This project is about gender and age detection. Nowadays, age and gender classification are being used in many applications like human-computer interaction, biometrics, video surveillance, forensic, etc. The main objective of this project is to compare the image with the reference images stored as templates in the database and to check how approximately it gives the accuracy by detecting the age and gender. The age and gender are detected by giving the picture and video file as input. It is used on the basis of deep learning and successfully applied to the image classification area. Gender and age detection from images of human faces and also the video is an essential part of our method for anomalous human behavior.