droidcam | GNU/Linux client for DroidCam

Here is the complete demo working code. Similar to you, I will be classifying 2 classes with softmax and use the sparse_categorical_crossentropy loss function. There is some issue in your model definition, so I will write my own, don't worry very simple and alike.

To make it useful, I will answer end-to-end, hopefully, new visitors may find it useful too. I will be classifying whether a digit even or odd - a binary classification. And lastly, we will use a raw sample image and preprocess it and try to find a class activation map to see where our model makes its attention. So, here is the content:

• Prepare 2 classes data set
• Build 2 class classifier and train it
• Find Grad-CAM

We will use MNIST and modify it for binary classes - even number or odd number.

The V4L2 device is a constant frame rate writer, so ffmpeg will duplicate frames to keep a constant frame rate if source frame rate is lower. In the absence of a user-set rate, ffmpeg sets it to the input's frame rate. For capture devices which don't signal a framerate explicitly, ffmpeg infers the rate by looking for the smallest interval between frames.

When ffmpeg detects 1000k tbr, that's when you get mass duplication. Either set a sane number -r 15 or disable frame duplication, -vsync vfr.

You can't change pixel formats when using -c:v copy. Change to -c:v rawvideo.