Normalized Convolution Upsampling for Refined Optical Flow Estimation

TL;DR

This paper introduces NCUP, a normalized convolution-based upsampling method that improves full-resolution optical flow estimation during training, outperforming existing methods in accuracy and efficiency.

Contribution

The paper proposes a novel normalized convolutional upsampler for optical flow CNNs, enabling efficient, high-quality full-resolution flow prediction during training.

Findings

Outperforms existing joint upsampling methods on FlyingChairs dataset.

Achieves state-of-the-art results on Sintel with fewer parameters.

Demonstrates better generalization across datasets.

Abstract

Optical flow is a regression task where convolutional neural networks (CNNs) have led to major breakthroughs. However, this comes at major computational demands due to the use of cost-volumes and pyramidal representations. This was mitigated by producing flow predictions at quarter the resolution, which are upsampled using bilinear interpolation during test time. Consequently, fine details are usually lost and post-processing is needed to restore them. We propose the Normalized Convolution UPsampler (NCUP), an efficient joint upsampling approach to produce the full-resolution flow during the training of optical flow CNNs. Our proposed approach formulates the upsampling task as a sparse problem and employs the normalized convolutional neural networks to solve it. We evaluate our upsampler against existing joint upsampling approaches when trained end-to-end with a a coarse-to-fine optical flow CNN (PWCNet) and we show that it outperforms all other approaches on the FlyingChairs dataset while having at least one order fewer parameters. Moreover, we test our upsampler with a recurrent optical flow CNN (RAFT) and we achieve state-of-the-art results on Sintel benchmark with ~6% error reduction, and on-par on the KITTI dataset, while having 7.5% fewer parameters (see Figure 1). Finally, our upsampler shows better generalization capabilities than RAFT when trained and evaluated on different datasets.