Blind Quality Assessment of Light Field Image Based on Spatio-Angular Textural Variation
Résumé
Light Field Image Quality Assessment (LF-IQA) is vitally important to facilitate the development of immersive technologies. However, current state-of-the-art LF-IQA metrics still struggle handle (LFI) with massive data in an efficient manner. To cope this challenge, we propose a simple yet effective Blind metric based on Spatio-Angular Textural Variation, named SATV-BLiF. Given distorted LFI, first apply Local Binary Pattern (LBP) operator measure textural variation spatial and angular domains respectively. Then generated matrices are merged further transformed into statistical histogram features. Finally, Support Vector Regression (SVR) employed construct nonlinear mapping function between features perceptual quality score LFI. Experimental results three representative light field databases show that proposed achieves evaluation performance, while having much lower complexity than existing No-Reference (NR) metrics. The code SATV-BLiF available at https://github.com/ZhengyuZhang96/SATV-BLiF.