# A Benchmark for Edge-Preserving Image Smoothing

## TL;DR

This paper introduces a comprehensive benchmark for edge-preserving image smoothing, including a dataset and baseline deep learning methods, enabling objective evaluation and faster, high-quality smoothing results.

## Contribution

It provides the first publicly available dataset and baseline deep learning algorithms for objective comparison of edge-preserving image smoothing methods.

## Key findings

- Deep networks outperform traditional algorithms in speed and quality.
- The benchmark enables consistent evaluation across diverse image contents.
- Baseline methods achieve state-of-the-art results both qualitatively and quantitatively.

## Abstract

Edge-preserving image smoothing is an important step for many low-level vision problems. Though many algorithms have been proposed, there are several difficulties hindering its further development. First, most existing algorithms cannot perform well on a wide range of image contents using a single parameter setting. Second, the performance evaluation of edge-preserving image smoothing remains subjective, and there lacks a widely accepted datasets to objectively compare the different algorithms. To address these issues and further advance the state of the art, in this work we propose a benchmark for edge-preserving image smoothing. This benchmark includes an image dataset with groundtruth image smoothing results as well as baseline algorithms that can generate competitive edge-preserving smoothing results for a wide range of image contents. The established dataset contains 500 training and testing images with a number of representative visual object categories, while the baseline methods in our benchmark are built upon representative deep convolutional network architectures, on top of which we design novel loss functions well suited for edge-preserving image smoothing. The trained deep networks run faster than most state-of-the-art smoothing algorithms with leading smoothing results both qualitatively and quantitatively. The benchmark is publicly accessible via https://github.com/zhufeida/Benchmark_EPS.