CN111652181B - Target Tracking Method and Device And Electronic Equipm…

Legal status (The legal status is an assumption and isn't a legal conclusion. Current Assignee (The listed assignees could also be inaccurate. Priority date (The priority date is an assumption and isn't a authorized conclusion. The application discloses a goal tracking methodology, ItagPro a goal tracking device and digital tools, and pertains to the technical subject of artificial intelligence. The strategy includes the following steps: a first sub-network in the joint monitoring detection network, a first characteristic map extracted from the target function map, and a second characteristic map extracted from the goal characteristic map by a second sub-network within the joint monitoring detection community; fusing the second characteristic map extracted by the second sub-community to the first function map to acquire a fused characteristic map corresponding to the primary sub-community; buying first prediction info output by a first sub-community primarily based on a fusion feature map, and iTagPro product buying second prediction info output by a second sub-community; and figuring out the present place and the movement path of the moving target in the goal video primarily based on the first prediction information and the second prediction info.

The relevance among all of the sub-networks which are parallel to one another could be enhanced by means of characteristic fusion, and the accuracy of the determined place and movement trail of the operation target is improved. The present utility relates to the field of synthetic intelligence, iTagPro locator and particularly, to a target tracking technique, apparatus, and electronic machine. Lately, artificial intelligence (Artificial Intelligence, AI) technology has been extensively utilized in the sphere of target monitoring detection. In some situations, a deep neural community is often employed to implement a joint trace detection (monitoring and ItagPro object detection) community, where a joint hint detection community refers to a community that is used to achieve goal detection and goal hint collectively. In the existing joint monitoring detection community, the position and motion path accuracy of the predicted shifting goal isn't excessive sufficient. The application supplies a target monitoring method, a target tracking device and electronic tools, which might improve the issues.

In one facet, iTagPro tracker an embodiment of the current software gives a goal tracking methodology, where the method consists of: a first sub-community in a joint monitoring detection network is used for extracting a first characteristic image from a goal characteristic image, and a second sub-community within the joint monitoring detection community is used for extracting a second characteristic image from the target characteristic image, iTagPro smart tracker whereby the target characteristic image is extracted from a video body of a goal video; fusing the second feature map extracted by the second sub-community to the first function map to acquire a fused feature map corresponding to the primary sub-community; buying first prediction data output by a primary sub-community according to the fusion characteristic map, and buying second prediction info output by a second sub-network; based on the first prediction info and the second prediction information, determining the present position and the movement trail of the transferring goal within the target video.

Optionally, in the tactic provided by the embodiment of the current software, the primary subnetwork is a classification subnetwork, and the second subnetwork is a regression subnetwork or a monitoring subnetwork. Optionally, in the strategy supplied by the embodiment of the present utility, the first subnetwork is a regression subnetwork, and the second subnetwork is a classification subnetwork or a tracking subnetwork. In one other side, an embodiment of the present application gives a goal tracking apparatus, together with: the system contains a feature acquisition module, a characteristic fusion module, a prediction module and a monitoring module. The feature acquisition module is used for detecting a primary sub-community in the community by way of joint tracking, extracting a primary characteristic map from a goal function map, and extracting a second characteristic map from the goal characteristic map by means of a second sub-network within the network by way of joint tracking, wherein the goal feature map is a function map extracted from a video frame of a goal video.