Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

2 results about "Basic class" patented technology

Library-based so file hooking method, device and medium

The application provides a so file hook method, device and readable medium based on a loading program library, the method comprises the following steps: acquiring all loading path information of a so file and acquiring difference path information; establishing an interface for providing data of a storage path, storing the difference path information; establishing an interface for providing exclusion path data, and encapsulating the interface for providing data of the storage path and the interface for providing exclusion path data through a basic class; acquiring a function name to be hooked and a name of a replacement function; then finding the basic class to parse a so file related to the function to be hooked; judging that the so file related to the function to be hooked is running through a loading program database, and performing a hook operation, so that the so file to be hooked can be quickly found and excluded.
Owner:WUHAN DOUYU NETWORK TECHNOLOGY CO LTD

Weakly supervised temporal action localization method and device based on semantic and saliency knowledge collaborative propagation

The weakly supervised temporal action localization method and device based on semantic and saliency knowledge collaborative propagation comprises the following steps: 1) extracting temporal features and saliency foreground features from an uncropped video; 2) constructing a basic branch and a saliency perception branch to respectively process the temporal features and the saliency target features to obtain a basic class activation sequence, a motion, appearance representation score and a saliency class activation sequence, and weighting and fusing the four sequences to obtain a fused action score sequence; 3) utilizing branch distillation and branch action consistency constraint to interact semantic information and saliency information, and perfecting the fused action score sequence; 4) extracting key segments and ambiguous segments of the basic branch and the saliency perception branch, and utilizing the key segments and the ambiguous segments between branches and within branches for comparative learning to improve feature representation, combining the distillation result to perfect the fused action score sequence and obtaining an action localization result. The present application can perceive subtle human actions and accurate temporal action boundaries in uncropped videos.
Owner:ZHEJIANG UNIV OF TECH