This object detection project takes input images and tries to identify the objects present in them. This particular model has been trained for ducky images.
To execute this project, make sure your directory inside tensorflow-gpu is: training_demo/ -annotations/ -exported-models/ -images/ -test/ -train/ -models/ -pre-trained-models/
Two of the folders have to be created manually, and are not provided above, that are the pre-trained-models and exported-models folders.