Planar Object Tracking via Weighted Optical Flow

TL;DR

WOFT is a new planar object tracking method that uses weighted optical flow to estimate homography, achieving robustness and state-of-the-art results without relying on traditional outlier rejection techniques.

Contribution

The paper introduces a differentiable weighted optical flow module that improves robustness in planar object tracking by automatically handling outliers, eliminating the need for RANSAC.

Findings

Achieves state-of-the-art performance on POT-210 and POIC benchmarks.

Automatically assigns zero weights to outliers, enhancing robustness.

Performs consistently across diverse tracking scenarios.

Abstract

We propose WOFT -- a novel method for planar object tracking that estimates a full 8 degrees-of-freedom pose, i.e. the homography w.r.t. a reference view. The method uses a novel module that leverages dense optical flow and assigns a weight to each optical flow correspondence, estimating a homography by weighted least squares in a fully differentiable manner. The trained module assigns zero weights to incorrect correspondences (outliers) in most cases, making the method robust and eliminating the need of the typically used non-differentiable robust estimators like RANSAC. The proposed weighted optical flow tracker (WOFT) achieves state-of-the-art performance on two benchmarks, POT-210 and POIC, tracking consistently well across a wide range of scenarios.