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

In recent years, convolutional neural networks (CNNs) have been used for optical flow estimation with great success. Convolutional neural networks are multilayer’s structures, highly competent to estimate the complex, nonlinear transformation between input imagery and the output. At present, it is one of the top and dynamic research domains in image processing. Modern researchers of optical flow have combined numerous strategies and ideas from traditional methods with deep learning settings, consequently producing massive amount of work within few years. In this paper we present an analysis of deep learning methods for optical flow estimation. We highlight the important features, key components and discuss factors for benefits and limitations of deep learning schemes of optical flow estimation. We also compare the performance in terms of accuracy, size and training procedures of different models based on convolutional neural networks proposed by modern researchers.