32 results about "Temporal similarity" patented technology

A method for generating spatial-temporal consistency depth map sequences based on convolutional neural networks for 2D-3D conversion of television works includes steps of: 1) collecting a training set, wherein each training sample thereof includes a sequence of continuous RGB images, and a corresponding depth map sequence; 2) processing each image sequence in the training set with spatial-temporal consistency superpixel segmentation, and establishing a spatial similarity matrix and a temporal similarity matrix; 3) establishing the convolution neural network including a single superpixel depth regression network and a spatial-temporal consistency condition random field loss layer; 4) training the convolution neural network; and 5) recovering a depth maps of a RGB image sequence of unknown depth through forward propagation with the trained convolution neural network; which avoids that clue-based depth recovery method is greatly depended on scenario assumptions, and inter-frame discontinuity between depth maps generated by conventional neural networks.

The invention discloses a method for classifying moving objects in a video, a method and a device for analyzing traffic flow. The method includes: extracting the motion trajectories of each target object in the video; performing the similarity modeling of the spatio-temporal relationship on the motion trajectories of each target object, and determining the spatio-temporal similarity between the motion trajectories of each target object; using the According to the spatio-temporal similarity between the trajectories of each target object, the trajectories of each target object are clustered to obtain groups of target objects that are similar in time and space. It can ensure the speed and accuracy of video data monitoring and the effectiveness of abnormal event detection.

The invention discloses a method for complementing continuity missing data in dam deformation monitoring. The method solves the problem of completing continuity missing data in dam deformation monitoring data. Firstly, the lack of continuity in the deformation monitoring data is preprocessed; secondly, from the perspective of global space, global time, local space, local time and semantics, the spatial-temporal similarity and functional Global and local space, time and semantic interpolation are performed on the missing data in the data; finally, a deep neural network model is constructed, and the preliminary results of the above-mentioned deformation monitoring missing data complement are used as input, and the deep neural network is used to realize non-linear fusion. Completion of continuous missing data in dam deformation monitoring.

This specification discloses a video matching method, a blockchain-based infringement proof storage method and a device, the blockchain-based infringement proof storage method includes: obtaining multiple feature vectors of the target video; multiple feature vectors based on the target video A feature vector, from the video database, retrieves a candidate video similar to the target video; based on the target video and the candidate video, constructs a temporal similarity matrix feature map between the target video and the candidate video ;Using the time domain similarity matrix feature map as the input of the deep learning detection model, to output video clips and corresponding similarities matched with the target video in the candidate video; when the similarity is greater than or equal to preset When the similarity threshold is reached, the summary of the target video, the video segments matching the target video in the candidate video and the corresponding similarity infringement evidence are uploaded to the block chain.

The invention discloses a method for dividing TVS events into clusters based on optical flow analysis. The method divides the TVS events into clusters by computing the optical flows of the TVS events, extracts the accurate position and the shape information of a moving target according to the information of the clusters, and is characterized by acquiring the optical flow speed of each spot by analyzing the optical flows event by event, further dividing the events into clusters according to the similarity of speed, positions, and output time, and dynamically updating pre-existing events in the clusters according to cluster speed. The method divides the TVS events into clusters according to the generated time, the spatial position, and the optical flow speed of the events, and has advantages that the combination between a characteristic that spatial moving targets have same projection speed on a 2D imaging plane and space-time similarity may solves overlapping and shielding problems of the targets so as to achieve higher accuracy; and cluster speed may update the pre-existing events in the clusters.

The present invention discloses a vegetation loss direction identification method based on a multi-remote-sensing index trend. The method comprises: calculating a temporal similarity of vegetation indexes between each year and a beginning year by using a JM distance to generate a track of temporal similarity of vegetation indexes; extracting a potential vegetation loss region according to a variation of the temporal similarity of the vegetation index, so as to define a region where the vegetation index is significantly decreased and an impervious surface index is significantly increased as a vegetation loss region; and on this basis, finally determining different vegetation loss directions such as urbanization, desertification and wetland formation according to a water body index and a bare soil index trend feature. In the method, the vegetation change region is determined by using the variation of the temporal similarity, and further, the vegetation loss direction is determined according to multiple remote sensing indexes, without depending on manual intervention for threshold setting, so that the method has the characteristics of high robustness, high classification precision, high automation and storing anti-interference ability, and so on.

The invention discloses a method for generating a spatial-temporal consistency depth map sequence based on a convolution neural network, which can be used in a film and television work 2D-to-3D technology. The method comprises the following steps: (1) collecting a training set, wherein each training sample in the training set is composed of a continuous RGB image sequence and a corresponding depth map sequence; (2) carrying out spatial-temporal consistency super pixel segmentation on each image sequence in the training set, and constructing a spatial similarity matrix and a temporal similarity matrix; (3) constructing a convolution neural network composed of a single super pixel depth regression network and a spatial-temporal consistency conditional random field loss layer; (4) training the convolution neural network; and (5) for an RGB image sequence of unknown depth, using the trained neural network to recover a corresponding depth map sequence through forward propagation. The problem that a depth recovery method based on clues relies too much on the scene hypothesis and the problem that the frames of a depth map generated by the existing depth recovery method based on a convolution neural network are discontinuous are avoided.