|
ИСТИНА |
Войти в систему Регистрация |
ИПМех РАН |
||
Applicability limitations of differentiable full-reference image-quality metricsстатья
- Авторы: Siniukov Maksim, Kulikov Dmitriy, Vatolin Dmitriy
- Сборник: 2023 Data Compression Conference (DCC)
- Год издания: 2023
- Место издания: IEEE
- DOI: 10.1109/dcc55655.2023.00082
- Аннотация: More and more visual-quality metrics are being developed to assess the quality of images, but little research has considered their limitations. In this paper, we demonstrate that image preprocessing before compression can artificially increase the quality scores provided by the popular metrics DISTS, LPIPS, HaarPSI, and VIF. We propose a series of neural-network preprocessing models that increase DISTS by up to 34.5%, LPIPS by up to 36.8%, VIF by up to 98.0%, and HaarPSI by up to 22.6% in the case of JPEG-compressed images. However, a subjective comparison of these preprocessed images showed that the visual quality either dropped or remained unchanged, indicating the limited applicability of these metrics. We used a ResNet-like lightweight CNN architecture for preprocessing and the differentiable DiffJPEG algorithm for compression.
- Добавил в систему: Синюков Максим Владимирович