33 results about "Video assessment" patented technology

A system obtains, by a video evaluator, a video clip generated by a video generator of the avatar generator. The system obtains, by the video evaluator, video features of a target person that the avatar is representing. The system compares the video clip with the video features of the target person using a set of video metrics. The system generates a video evaluation score for the video clip based on a comparison of the video clip and the video features.

This invention proposes a method, system, and computer device for generating text-based videos, relating to the technical field of video processing. The method includes: acquiring user input parameters; generating anchor point prompts and anchor point character diagrams based on the user input parameters; evaluating the anchor point character diagrams until they pass evaluation; extracting information from the anchor point prompts to obtain a storyboard; evaluating the storyboards until they pass evaluation; generating audio and storyboard video based on the storyboards; evaluating the storyboard video until it passes evaluation; post-processing the storyboard video to obtain a post-processed video; and combining the post-processed video with the audio to generate a target video. This invention effectively provides a closed-loop evaluation node for text-based video generation, improving the quality of generated text-based videos.

The invention discloses a video-based omni-directional remote rehabilitation training system, a video-based omni-directional remote rehabilitation training method and a medium, which are characterized in that a digital standardized training video library classified according to four stages of PT physical therapy is established, and more than 700 digital standardized training videos such as joint activity, balance training, gait correction and the like are covered; it is ensured that the training content is scientific and covers the whole period and all directions of rehabilitation training; rehabilitators generate personalized training schemes in combination with rehabilitation scene types (hospitalization / home / community / old-age care institutions) and clinical data and in combination with remote video evaluation of the rehabilitators, and training requirements in different environments are met. By capturing actions of limb joint angles, joint point motion trails and muscle force changes, a training action deviation rate is calculated and is fed back and output to a rehabilitation teacher in a grading manner, so that the training standardability is remarkably improved; rehabilitators can manage multiple patients online at the same time, remotely check training percentage data, adjust schemes and generate rehabilitation notes, and efficient, accurate and personalized rehabilitation guidance services are provided for the patients.

The invention provides a video evaluation method and device based on artificial intelligence, an intelligent agent, electronic equipment, a storage medium and a program product, relates to the technical field of artificial intelligence, in particular to the technical fields of computer vision, deep learning, large models and the like, and can be applied to scenes such as content generation based on artificial intelligence. The specific implementation scheme is as follows: in response to a received evaluation request, determining a to-be-evaluated video and an evaluation index of the to-be-evaluated video; under the condition that the evaluation index comprises a time sequence evaluation index, analyzing the to-be-evaluated video to obtain content information used for describing the to-be-evaluated video; based on the content information and the evaluation indexes, evaluation prompt information is generated; and inputting the assessment prompt information and the to-be-assessed video into the multi-modal large model to obtain a time sequence assessment result for the to-be-assessed video.

The application discloses a kind of based on deep learning network's ultra-high-definition video compression damage grade evaluation method, comprising the following steps;Step 1, establish multi-resolution compression damage training set, verification set, test set: step 2, construct the compression damage grade evaluation network of time-frequency domain combination: step 3, train the compression damage grade evaluation network of time-frequency domain combination: step 4, test the accuracy of multi-resolution compression damage test set;Step 5, evaluate the compression damage grade evaluation network of time-frequency domain combination.The application is evaluated to ultra-high-definition video compression damage, solves the problem that compression damage data set production cost is high, not easy to expand, video compression damage evaluation network does not adopt end-to-end design, and generalization ability is insufficient and the problem that ultra-high-definition video evaluation is time-consuming.

Systems and methods in accordance with many embodiments of the invention include a motion evaluation system that trains a model to evaluate motion (such as, but not limited to, gait) through images (or video) captured by a single image capture device. In certain embodiments, motion evaluation includes predicting clinically relevant variables from videos of patients walking from keypoint trajectories extracted from the captured images.

The present disclosure provides a method, apparatus, and electronic device for video quality assessment, relating to the field of artificial intelligence technology, particularly video assessment. The implementation scheme comprises: obtaining quality assessment data for a video comprising multiple video segments, wherein the quality assessment data includes a score on at least one quality assessment dimension; determining, for each quality assessment dimension, a score for at least one video segment evaluated on the quality assessment dimension among the multiple video segments; and determining a quality assessment result for the video based on one or more video segments in the at least one video segment having the same predetermined score.

The application discloses a configuration method of a formation performance scheme, and the method is applied to a user terminal device, and the method comprises the following steps: obtaining a take-off scheme determined by a user; obtaining a performance scheme determined by the user; generating a formation flight scheme based on the take-off scheme and the performance scheme; performing flight simulation on an aircraft fleet according to the formation flight scheme, obtaining simulation data, and generating a performance video corresponding to the formation flight scheme according to the simulation data, so that the user can evaluate the formation flight scheme based on the performance video. The application can save communication cost and generate the formation flight scheme based on the user demand.

The application provides a CIPN information fusion method and system based on multi-modal information, comprising: evaluating a subjective questionnaire of a to-be-tested person; evaluating walking speed and gait according to a walking video of the to-be-tested person; evaluating finger fine motor skills of the to-be-tested person; evaluating nerve conduction velocity of the to-be-tested person; evaluating brain function of the to-be-tested person; evaluating brain structure of the to-be-tested person; and constructing information fusion according to the subjective questionnaire, the walking speed and gait, the finger fine motor skills, the nerve conduction velocity, the brain function and the brain structure. Compared with a traditional NCI-CTCAE grading system and an EORTC QLQ-CIPN20 questionnaire, the application overcomes subjectivity and repeatability problems, and improves diagnostic accuracy and treatment effect.

The system according to the embodiment aims to allow participants to create and evaluate music and video in real time. [Solution] A system according to an embodiment includes a reception unit, a generation unit, a video generation unit, a performance unit, and an evaluation unit. The reception unit receives input of music genre and tempo from participants. The generation unit generates music based on the information received by the reception unit. The video generation unit generates video based on the music generated by the generation unit. The performance unit plays the music and video generated by the generation unit and the video generation unit in real time. The evaluation unit evaluates the music played by the performance unit.

The invention discloses a modularized industrial robot intelligent process generation method and system based on video language interactive learning, and aims at solving the technical bottleneck that a traditional industrial robot is complex in programming and difficult in task migration. The method comprises the following steps: analyzing space-time association between a natural language instruction and video demonstration by constructing a video language contrast learning model, and generating cross-task reusable semantic representation; combining a reinforcement learning reward mechanism to fuse sparse task signals and dense video evaluation, and driving adaptive optimization of a strategy model; and dynamically mapping the semantic instruction into an executable workflow of the industrial robot based on a componentized architecture, and verifying logic completeness through an automatic state machine. According to the method, the operation threshold of non-professionals is remarkably reduced, rapid task deployment of cross-entity platforms is realized, the system flexibility and execution precision in a complex manufacturing scene are improved, and a closed-loop self-optimization process control solution is provided for intelligent manufacturing.

The present disclosure relates to a system to evaluate an avatar generated by an avatar generator. The system comprises an evaluation module including an audio evaluation module for evaluating audio features and a video evaluation module for evaluating video features. Evaluation of the avatar includes extracting audio and video features from the avatar and applying a set of evaluation metrics for generating audio and video evaluation scores. The scores are combined to generate a final score. For avatar generator evaluation, audio clip and video clip are provided to the audio evaluation module and video evaluation module, respectively. A set of evaluation metrics is applied for evaluation. Each metric can generate a score. All scores are combined to generate a final evaluation score.

The invention provides an audio-visual quality evaluation method and system for audios and videos, electronic equipment and a storage medium, and relates to the technical field of audio-video evaluation, and the method comprises the steps: collecting the biometric data of a user in real time in the playing process of a target audio and video, and the environment data of a playing environment; performing feature extraction according to the biometric data to obtain biometric features, and converting the environmental data into environmental features; performing feature extraction on the target audio and video to obtain audio-visual signal features; performing multi-modal feature fusion on the biometric features, the environment features and the audio-visual signal features to obtain a fusion feature vector; and inputting the fusion feature vector into a pre-trained audio-visual quality evaluation model to obtain an audio-visual quality evaluation result of the target audio and video. According to the method, subjective perception and accurate prediction of the audio-visual quality are realized, the limitation that only signal indexes are relied on traditionally is broken through, and the accuracy and individuation level of audio-visual quality evaluation are remarkably improved.

To provide a nursing care state determination device contributing to reduction of a falling risk of a care-receiver.SOLUTION: A nursing care state determination device 1 for determining a nursing care state of a person to be cared for includes a moving image acquisition unit 10 that acquires a moving image obtained by photographing a determination target person, a moving image analysis unit 20 that determines whether or not the acquired moving image is a target moving image obtained by photographing a standing-up motion of the determination target person from a side direction, a determination unit 30 that determines a falling risk of the determination target person based on the target moving image, and a determination result output unit 40 that outputs a determination result of the falling risk. The video analysis unit 20 includes the coordinate providing unit 21 that provides the target video with the coordinate information regarding the earlobe, the acromion, the greater trochanter, the posterior surface of the knee joint, and the lateral malleolus of the determination target person, and the determination unit 30 includes the movement amount measurement unit 31 that measures the movement amount of the body part from the coordinate information and the angle measurement unit 33 that measures the head inclination angle and the knee joint angle and determines the risk of falling on the basis of the measured movement amount and angle.SELECTED DRAWING: Figure 1

Systems and methods for improving video generation models are provided. In some implementations, a video evaluation input prompt is assembled, including data indicative of a synthetic digital video and a request to evaluate the video based on one or more criteria. The synthetic digital video comprises a temporal sequence of frames generated by a video generation model and portrays simulated events. The video evaluation input prompt is processed using a generative model, such as a vision-language model, to generate evaluation output indicative of an outcome of the evaluation. Based on the evaluation output, the video generation model is adapted. The adaptation can be performed using reinforcement learning with the evaluation output as reward data or using direct preference optimization with the evaluation output as preference data. This process can improve the physical plausibility and logical consistency of videos subsequently generated by the adapted model.

A method for evaluating video quality based on a non-reference video to evaluate the quality of a received video using artificial intelligence composed of convolutional neural networks (CNN) and recurrent neural networks (RNN) capable of setting a learning range, includes: (a) extracting RGB values for one frame of a video; (b) obtaining an output by providing the extracted RGB values to CNN No. 1; (c) obtaining an output by providing the extracted RGB values n times to a CNN; (d) repeating (a) to (c) for all frames and merging outputs of all CNNs; (e) obtaining the output of the RNN with the time dimension reduced to 1 after passing the merged output value to the RNN; and (f) after applying a regression algorithm to the final output of the RNN, predicting this value as video quality value.

The invention relates to the technical field of integrated platforms, in particular to an autism early intervention management integrated platform based on telemedicine, which comprises a cloud service layer, an application layer, a network adaptability module and a front-end display layer, the cloud service layer is deployed on at least one server, and the cloud service layer further comprises a child information database used for storing child personal information, an evaluation scale and behavior video multi-source heterogeneous data; the AI model service unit comprises a deep learning sub-model for behavior recognition and a statistical sub-model for evaluation scale analysis; the recommendation engine unit is used for outputting a personalized IEP target and an intervention scheme by adopting a collaborative filtering and rule matching algorithm based on the child characteristics and historical intervention data loaded in the database; according to the invention, a client network adaptability module and a local caching strategy are introduced, so that the continuity and integrity of video evaluation and interaction data under network fluctuation are ensured.

A video assessment method includes: obtaining a first video group, where the first video group includes a plurality of first videos; processing watch times of the plurality of first videos to obtain parameters of the first video group, where the parameters of the first video group indicate biases of the watch times of the plurality of first videos and noise of the watch times of the plurality of first videos; and obtaining assessment values of the plurality of first videos based on the parameters of the first video group and the watch times of the plurality of first videos, where the assessment values of the plurality of first videos indicate degrees of interest to a user for the plurality of first videos.

The embodiment of the invention provides a video evaluation method and electronic equipment, and the method comprises the steps: carrying out the sampling of a video based on a plurality of sampling rates, and obtaining a plurality of sub-videos of the video; performing feature extraction on each sub-video by using a video evaluation model to obtain a first feature vector corresponding to each sub-video; combining the first feature vectors corresponding to all the sub-videos to obtain a combined feature vector; performing feature extraction on the combined feature vector by using the video evaluation model to obtain a second feature vector; and based on preset prompt information and the second feature vector, evaluating the video by using the video evaluation model to obtain an evaluation result of the video. According to the invention, the robustness and adaptability of a video quality evaluation task can be improved, so that the accuracy of a video evaluation result is improved.

Disclosed in the present invention is a mental disorder rehabilitation training system based on a virtual human. The rehabilitation training system comprises a patient end and a doctor end, which are in communication connection with each other, wherein the patient end comprises a login module, an assessment module, a rehabilitation module and a management module; and the doctor end comprises a patient list module, a video assessment module and a live-streaming course module. A patient completes account registration and logs into the system by means of the login module, and completes a psychiatric assessment in the assessment module; a rehabilitation training task is pushed on the basis of a psychiatric assessment result, and training is completed in the rehabilitation module; and in addition to rehabilitation training, the patient needs to record rehabilitation data in the management module every day, and the system pushes a follow-up visit reminder and health-care knowledge on the basis of the physical condition of the patient. Correspondingly, a doctor can log into the doctor end and view the rehabilitation situation of the patient in the patient list module, so as to provide comments and guidance.

The invention relates to a video playing method and device, electronic equipment and a computer program product, and relates to the technical field of streaming media. The method comprises the following steps: acquiring a video watching data set; according to data features corresponding to the video watching data, determining video playing quality and watching exit probabilities under different video playing parameter values; constructing a video evaluation model based on mapping relations between the data features and the video playing quality and between the data features and the watching exit probability; determining the estimated playing quality and the predicted exit probability of the target video through the video evaluation model, and predicting according to the estimated playing quality and the predicted exit probability to obtain a user experience score; and performing a playing decision operation according to the user experience score, and determining a target playing parameter value of the target video. According to the invention, by analyzing the mapping relation between the data features of the video watching data and the video playing quality and the watching exit probability, differential algorithm strategy dynamic adjustment is realized for different users, so that the subjective experience of the users is optimized.

The application discloses a short video evaluation and creation assistance method and system based on visual understanding, the method comprising: acquiring short video data, extracting key frames from the video; detecting the key frames to obtain the proportion information of the face region position in the picture; identifying the expression category of the detected face region, and counting the number of expression category changes in adjacent time windows; estimating the global motion amount of the picture based on the optical flow information between adjacent frames, calculating the lens shake degree and forming the picture stability index; obtaining the video frame change rate by calculating the pixel difference and color histogram difference between adjacent frames; and fusing the above indexes to output a comprehensive score of the video hit potential. The application analyzes visual features such as the state of the character in the video picture, the expression change, the picture stability and the rhythm change, constructs an interpretable hit potential evaluation model, and provides objective, real-time and executable intelligent assistance decision for short video and live content creation.

The present invention relates to a method for assessing human posture videos based on a spatiotemporal graph convolutional network (STGCN), belonging to the field of computer vision technology. The method comprises the following steps: obtaining a human posture point cloud video, preprocessing it, and constructing a dataset; then improving the STGCN model to generate highly accurate human posture assessment scores; training the improved STGCN model and saving the final trained network state; and using the trained generative network to assess human posture videos. The method can generate highly accurate human posture assessment scores that are more suitable for practical application scenarios and can play an important guiding role in autonomous rehabilitation assessment.

A device for ensuring correct results in joint self-tests, characterized in that this invention, available to each participant and consisting of various system parts – as well as its logical, physical, and data-related interconnection – autonomously regulates and controls the procedure of a participating group of people both temporarily and logically, such that the control unit (SE) of the device informs the group of people around the time of the desired reporting level (M) with infection status (INFQ) (VI), at which time (S) an optionally video-documented self-test (T1, VS) – alternatively (self-)tests with other personal identification – is to be carried out immediately, wherein an included random function (RAN) ensures that this time is not predictable, and in the case of the video-documented self-test, the test stick of the device is marked (K) for identification in the video evaluation.which can also be AI-based, and the swab (TU) has an additive / impregnation with an acid indicator (INDI) so that deliberate wetting by acidic substances by the test subject – and the associated discoloration – leads to an invalid test result – and the non-distribution of a reward; and the device determines the sum of all positive test results (INFM) from the tests (T1), whereby the device then calculates the time interval (INT) after which, in the case of a positive initial test (T1), a second test (T2) must be carried out in order to obtain the reward for the test subject and their housemates if the test result is again positive and quarantine (Q1) has been observed, wherein the calculation is reciprocal – for example, with the calculation approach Q1*(1-INFQ / INFM) – and the resulting adaptive shortening of the test interval (INT) is intended to achieve as many justified reward distributions as possible.

The invention belongs to the technical field of artificial intelligence, is applied to a financial business marketing propaganda scene, and relates to a video evaluation method, device and equipment based on an AI model and a storage medium of the video evaluation method and device. Obtaining aesthetics and content evaluation results fed back by batch users according to a preset quality questionnaire; based on the key evaluation result, generating an evaluation label for evaluating the video data; setting an evaluation output node in a preset AI evaluation model according to the evaluation label; performing learning training on the AI evaluation model to obtain an AI evaluation model after learning training is completed; and inputting a to-be-evaluated video into the AI evaluation model after learning training is completed, and obtaining an aesthetics and content evaluation result output by the AI evaluation model. When the video evaluation method is applied to financial or medical marketing propaganda scenes, the quality of marketing propaganda videos can be controlled, and related businesses are assisted to provide propaganda videos with better quality.

A method of video evaluation for a video comprising successive video frames comprises the steps of generating optical flow vectors for at least a subset of corresponding pixels between frame pairs, s6

The invention discloses a human motion assessment method based on a multi-modal large language model, and belongs to the field of computer vision and artificial intelligence. Aiming at the problems of lack of professional training data and low video evaluation calculation efficiency in the prior art, the method comprises the following steps: firstly, generating diversified moving images in batches by using an image generation model taking a human body skeleton diagram as a control condition, and performing automatic professional labeling by means of a commercial large language model to construct a training data set; performing multi-task fine tuning on the open source visual language model by using the data set to obtain a professional exercise evaluation model; in the application stage, key frames are extracted by detecting periodic extreme points of human body joint angles in a user motion video, and only the key frames are input into the evaluation model to generate final scores and text feedback. According to the invention, automatic construction of low-cost and high-quality training data is realized, and efficient and real-time professional exercise evaluation can be carried out.

The invention belongs to the technical field of video evaluation, and discloses a video quality VMAF score evaluation method, which comprises the following steps of: 1, cutting a video by using ffmpeg, and assuming that the video in the idxth second needs to be extracted at present; according to the invention, matching is carried out based on a frame in a certain time interval in a reference video and a time interval corresponding to a distorted video after a range is expanded, a corresponding frame is found according to a PSNR value, vmaf frame alignment limitation is broken through, a matching frame searching algorithm is optimized, the corresponding frame is searched for in each iteration, the corner mark position of the current frame is recorded, and the next iteration searching is carried out based on the corner mark. And if the search result reaches the set score threshold, the corresponding frame is matched, and the search is stopped, the frame is matched one by one according to the method, so that the efficiency can be improved by 40% compared with the total matching efficiency, the matching search time is shortened to the maximum extent, and for the problem of inconsistent resolution, on the basis of an FFmpeg scaling filter, the matching efficiency is improved by 40%, and the matching search time is shortened to the maximum extent. And adjusting the reference video and the distorted video to be uniform in resolution.