classification - Need Advice on Image Classfication using Autoencoders

Question

(Excuse me for errors for it is my first time posting)

For my project I am aiming to use autoencoders along with an another model such as MLP for classification of pedestrians and non pedestrians. My dataset is 128x64 boxes that either just human, or background such as cars etc. I am having trouble on deciding on what sort of Autoencoder model should I use and how should I choose the parameters. My desire is to get an output of either [1;0] or [0;1] at the end of the classifier showing whether if the image is of a pedestrian or not.

If anyone can make a suggestion of a related work, or can give me advice on the autoencoder, I will be very happy.

question from:https://stackoverflow.com/questions/65938010/need-advice-on-image-classfication-using-autoencoders

Answer 1

Waitting for answers