ipynb
As always, I welcome your comments and suggestions.