Native Resolution Detection for 4K-UHD Videos
| Autor: | Lin Li, Rong Xie, Feng Yanan, Li Song, Zaixin Yang, Yu Dong |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
Native resolution
Computer science business.industry Deep learning ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION 02 engineering and technology Resolution (logic) Feature (computer vision) Frequency domain 0202 electrical engineering electronic engineering information engineering Discrete cosine transform 020201 artificial intelligence & image processing Computer vision Artificial intelligence Aliasing (computing) business Interpolation |
| Zdroj: | BMSB |
| DOI: | 10.1109/bmsb49480.2020.9379531 |
| Popis: | 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 detection method for 4K-UHD videos. The method is based on extracting feature from the DCT frequency domain and integrating this feature in all frames. Our method can determine authenticity of 4K-UHD videos as well as native resolution, no matter which kind of SR model is applied for upscaling. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|