Scene adaptive target detection method based on motion foreground

Abstract

With the increasing development of the deep learning technology, the requirement on the generalization performance of the model in a real environment is increased increasingly, and the influence of the differences of illumination, background and the like on the generalization performance of the model has aroused wide attention. The invention discloses a scene adaptive target detection method based on a motion foreground. According to the method, the target frame of the motion foreground is effectively utilized by utilizing the priori of the distribution consistency of the motion foreground and the global target data, and meanwhile, the instance feature similarity is calculated through the decoder, so that the effect of the model in the target domain is greatly improved. Experimental results show that the target detection effect of the method provided by the invention is greatly improved in a real environment.

Description

technical field [0001] The invention relates to a scene adaptive target detection method based on moving foreground. Background technique [0002] In the field of computer vision, object detection is an important topic. His task is to find the region of interest in the image and video, and determine its category and location. At present, many methods based on deep learning can achieve good results on benchmark datasets. However, due to the existence of domain differences, that is, when the target size, camera angle, illumination, and background environment change, the effect of the model will decline to varying degrees. . The simplest and most effective way to solve this problem when training models on the same domain is data-driven training. However, on the one hand, manual labeling of datasets costs a lot of manpower and material resources. On the other hand, many practical fields cannot be manually labeled. Therefore, in order to solve the degradation of model generaliz...

AI Technical Summary

Problems solved by technology

[0004] However, the above algorithm only considers the domain difference in classification, and does not consider the domain difference in regression, resulting in unsatisfactory effects after scene changes.

In addition, for images with large domain differences, due to the unknown data distribution, it is impossible to effectively extract suitable candidate frame targets in the stage of extracting candidate regions RPN in the first stage of the two-stage target detection, and it is also impossible to distinguish which features of the regions need to be aligned during feature alignment.

Images