Fast positioning method for video events from rough state to fine state

Abstract

The invention discloses a fast positioning method for video events from rough state to fine state, comprising the following steps of: roughly matching an inquired event with an actual video to obtain the start and ending points for dividing the actual video time, extracting an interesting region of each frame of image in the actual video to finish space division of the actual video, performing space division and time division to the actual video to obtain a series of interesting time bodies, finely matching the interesting time bodies with the inquired event to construct a relative body, using a global significance test to identify whether each interesting time body has the event related to the inquired sample on the relative body, and finally, using a post-process method to control the time region positioning and displaying around the best matched obvious points. The method adopts the roughly matching method to remove large amount of irrelevant time regions to effectively reduce the matching and searching space so that the finely matching is only performed between the interesting time body and the inquired event, so the method improves the searching speed.

Description

technical field [0001] The invention belongs to the research field of video event positioning methods, in particular to a rapid positioning method of video events from coarse to fine. Background technique [0002] The localization of video events has a wide range of applications in video retrieval, video browsing, intelligent monitoring, and human motion analysis. Current video event localization methods are mainly divided into two categories: learning-based methods and learning-free methods. The learning-based method needs to establish a training model for each query event, but because the training of the model needs to adjust multiple parameters, over-fitting may occur. The learning-free method does not require training and only needs the user to provide the corresponding query event to locate the video event, and the space-time location of the query event in the real video can be obtained by searching. This type of method uses query events as templates to perform matchi...

AI Technical Summary

Problems solved by technology

[0005] The above two methods only provide visual experimental results, and do not count the detection rate, so they cannot be compared quantitatively and accurately

In addition, using the above two methods to realize the positioning of the query event in the video must be searched in the complete space of X-Y-T. In order to reduce the huge search space, they both down-sample the original video, which is more likely to cause missed detection, and Significant computational cost is still spent on space-time locations that are unlikely to contain query events

The above technical defects make the existing video event positioning method not meet the practical requirements in terms of performance and time efficiency, which limits the scope of application of this method

Images