CN111652181B - Target Tracking Method and Device And Electronic Equipment - Google Patents > 자유게시판

Legal status (The authorized status is an assumption and ItagPro 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 appliance discloses a target monitoring methodology, iTagPro locator a goal tracking device and digital gear, and relates to the technical subject of synthetic intelligence. The tactic comprises the following steps: iTagPro features a primary sub-community within the joint tracking detection network, a first feature map extracted from the target characteristic map, and a second feature map extracted from the target feature map by a second sub-community in the joint tracking detection community; fusing the second feature map extracted by the second sub-community to the primary characteristic map to obtain a fused function map corresponding to the primary sub-community; acquiring first prediction info output by a primary sub-network based on a fusion characteristic map, and acquiring second prediction info output by a second sub-network; and determining the present position and the motion trail of the shifting goal within the target video based mostly on the first prediction data and the second prediction data.

The relevance among all the sub-networks which are parallel to each other will be enhanced by means of characteristic fusion, and the accuracy of the determined place and iTagPro locator movement path of the operation target is improved. The current application pertains to the field of artificial intelligence, and in particular, to a target tracking methodology, apparatus, iTagPro locator and digital system. Lately, iTagPro locator artificial intelligence (Artificial Intelligence, AI) expertise has been widely used in the sphere of target tracking detection. In some scenarios, a deep neural community is usually employed to implement a joint trace detection (monitoring and object detection) network, the place a joint hint detection community refers to a community that's used to achieve goal detection and goal hint collectively. In the present joint monitoring detection community, iTagPro locator the position and motion trail accuracy of the predicted shifting goal will not be excessive sufficient. The application gives a goal monitoring method, iTagPro technology a goal tracking device and digital gear, which can enhance the issues.

In one facet, an embodiment of the present utility supplies a goal monitoring methodology, where the tactic contains: a first sub-community in a joint monitoring detection network is used for iTagPro locator extracting a first characteristic picture from a goal characteristic image, and a second sub-network in the joint monitoring detection network is used for extracting a second characteristic image from the goal characteristic image, wherein the goal characteristic image is extracted from a video frame of a goal video; fusing the second function map extracted by the second sub-network to the first function map to acquire a fused characteristic map corresponding to the first sub-community; buying first prediction information output by a first sub-network in response to the fusion function map, and buying second prediction data output by a second sub-community; based on the primary prediction data and iTagPro support the second prediction info, figuring out the present position and the movement path of the shifting goal within the target video.

Optionally, in the strategy offered by the embodiment of the present software, the first subnetwork is a classification subnetwork, and the second subnetwork is a regression subnetwork or a monitoring subnetwork. Optionally, in the strategy provided by the embodiment of the current application, the primary subnetwork is a regression subnetwork, and the second subnetwork is a classification subnetwork or a monitoring subnetwork. In one other aspect, iTagPro smart tracker an embodiment of the current software gives a target tracking apparatus, including: the gadget contains a function acquisition module, a feature fusion module, a prediction module and a tracking module. The feature acquisition module is used for detecting a primary sub-network in the community by joint tracking, extracting a first function map from a goal function map, and extracting a second function map from the goal function map by a second sub-community in the community via joint monitoring, wherein the goal function map is a characteristic map extracted from a video body of a target video.