Target tracking method based on difficult positive sample generation

Abstract

The invention discloses a target tracking method based on difficult positive sample generation. According to the method, for each video in training data, a variation auto-encoder is utilized to learna corresponding flow pattern, namely a positive sample generation network, codes are slightly adjusted according to an input image obtained after encoding, and a large quantity of positive samples aregenerated; the positive samples are input into a difficult positive sample conversion network, an intelligent body is trained to learn to shelter a target object through one background image block, the intelligent body performs bounding box adjustment continuously, so that the samples are difficult to recognize, the purpose of difficult positive sample generation is achieved, and sheltered difficult positive samples are output; and based on the generated difficult positive samples, a twin network is trained and used for matching between a target image block and candidate image blocks, and positioning of a target in a current frame is completed till processing of the whole video is completed. According to the target tracking method based on difficult positive sample generation, the flow pattern distribution of the target is learnt directly from the data, and a large quantity of diversified positive samples can be obtained.

Description

technical field [0001] The invention relates to a visual tracking technology, in particular to a target tracking method based on generation of difficult positive samples. Background technique [0002] At present, mainstream deep learning method tracking usually includes the following steps: the first step is to collect a large number of hand-labeled videos; the second step is to perform dense sampling of positive and negative samples near the first frame label box on each video; The third step is to use the samples sampled in the previous step to train a binary classifier; the fourth step is to confirm the candidate area near the search box, perform classification, and select the area with the highest score as the tracking result; the fifth step is to repeat the above steps until the video Finish. [0003] The deficiencies in the prior art are: figure 1 As shown, the sample diversity obtained by the existing dense sampling method is insufficient; there are few difficult sa...

AI Technical Summary

Problems solved by technology

[0005] The technical problem to be solved by the present invention is: the sample diversity obtained by the conventional dense sampling method is insufficient and the difficult samples are few, and the model is too sensitive to the challenge factors, and a target tracking method based on the generation of difficult positive samples is provided

Images