iGYM: Implementation of Image Recognition Using Silhouette Extraction and Artificial Neural Network as Gym Instructor
Iluminada Vivien R. Domingo, Christian James Sunga, Miguell Comia · 2022
The researchers aimed at creating an App-Based Gym Workout Instructor using Image Recognition via Artificial Neural Network which can recognize the body type of a male person using images and show the workout for the body type. The input is a whole-body image of a male person and the output is the workout for the detected body type. Using MATLAB, the researchers created an Artificial Neural Network that is trained to recognize body types and C# platform to implement the ANN. The results of the study showed that the developed system was able to determine the body type of the user. In terms of the over-all accuracy of the developed igym instructor for all of the body type defined, it was fairly moderate with an average of 64.38%. The effectivity and accuracy of the iGYM does not only depend on the number of training data but also with the quality of the data set.