This paper proposes a novel no-reference quality assessment method for text-to-image generation. Text-to-image refers to the process of generating image content from textual descriptions using deep learning models. Although advances in technology and improvements in models have made it possible to generate some high-quality images, some generated images still exhibit unique distortions that reflec...Show More
The evaluation of free-viewpoint video (FVV) quality is essential for improving the quality of experience (QoE). Prior deep video quality assessment (VQA) approaches for FVV typically focused on either spatial or temporal distortions and lacked a comprehensive assessment considering the two aspects. In this paper, we provide an end-to-end no-reference video quality assessment (NRVQA) model for FVV...Show More
The rapid development of display devices and the continuous improvement of visual quality requirements put urgent demands on high bit-depth (HBD) image, but most legacy data are stored and transmitted at low bit-depth (LBD). Bit-depth expansion (BDE) methods are designed to handle this dilemma by reconstructing HBD contents from LBD. But the expansion process often introduces artifacts such as fal...Show More
There are many resolution enhancement methods to produce 4K-UHD videos including interpolation and deep learning based Super-Resolution, but lack of details, aliasing and unnatural transition are still inevitable and affect viewer's immersive experience. Therefore, it is necessary to detect authenticity and native resolution of 4K-UHD videos. In this paper, we propose a new native resolution detec...Show More