| 摘要 |
The present invention relates to the field of imaging. Disclosed are a method and electronic device for aesthetic enhancements of a face in a real-time video, the method comprising: predicting at least one detail macroblock of a face in a current real-time video frame to obtain a detailed image corresponding to the current real-time video frame (301); filtering an actual face in the current real-time video frame to obtain a filtered image corresponding to the current real-time video frame (302); and combining the detailed image and the filtered image to generate an aesthetically enhanced current real-time video, wherein the aesthetically enhanced current real-time video at least comprises an aesthetically enhanced face (303). By predicting the detail macroblock in the current real-time video frame, combining the detailed image and the filtered image generated after prediction, and generating the aesthetically enhanced current real-time video, efficiency for enhancing facial aesthetics of the face in the real-time video and user experience can be improved. |
