Hi HJ,
When starting to train AI, the choice between using videos or images depends on the specific task at hand. If the AI needs to understand temporal patterns or motion, training with videos might be more beneficial. On the other hand, if the task is more focused on static features, images could suffice.
For neural networks, a popular choice is Convolutional Neural Networks (CNNs) for image-related tasks due to their ability to extract spatial features effectively. You can implement CNNs in MATLAB using functions like convolution2dLayer, maxPooling2dLayer, and fullyConnectedLayer from the Deep Learning Toolbox.
If dealing with video data, Recurrent Neural Networks (RNNs) or 3D Convolutional Neural Networks (3D CNNs) could be suitable. MATLAB provides functions like lstmLayer for RNNs and 3dConvolutionLayer for 3D CNNs in the Deep Learning Toolbox.
Hope this will help achieve your goals.
