A No-Reference Metric for Quality Assessment of Tone-Mapped High Dynamic Range Images

semanticscholar(2017)

引用 0|浏览0
暂无评分
摘要
Tone mapping operations reduce the high dynamic range of an image to a range suitable for the display capabilities. This mapping can accentuate some details in the original scene while some others can be lost. Objective full-reference quality metrics have recently been proposed to assess the quality of tone-mapped images. Since the original scene content cannot be visualized on a low dynamic range display, the quality of the output image can be mainly evaluated based on the mapped output without referring to the original high dynamic range image. This paper proposes a new no-reference metric for quality assessment of tonemapped images. Results show that the new proposed metric outperforms recently developed metrics with reference to the Mean Opinion Score using the same image database, while at the same time significantly reducing the required computational time.
更多
查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要