Identifying pitfalls in the evaluation of saliency models for videos

Dong, Zhengyan, Wu, Xinbo, Zhao, Xin, Zhang, Fan and Liu, Hantao ORCID: https://orcid.org/0000-0003-4544-3481 2022. Identifying pitfalls in the evaluation of saliency models for videos. Presented at: IEEE Image, Video, and Multidimensional Signal Processing Workshop (IVMSP), Nafplio, Greece, 26-29 June 2022. Proceedings of IEEE 14th Image, Video, and Multidimensional Signal Processing Workshop (IVMSP). IEEE, 10.1109/IVMSP54334.2022.9816306

PDF - Accepted Post-Print Version Download (913kB)

Abstract

Saliency prediction has been extensively studied for natural images. In the area of video coding and video quality assessment, researchers attempt to integrate a saliency model to individual frames of a video sequence. In selecting best-performing saliency models for these applications, the evaluation only considers the average model performance over all frames of a video. This may mask the defects of a saliency model and consequently hinder further improvement of the model. In this paper, we present the identification of pitfalls in the evaluation of saliency models for videos. We demonstrate the importance of considering the video content classification and temporal effect. Building on the findings, we make recommendations for saliency model evaluation and selection for videos.

Item Type:	Conference or Workshop Item (Paper)
Date Type:	Published Online
Status:	Published
Schools:	Computer Science & Informatics
Publisher:	IEEE
ISBN:	9781665478236
Date of First Compliant Deposit:	8 August 2022
Last Modified:	31 Mar 2023 06:17
URI:	https://orca.cardiff.ac.uk/id/eprint/151812

Actions (repository staff only)

Edit Item