This simple classifier can identify if an image is take indoors or outdoors. This is a model developed to test a hypothesis that such a classification can be made using 3 things:
This is definitely sub-optimal. However, I want to use this to test my hypothesis. So this is good enough as a first iteration.
For this use-case, I am using the pre-trained model
MobileNetV2. In my use case, I have just 50 images each for the 2 labels that I will be training. Due to the small data set, a smaller model like MobileNet would be better suited.