Abstract:Bodybuilding operation is a favorite sports item, which is beneficial to enhance physical fitness, improve coordination and flexibility of movement, and enhance cardiopulmonary function. Standard aerobics movements and nutrition matching strategies can more effectively enhance the exercise effect. In this paper, the characteristics of aerobics, sports, and nutrition strategies are researched through the human body recognition technology of the depth image; the subjects were divided into four groups by statistical method and control variable experiment method. They were the control group, the experimental group for aerobics training, the experimental group for improving the nutritional balance under diet, and the experimental group for aerobics training. After two and a half months of experimental training and observation, the students in the four groups were tested for physical fitness and physical function; analyze the obtained experimental test data using the human body recognition technology of depth image, then obtain the experimental test data, and then use the data fusion method to combine the data and information for more accurate evaluation. The results showed that after 10 weeks of aerobics learning and nutrition, the average height of students increased by 1.19 cm, the average weight decreased by 1.17 kg, the number of sit-ups increased from 23.9 before the experiment to 31.2, and the results of 50-meter race and 800-meter run were 0.2 seconds and 2.9 seconds, respectively. It can be concluded that aerobics and nutrition can speed up metabolism, promote the growth and development of bones, and supplement the nutrition needed by the human body, to improve students’ physical quality. This study contributes to the research of sports and nutrition matching in improving physical conditions.

REAL-TIME MOTION IMAGE POSE DECOMPOSITION, CLASSIFICATION AND ANALYSIS

Posture Tracking Meets Fitness Coaching: A Two-Phase Optimization Approach with Wearable Devices

Healthy Human Sitting Posture Estimation in RGB-D Scenes Using Object Context

Fitness Done Right: a Real-time Intelligent Personal Trainer for Exercise Correction

AI Coach: Deep Human Pose Estimation and Analysis for Personalized Athletic Training Assistance

Pose Trainer: Correcting Exercise Posture using Pose Estimation

Multiview Human Pose Estimation and Corrective Feedback

Human posture estimation and action recognition on fitness behavior and fitness

Low-Budget Real-time Pose Estimation Method for Personal Physical Exercise Assistant

Characteristics of Aerobics and Nutrition Strategy Based on Depth Image Human Body Recognition

Real-time AI-assisted visual exercise pose correctness during rehabilitation training for musculoskeletal disorder

Single-Stage Pose Estimation and Joint Angle Extraction Method for Moving Human Body

A Machine Learning Method for Identification of Key Body Poses in Cyclic Physical Exercises

A real-time algorithm for weight training detection and correction

Video based exercise recognition and correct pose detection

Analysis of Artificial Intelligence Technology and Its Application in Improving the Effectiveness of Physical Education Teaching

Deep Learning for Fitness

Pose Detection Using OpenCV and Media Pipe

Image-Based Fitness Yoga Pose Recognition: Using Ensemble Learning and Multi-head Attention

Analysis of Students' Sports Exercise Behavior and Health Education Strategy Using Visual Perception–Motion Recognition Algorithm

Towards Automating Personal Exercise Assessment and Guidance with Affordable Mobile Technology