A New Adaptive Video Super-Resolution Algorithm With Improved Robustness to Innovations
This work addresses a specific bottleneck in video super-resolution for applications requiring efficient and robust processing, though it is incremental as it builds on existing R-LMS methods.
The paper tackles the problem of video super-resolution reconstruction (SRR) algorithms degrading in the presence of innovation outliers, proposing a new method that improves robustness to outliers while maintaining computational costs comparable to R-LMS. Monte Carlo simulations show it outperforms traditional and regularized LMS versions and is competitive with state-of-the-art SRR methods at a much smaller computational cost.
In this paper, a new video super-resolution reconstruction (SRR) method with improved robustness to outliers is proposed. Although the R-LMS is one of the SRR algorithms with the best reconstruction quality for its computational cost, and is naturally robust to registration inaccuracies, its performance is known to degrade severely in the presence of innovation outliers. By studying the proximal point cost function representation of the R-LMS iterative equation, a better understanding of its performance under different situations is attained. Using statistical properties of typical innovation outliers, a new cost function is then proposed and two new algorithms are derived, which present improved robustness to outliers while maintaining computational costs comparable to that of R-LMS. Monte Carlo simulation results illustrate that the proposed method outperforms the traditional and regularized versions of LMS, and is competitive with state-of-the-art SRR methods at a much smaller computational cost.