What is it about?
This paper poposed a new global optimization-based smoother, named iterative least squares (ILS), for efficient edge-preserving image smoothing. ILS can produce high-quality results but at a much lower computational cost. ILS is simple where only two steps need to be iterated for a few times. It is also highly parellel and easily accelerated. With the acceleration of a GTX 1080 GPU, it is able to process images of 1080p resolution (1920 × 1080) at the rate of 20fps for color images and 47fps for gray images. Promissing results are produced in image detail enhancement and HDR tone mapping.
Featured Image
Why is it important?
The proposed method narrows down the gap between smoothing quality and processing speed. It can produce high-quality results and high processing efficiency at the same time, which makes it practical for real applications.
Perspectives
I think this is a good example of achieving both smoothing quality and processing efficiency. I have also made the code publicly available at: https://github.com/wliusjtu/Real-time-Image-Smoothing-via-Iterative-Least-Squares
Wei Liu
University of Adelaide
Read the Original
This page is a summary of: Real-time Image Smoothing via Iterative Least Squares, ACM Transactions on Graphics, June 2020, ACM (Association for Computing Machinery),
DOI: 10.1145/3388887.
You can read the full text:
Resources
Contributors
The following have contributed to this page
