A method for hand lifter detection based on object detection and attitude estimation

Abstract

The invention relates to a hand raiser detection method based on object detection and attitude estimation. The method comprises the following steps: 1) obtaining a teaching video to be tested; 2) evenly extract picture frames in that teach video to be tested, inputting the trained hand-raising motion detection model, and obtain a picture frame containing the hand-raising motion and a first text file for recording the position of the hand-raising frame; 3) carry out pose estimation on that picture frame containing the hand-raising action to obtain the key point of the human body of all personsin each picture frame and forming a second text file for recording the position of the key point; 4) detect that obtained hand raise according to the first text file and the second text file by usinga heuristic matching strategy. Compared with the prior art, the invention solves the problems of low resolution and motion distortion by improving the attitude estimation algorithm, adopts the heuristic matching strategy to accurately obtain the real hand raiser, and has the advantages of high detection accuracy and detection rate.

Description

technical field [0001] The invention relates to a video intelligent detection method, in particular to a hand-raiser detection method based on object detection and attitude estimation. Background technique [0002] Object detection and pose estimation have also flourished in recent years with the rapid rise of artificial intelligence and deep neural networks. Thanks to a large number of image data sets and computing resources, many excellent object recognition algorithms have emerged, including Fast R-CNN, Faster R-CNN, YOLO, SSD and R-FCN. At the same time, by detecting key points of the human body The pose estimation algorithm has also made a breakthrough, and the pose estimation algorithm is usually used in multi-person scenes. [0003] The general multi-person pose estimation algorithm mainly includes two categories of methods, Top-down (top-down): first detect multiple people, and then perform single-person pose estimation for each person, Bottom-up (bottom-up): first ...

AI Technical Summary

Problems solved by technology

Low resolution will cause key point detection errors and missing situations, which may lead to errors in final action detection, such as figure 1 as shown in (1a)

Actions such as raising hands may bring about very exaggerated human body postures, such as raising the arm high to cause a disproportionate body length, raising the arm on one side of the body and then deflecting it to the other side of the body, and raising the arm in other directions to make the human body disproportionate. Instead of raising the arm vertically upwards, as shown in Figure (1b), these rare gestures not only bring difficulties to key point detection, but also cause new problems for the subsequent matching of the hand raiser

[0005] All in all, the existing original multi-person pose estimation algorithm faces the above two types of problems, and the detection effect is not ideal, and there will be many missed and false detections about the key points of the human body, which brings great difficulties to the subsequent matching of the hand-raiser. big difficulty

Images