Optical Flow Estimation using a Spatial Pyramid Network

Abstract

We learn to compute optical flow by combining a classical spatial-pyramid formulation with deep learning. This estimates large motions in a coarse-to-fine approach by warping one image of a pair at each pyramid level by the current flow estimate and computing an update to the flow. Instead of the standard minimization of an objective function at each pyramid level, we train one deep network per level to compute the flow update. Unlike the recent FlowNet approach, the networks do not need to deal with large motions; these are dealt with by the pyramid. This has several advantages. First, our Spatial Pyramid Network (SPyNet) is much simpler and 96% smaller than FlowNet in terms of model parameters. This makes it more efficient and appropriate for embedded applications. Second, since the flow at each pyramid level is small (< 1 pixel), a convolutional approach applied to pairs of warped images is appropriate. Third, unlike FlowNet, the learned convolution filters appear similar to classical spatio-temporal filters, giving insight into the method and how to improve it. Our results are more accurate than FlowNet on most standard benchmarks, suggesting a new direction of combining classical flow methods with deep learning.

10 pages

References in corpus (6)

• Adam: A Method for Stochastic Optimization
• Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs
• Deep Generative Image Models using a Laplacian Pyramid of Adversarial Networks
• Understanding Deep Convolutional Networks
• EpicFlow: Edge-Preserving Interpolation of Correspondences for Optical Flow
• Back to Basics: Unsupervised Learning of Optical Flow via Brightness Constancy and Motion Smoothness

Cited by in corpus (12)

• Beyond Skip Connections: Top-Down Modulation for Object Detection
• Flow-Guided Feature Aggregation for Video Object Detection
• EDVR: Video Restoration with Enhanced Deformable Convolutional Networks
• Coherent Online Video Style Transfer
• Guided Optical Flow Learning
• Hierarchical Surface Prediction for 3D Object Reconstruction
• Learning Human Optical Flow
• Dual Motion GAN for Future-Flow Embedded Video Prediction
• Dynamic Filtering with Large Sampling Field for ConvNets
• Deep Optical Flow Estimation Via Multi-Scale Correspondence Structure Learning
• Dynamic Video Segmentation Network
• Depth Assisted Full Resolution Network for Single Image-based View Synthesis