Displays ( IF 4.3 ) Pub Date : 2021-09-28 , DOI: 10.1016/j.displa.2021.102092 Pan Da 1 , GuiYing Song 1 , Ping Shi 1 , HaoCheng Zhang 1
The quality of nighttime videos is important for consumer photography and monitoring of video clearness. However, little work has been done on the study of nighttime video quality assessment. In this paper, to the best of our knowledge, we explore the study on the nighttime video quality assessment for the first time. First, we build a real-world nighttime video quality assessment database (NVQA) containing 200 videos with abundant content and diverse distortion. Additionally, we carry out subjective tests to rate all nighttime videos in the NVQA database. Thereafter, we proposed a blind nighttime video quality assessment model based on feature fusion and conducted experiments to evaluate the performance and efficiency of our proposed model. The experiment results demonstrate that our model outperforms most traditional methods.
中文翻译:
夜间视频的感知质量评估
夜间视频的质量对于消费者摄影和视频清晰度监控非常重要。然而,关于夜间视频质量评估的研究工作很少。在本文中,据我们所知,我们首次探索了夜间视频质量评估的研究。首先,我们构建了一个真实世界的夜间视频质量评估数据库(NVQA),其中包含 200 个内容丰富且失真多样的视频。此外,我们还进行了主观测试,以对 NVQA 数据库中的所有夜间视频进行评分。此后,我们提出了一种基于特征融合的夜间盲人视频质量评估模型,并进行了实验以评估我们提出的模型的性能和效率。实验结果表明,我们的模型优于大多数传统方法。