Content-oriented animation video no-reference quality evaluation method

The invention discloses a content-oriented animation video no-reference quality evaluation method. The method comprises the following steps: firstly, establishing a CG animation video quality database, performing feature vector extraction and labeling on video samples, and dividing the samples in the database into two parts, namely a training set and a test set; secondly, training a video content classifier based on the convolutional neural network by using the training set, and during training, inputting a frame image of the video and outputting a content category of the video; then, each type of videos in the training set is used for training a corresponding quality scoring model based on a BP neural network, in the training process, feature vectors of the videos are input, and quality scores of the videos are output; and finally, implementing the content-oriented CG animation video no-reference quality evaluation method by applying the trained convolutional neural network and BP neural network, and carryin