Scholarlink

<p>In recent years, windowed six degrees of freedom (DoF) video is becoming a crucial medium for delivering immersive visual experiences. However, the increased user freedom in viewpoint selection introduces unique challenges for quality assessment, particularly concerning viewpoint transitions and associated distortions. Existing quality assessment methods fail to adequately account for the characteristics of dynamic distortions in windowed 6DoF video. To address this issue, we propose a quality assessment method for windowed 6DoF video based on spatiotemporal dynamic adjustment. The proposed method contains three components, a basic quality predictor, a spatial regulator, and a temporal regulator. Firstly, the basic quality predictor extracts high-level semantic information to generate a basic quality score. Then, the spatial regulator effectively captures complex spatial distortion information using the non-subsampled contour-let transform and saliency-driven methods. The temporal regulator extracts visual discomfort distortions arising from viewpoint changes. Finally, the extracted spatial and temporal information are used to dynamically regulate the basic quality score and obtain the eventual quality score. The experimental results demonstrate that the proposed method exhibits excellent performance and has high competitiveness compared to the state-of-the-art quality assessment methods.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Windowed 6DoF video quality assessment based on spatiotemporal dynamic adjustment

  • 期刊论文

  • 发表日期：

    2025年9月5日

查看全文
  • Weihua Chen,
  • Fen Chen,
  • Qian Wang,
  • Youshuang Zhao,
  • Zongju Peng

摘要

In recent years, windowed six degrees of freedom (DoF) video is becoming a crucial medium for delivering immersive visual experiences. However, the increased user freedom in viewpoint selection introduces unique challenges for quality assessment, particularly concerning viewpoint transitions and associated distortions. Existing quality assessment methods fail to adequately account for the characteristics of dynamic distortions in windowed 6DoF video. To address this issue, we propose a quality assessment method for windowed 6DoF video based on spatiotemporal dynamic adjustment. The proposed method contains three components, a basic quality predictor, a spatial regulator, and a temporal regulator. Firstly, the basic quality predictor extracts high-level semantic information to generate a basic quality score. Then, the spatial regulator effectively captures complex spatial distortion information using the non-subsampled contour-let transform and saliency-driven methods. The temporal regulator extracts visual discomfort distortions arising from viewpoint changes. Finally, the extracted spatial and temporal information are used to dynamically regulate the basic quality score and obtain the eventual quality score. The experimental results demonstrate that the proposed method exhibits excellent performance and has high competitiveness compared to the state-of-the-art quality assessment methods.