In today’s world, cameras are used in various field like surveillance, medical, defence, etc. Most of these cameras are either visible image or infrared image cameras. For improving in this field there has been many attempts to create a better definition and better quality images and companies spend billions of dollars in research and development. We also got into the direction of multi modal image fusion. In this we create a FusionNet, a CNN which is used to integrate the information from both infrared and visible images to create a much more informative and have more details which individual cameras are not able to do individually. In this, through a advanced neural network we are going to generate a fused output which will be much more clear perception and will offer a more information that provided before fusion of images. It is quite different from the traditional fusion of images as it used deep learning techniques which tends to create a much better and tends to improve over time. This research tends to broaden our horizon in field of image fusion using deep learning and in future it will help us in increasing efficiency and accuracy of images in various fields.