Over four decades, the majority addresses the problem of optical flow estimation using variational methods. With the advance of machine learning, some recent works have attempted to address the problem using convolutional neural network (CNN) and have showed promising results. FlowNet2 [1], the state-of-the-art CNN, requires over 160M parameters to achieve accurate flow estimation. Our LiteFlowNet2 outperforms FlowNet2 on Sintel and KITTI benchmarks, while being 25.3 times smaller in the model size and 3.1 times faster in the running speed. LiteFlowNet2 is built on the foundation laid by conventional methods and resembles the corresponding roles as data fidelity and regularization in variational methods. We compute optical flow in a spatial-pyramid formulation as SPyNet [2] but through a novel lightweight cascaded flow inference. It provides high flow estimation accuracy through early correction with seamless incorporation of descriptor matching. Flow regularization is used to ameliorate the issue of outliers and vague flow boundaries through feature-driven local convolutions. Our network also owns an effective structure for pyramidal feature extraction and embraces feature warping rather than image warping as practiced in FlowNet2 and SPyNet. Comparing to LiteFlowNet [3], LiteFlowNet2 improves the optical flow accuracy on Sintel Clean by 23.3%, Sintel Final by 12.8%, KITTI 2012 by 19.6%, and KITTI 2015 by 18.8%, while being 2.2 times faster. Our network protocol and trained models are made publicly available on https://github.com/twhui/LiteFlowNet2. Fig. 1: Examples demonstrate the effectiveness of the proposed components in LiteFlowNet for i) feature warping, ii) cascaded flow inference, and iii) flow regularization. Enabled components are indicated with bold black fonts.image. The model, as a result, comprises over 160M parameters and has a slow runtime, which could be formidable in many applications. Another work, SPyNet [2], uses a spatial pyramid network with only 1.2M parameters by adopting image warping in each pyramid level. Nonetheless, its performance can only match that of FlowNet but not FlowNet2.FlowNet2 [1] and SPyNet [2] showed the potential of solving the optical flow problem by using CNNs. Our earlier work, LiteFlowNet [13], is inspired by their successes, but we further drill down some of the key elements of solving the flow problem by adopting data fidelity and regularization in classical variational methods to CNN more closely. In this work, we provide more details on the correspondences between conventional methods and arXiv:1903.07414v2 [cs.CV]