I have a school project where I need to build an application for detecting plastic bottles and soda cans. I found a bunch of models that detect bottles but they don't detect cans. I trained TFLite model on Google Cloud, the detection model works perfectly on the cloud but when I run it on the raspberry pi it detects only cans(both bottles and cans are detected as bottles). I copied the code from: https://github.com/EdjeElectronics/TensorFlow-Lite-Object-Detection-on-Android-and-Raspberry-Pi/blob/master/TFLite_detection_image.py
Can someone please help with this problem? Is there a problem with the code because when I run it the image is shown extra large and cannot be displayed on the whole screen? Is there another model that detects soda cans? Thank you in advance!