Abstract: Optical coherence tomography (OCT) angiography (OCTA) data is generated by one or more machine learning systems to which OCT data is input. The OCTA data is capable of visualization in three dimensions (3D) and can be generated from a single OCT scan. Further, motion artifact can be removed or attenuated in the OCTA data by performing the OCT scans according to special scan patterns and/or capturing redundant data, and by the one or more machine learning systems.

Abstract: The present disclosure provides methods, apparatuses and systems for transcoding a video. One exemplary method for transcoding a video includes: receiving a model file issued by a machine learning framework; converting the model file to a file identifiable by a filter; setting one or more filtering parameters according to the file; and processing the video based on the filtering parameters. According to the embodiments of the present disclosure, no extra storage will be consumed in the process of video transcoding, and system resource consumption can be reduced without adding any extra system deployment costs.

Abstract: A method including selecting a first definition and a second definition from multiple levels of definition, and determining a first resolution and a second resolution that respectively correspond to the first definition and the second definition; transcoding a first part of content of a media file based on the first resolution and the second resolution and according to a preset transcoding rule, and recording a first real-time quantization parameter (QP) value and a second real-time QP value in the transcoding process; determining whether the first real-time QP value and the second real-time QP value meet a preset detection rule; if the determining result is no, adjusting the first resolution and/or adjusting the second resolution; and transcoding the non-transcoded part in the media file according to the adjusted first resolution and/or second resolution. The transcoding method and apparatus increase the transcoding speed of a media file.

Abstract: The present disclosure provides methods, apparatuses and systems for transcoding a video. One exemplary method for transcoding a video includes: receiving a model file issued by a machine learning framework; converting the model file to a file identifiable by a filter; setting one or more filtering parameters according to the file; and processing the video based on the filtering parameters. According to the embodiments of the present disclosure, no extra storage will be consumed in the process of video transcoding, and system resource consumption can be reduced without adding any extra system deployment costs.

Abstract: A method including selecting a first definition and a second definition from multiple levels of definition, and determining a first resolution and a second resolution that respectively correspond to the first definition and the second definition; transcoding a first part of content of a media file based on the first resolution and the second resolution and according to a preset transcoding rule, and recording a first real-time quantization parameter (QP) value and a second real-time QP value in the transcoding process; determining whether the first real-time QP value and the second real-time QP value meet a preset detection rule; if the determining result is no, adjusting the first resolution and/or adjusting the second resolution; and transcoding the non-transcoded part in the media file according to the adjusted first resolution and/or second resolution. The transcoding method and apparatus increase the transcoding speed of a media file.

Abstract: Embodiments of the disclosure provide a video processing method and a video processing apparatus. The video processing method may include estimating noise of an input video to obtain noise level information of the input video; and selecting, according to the noise level information, a denoising scheme to process the input video. The denoising scheme may merge a plurality of denoising algorithms.