# Iterative Residual Refinement for Joint Optical Flow and Occlusion Estimation

## TL;DR

This paper introduces an iterative residual refinement scheme for optical flow and occlusion estimation that reduces parameters and enhances accuracy by integrating classical energy minimization ideas with residual networks.

## Contribution

The proposed IRR scheme with weight sharing is a novel approach that improves accuracy and reduces parameters, and it effectively incorporates occlusion prediction and bi-directional flow.

## Key findings

- Achieves state-of-the-art results on standard datasets.
- Reduces model parameters while maintaining or improving accuracy.
- Enhances optical flow and occlusion estimation through integrated scheme.

## Abstract

Deep learning approaches to optical flow estimation have seen rapid progress over the recent years. One common trait of many networks is that they refine an initial flow estimate either through multiple stages or across the levels of a coarse-to-fine representation. While leading to more accurate results, the downside of this is an increased number of parameters. Taking inspiration from both classical energy minimization approaches as well as residual networks, we propose an iterative residual refinement (IRR) scheme based on weight sharing that can be combined with several backbone networks. It reduces the number of parameters, improves the accuracy, or even achieves both. Moreover, we show that integrating occlusion prediction and bi-directional flow estimation into our IRR scheme can further boost the accuracy. Our full network achieves state-of-the-art results for both optical flow and occlusion estimation across several standard datasets.