The paper discusses the computation of optical flow, which approximates two-dimensional image motion as either instantaneous velocities or discrete displacements. Optical flow can be used to estimate three-dimensional motion and surface structure, as well as for motion detection, object segmentation, and other tasks. The paper surveys various methods for computing optical flow, including differential, frequency-based, correlation-based, and multiple motion methods. It also discusses the challenges and issues in optical flow computation, such as the aperture problem, occlusion, and transparency. The paper concludes with a discussion of current research issues in optical flow computation. The paper also discusses the importance of optical flow in visual processing and its applications in various fields. The paper is organized into sections on motion and structure paradigms, optical flow, hierarchical processing, problems and issues, and scope and purpose. The paper also discusses the use of optical flow in motion and structure algorithms, as well as in motion compensation and stereo disparity measurement. The paper concludes with a discussion of the current state of optical flow research and its potential applications.The paper discusses the computation of optical flow, which approximates two-dimensional image motion as either instantaneous velocities or discrete displacements. Optical flow can be used to estimate three-dimensional motion and surface structure, as well as for motion detection, object segmentation, and other tasks. The paper surveys various methods for computing optical flow, including differential, frequency-based, correlation-based, and multiple motion methods. It also discusses the challenges and issues in optical flow computation, such as the aperture problem, occlusion, and transparency. The paper concludes with a discussion of current research issues in optical flow computation. The paper also discusses the importance of optical flow in visual processing and its applications in various fields. The paper is organized into sections on motion and structure paradigms, optical flow, hierarchical processing, problems and issues, and scope and purpose. The paper also discusses the use of optical flow in motion and structure algorithms, as well as in motion compensation and stereo disparity measurement. The paper concludes with a discussion of the current state of optical flow research and its potential applications.