Gymnasium body-building movement identification method and device based on depth learning

Abstract

The invention relates to a gymnasium body-building movement identification method and device based on depth learning, takes real-time videos as media and employs a method based on depth learning to carry out movement normalization identification. The method comprises steps that (1), data acquisition, normative movements are recorded as normative movement images; (2), data annotation, normative movement classification of the normative movement images is carried out; (3), data training, an object detection identification framework based on the convolutional neural network of the depth learning method Caffe is employed to acquire a normative movement identification model; (4), movement identification, user shooting is carried out, user movements in line with normative movement classificationof the normative movement identification model are identified; and (5), movement scoring, normative similarity scores and correction plans are outputted. The method is advantaged in that a user has noneed to wear or replace expensive equipment, cost is controllable, videos which are popular in intelligent terminals at present and are convenient to use are taken as media, precise movement data isgenerated for the user, false movements are corrected, and the better body-building effect is realized.

Description

technical field [0001] The present invention relates to the technical field of image recognition, in particular to a deep learning-based gym fitness action recognition method and device. Background technique [0002] The standardization of movements is especially important in fitness activities, and how to exercise correctly has become a crucial issue. Irregular movements will result in half the effort of fitness activities, making the fitness effect asymmetrical to expectations. In addition, there is a risk of sports injuries due to irregular movements, and hiring a fitness coach is also a big expense for many people. [0003] In order to solve this problem, relatively high-end fitness systems have begun to try motion recognition, but most of them are assisted by wearable devices, such as sports bracelets, shoes, clothes, armbands, glasses, mobile phones, etc., which require active participation of users, which is cumbersome . There are other non-video sensors installed o...

AI Technical Summary

Problems solved by technology

Irregular movements will lead to half the effort of fitness activities, making the fitness effect asymmetrical to expectations. In addition, there is a risk of sports injuries due to irregular movements, and hiring a fitness coach is also a big expense for many people.

[0003] In order to solve this problem, relatively high-end fitness systems have begun to try motion recognition, but most of them are assisted by wearable devices, such as sports bracelets, shoes, clothes, armbands, glasses, mobile phones, etc., which require active participation of users, which is cumbersome

There are other non-video sensors mounted on fitness equipment that are highly coupled and very costly

At present, most of the mainstream gyms are traditional equipment. If you want to replace them with the above equipment, the price will be expensive, far exceeding the budget of ordinary users.

Images