Traditional and modern strategies for optical flow: an investigation

Abstract Optical Flow Estimation is an essential component for many image processing techniques. This field of research in computer vision has seen amazing development recent years. In particular, the introduction Convolutional Neural Networks optical flow estimation shifted paradigm from classical traditional approach to deep learning side. At present, state art techniques are based on convolutional neural networks and almost all top performing methods incorporate architectures their schemes. paper presents a brief analysis highlights most developments this field. A comparison majority pertinent methodologies been undertaken resulting detailed establishment respective advantages disadvantages categories. An insight provided into significant factors that affect success or failure two classes estimation. establishing foremost existing inherent challenges with schemes, probable solutions have proposed indeed.