Drift Robust Non-rigid Optical Flow Enhancement for Long Sequences
This addresses a fundamental issue in computer vision for applications like motion analysis, though it appears incremental as it enhances existing tracking methods.
The paper tackles the problem of drift in long-term dense tracking of non-rigid objects by introducing an optimization framework with an Anchor Patch constraint, resulting in significant error reduction across multiple optical flow algorithms on real-world benchmarks.
It is hard to densely track a nonrigid object in long term, which is a fundamental research issue in the computer vision community. This task often relies on estimating pairwise correspondences between images over time where the error is accumulated and leads to a drift issue. In this paper, we introduce a novel optimization framework with an Anchor Patch constraint. It is supposed to significantly reduce overall errors given long sequences containing non-rigidly deformable objects. Our framework can be applied to any dense tracking algorithm, e.g. optical flow. We demonstrate the success of our approach by showing significant error reduction on 6 popular optical flow algorithms applied to a range of real-world nonrigid benchmarks. We also provide quantitative analysis of our approach given synthetic occlusions and image noise.