Fast Piecewise-Affine Motion Estimation Without Segmentation
This work addresses motion estimation in computer vision, offering an incremental improvement by eliminating segmentation steps and reducing computational dependency on motion field complexity.
The authors tackled the problem of piecewise affine motion estimation by proposing a method that directly estimates motion fields without intermediate segmentation, resulting in lower computational cost and competitive accuracy on standard benchmarks.
Current algorithmic approaches for piecewise affine motion estimation are based on alternating motion segmentation and estimation. We propose a new method to estimate piecewise affine motion fields directly without intermediate segmentation. To this end, we reformulate the problem by imposing piecewise constancy of the parameter field, and derive a specific proximal splitting optimization scheme. A key component of our framework is an efficient one-dimensional piecewise-affine estimator for vector-valued signals. The first advantage of our approach over segmentation-based methods is its absence of initialization. The second advantage is its lower computational cost which is independent of the complexity of the motion field. In addition to these features, we demonstrate competitive accuracy with other piecewise-parametric methods on standard evaluation benchmarks. Our new regularization scheme also outperforms the more standard use of total variation and total generalized variation.