Deep Learning techniques are a very effective tool in face detection. In this book, deep learning techniques have been used to identify dual faces nothing but by detecting dual shot faces. As the data is emerging day by day with high dimensionality, recognizing dual faces is a major problem. So wasting time on identifying images is like fiddling around. In order to save time and get absolute accuracy, we have implemented a fast preprocessing technique known as convolution neural network along with feature extraction which is used to know relevant features to detect faces/images. By performing this robust method, our intention is to detect dual images in an efficient way. The experiment is performed on extensive good face detecting benchmark datasets like FDDB (Face detection dataset and benchmark), wider face datasets. CNN with FE demonstrates the results with superiority and the accuracy was in-depth analyzed by CNN classifier.