5 results about "Subjective quality" patented technology

The embodiment of the application discloses a kind of encoding method, device, encoder, medium and product, belong to video coding technique field.The encoding method includes: obtaining multiple candidate encoding parameters, candidate encoding parameters are used to indicate the encoding mode of current frame;Obtain the encoding quality corresponding to multiple candidate encoding parameters respectively by subjective quality prediction model, subjective quality prediction model is the machine learning model obtained by training sample image frame, sample image frame is labeled with reference encoding quality, reference encoding quality is labeled from subjective view, and reference encoding quality is used to express the data consistency before and after sample image frame encoding;From multiple candidate encoding parameters, obtain the target encoding parameter of the encoding quality that meets quality requirement;Output the code stream data of current frame encoding under target encoding parameter.Through subjective quality prediction model, the efficiency of obtaining the encoding mode adapted to current frame is improved, and the accuracy and stability of encoding process are improved.

The invention discloses a cross-platform video conference fusion method and system, and the method comprises the steps: collecting the coding format, equipment type and network state of each terminal, generating a coding capability configuration table, constructing a coding perception quality mapping model in combination with historical subjective quality score data and video content features, and achieving the subjective quality prediction of multiple terminals in a conference scene; the system automatically triggers code control parameter optimization and code rate allocation dynamic adjustment according to the quality score difference of each terminal, a quantization matrix is optimized for low-quality nodes, code rate compression is implemented for high-quality nodes, and the overall visual quality consistency is improved; the prediction model is dynamically corrected based on real-time feedback data, the adaptability and robustness of the model are continuously improved through periodic incremental training, dynamic consistency of video quality of multiple coding nodes in a conference scene is effectively achieved, and collaborative experience is optimized.

The invention relates to the technical field of video coding, in particular to a monitoring scene video coding code rate control method and system suitable for hardware implementation, and the method comprises the following steps: in a low-delay P-frame coding mode, initializing the state of a buffer area, and calculating the average target code rate of each frame and the QP of the first frame according to the target code rate, frame rate and resolution input by a user; dynamically adjusting a GOP-level target code rate, a GOP-level QP, a frame-level target code rate and a frame-level QP based on a buffer area state, a buffer area size, an average target code rate of each frame, a first frame QP and preorder coding information; and after one frame is coded, further adjusting the frame-level target code rate based on the motion degree of the current frame, further updating the state of the buffer area, and adjusting the length of the GOP. According to the method, the code rate control with low calculation complexity can be realized, the subjective quality stability of coding is ensured, the intelligent code rate adjustment is carried out by using the motion information, and the compression efficiency of the encoder is further improved on the premise of ensuring the subjective quality.

The invention relates to the technical field of video quality evaluation, in particular to a subjective quality evaluation method for NeRF scene reconstruction, and the method comprises the steps: S1, selecting a real multi-view image sequence from a plurality of open real scenes; s2, performing multi-level illumination distortion on the real multi-view image sequence of each open real scene; s3, aiming at each open real scene, constructing a plurality of track viewpoint paths for simulating a real watching behavior; s4, reconstructing an open real scene based on the distorted image sequence through a plurality of NeRF models, and rendering a plurality of track viewpoint paths in the reconstructed scene to generate a corresponding pre-rendered video sequence; s5, constructing a subjective evaluation environment, and organizing subjects to perform subjective quality scoring on each pre-rendered video sequence; and S6, cleaning and screening the subjective quality score of each pre-rendered video sequence, and constructing a subjective quality evaluation data set. The data set constructed by the method has the advantages of high authenticity, comprehensive distortion types, rich evaluation dimensions and the like.