The Effect of Hyperparameters on Faster R-CNN inFace Recognition Systems

The Effect of Hyperparameters on Faster R-CNN inFace Recognition Systems

face recognition;faster R-CNN;hyperparameter optimization;deep learning;grid search

Facial recognition remains a significant challenge in the advancement of computer vision technologies. This research seeks todevelop a facial recognition system utilizing the Faster R-CNN architecture, with performance enhancement achieved through hyperparameter optimization. This research utilizes the "Face Recognition Dataset" from Kaggle, which comprises 2,564 face images across 31 classes. The development process involves creating bounding boxes using the LabelImg application and implementing the Grid Search method. The Grid Search is applied with predefined hyperparameter combinations (3 epochs [10, 25, and 50] × 3 learning rates [0.001, 0.0001, and 0.00001] × 3 optimizers [SGD, Adam, and RMS], resulting in 27 models). The evaluation of the model was conducted using accuracy, precision, recall, and F1-score as performance metrics. The experimental findings indicate that hyperparameter selection has a substantial impact on model performance. Among the tested configurations, the combination of a learning rate of 0.00001, 50 training epochs, and the Adam optimizer achieved thehighest accuracy, resulting in an 8.33% improvement over the baseline model. The results indicate that hyperparameter optimization enhances the ability of the model to recognize faces. Compared to conventional models, theFaster R-CNN performs better in detecting faces more accurately. Future research could further enhance the face recognition efficiency andaccuracy by exploring other deep learning architectures and more advanced hyperparameter optimization techniques

Jasman Pardede1*, Khairul Rijal2

https://jurnal.iaii.or.id/index.php/RESTI/article/view/6405/1061

Department of Informatics, Faculty of Industrial Technology, Institut Teknologi Nasional, Bandung, Indonesia

May, 28, 2025

FAJAR BAGUS W

PDF

ENGLISH

TEXT