SKRIPSI IF

DETEKSI EKSPRESI WAJAH PADA CITRA REKAMAN VIDEO DENGAN PENDEKATAN TRANSFER LEARNING BERBASIS MOBILENET

---

ABSTRAK

Emosi berperan penting dalam komunikasi manusia, dan ekspresi wajah merupakan indikator utama dalam mengidentifikasi kondisi emosional. Mayoritas penelitian Facial Expression Recognition (FER) masih berfokus pada citra statis atau deteksi real-time berbasis webcam, sementara pendekatan evaluatif berbasis rekaman video masih jarang dikembangkan. Penelitian ini bertujuan merancang pipeline sederhana namun fungsional untuk mengevaluasi model MobileNetV2 berbasis transfer learning pada data citra video rekaman. Dataset Karolinska Directed Emotional Faces (KDEF) digunakan untuk pelatihan dengan tujuh kelas emosi dasar, sedangkan data uji berupa rekaman video diproses frame-by-frame melalui tahap ekstraksi, deteksi wajah dengan Haar Cascade, preprocessing, dan klasifikasi menggunakan MobileNetV2 yang telah di fine-tune. Evaluasi dilakukan menggunakan akurasi, precision, recall, dan F1-score. Hasil penelitian menunjukkan model mencapai akurasi validasi sebesar 87% dan mampu mengidentifikasi ekspresi dominan pada video, meskipun masih terdapat bias prediksi terhadap kelas neutral pada ekspresi subtil seperti angry dan disgust. Sebaliknya, ekspresi dengan ciri visual lebih jelas seperti happy dapat dikenali lebih baik. Kesimpulannya, pipeline ini berhasil menjembatani model citra statis dengan data rekaman video, menawarkan pendekatan evaluasi yang sederhana, efisien, dan relevan untuk aplikasi Human-Computer Interaction (HCI) berbasis perangkat terbatas.
Kata Kunci: Deteksi Ekspresi Wajah, MobileNetV2, Transfer Learning, Interaksi Verbal, Video Analysis

ABSTRACT


Emotions play an important role in human communication, and facial expressions are one of the main indicators for recognizing emotional states. Most studies in Facial Expression Recognition (FER) still focus on static images or real-time webcam tracking, while evaluation approaches based on recorded video remain less explored. This study aims to design a simple but functional pipeline to evaluate the performance of MobileNetV2 with transfer learning on verbal interaction video data. The Karolinska Directed Emotional Faces (KDEF) dataset was used for training with seven basic emotion classes, while the test data came from video recordings processed frame-by-frame. The pipeline includes frame extraction, face detection using Haar Cascade, image preprocessing, and classification with the fine-tuned MobileNetV2 model. Evaluation metrics such as accuracy, precision, recall, and F1-score were applied. The results show that the model reached 87% validation accuracy and was able to identify dominant emotions in video, although predictions tended to be biased toward the neutral class in subtle expressions such as angry and disgust. On the other hand, clearer expressions such as happy were detected more reliably. In conclusion, the proposed pipeline successfully bridges static-image models with video data, offering a practical and efficient evaluation approach that can support Human-Computer Interaction (HCI) applications on resource-limited devices.

Keywords: Facial Expression Recognition, MobileNetV2, Transfer Learning, Verbal Interaction, Video Analysis

---

Ketersediaan

S250925002

607.2 MAU d

Perpustakaan STMIK AMIKBANDUNG

Tersedia

Informasi Detil

Judul Seri

-

No. Panggil

607.2 MAU d

Penerbit

: ., 2025

Deskripsi Fisik

-

Bahasa

Indonesia

ISBN/ISSN

-

Klasifikasi

NONE

Tipe Isi

-

Tipe Media

-

Tipe Pembawa

-

Edisi

-

Subyek

-

Info Detil Spesifik

-

Pernyataan Tanggungjawab

-

Versi lain/terkait

Lampiran Berkas

Komentar

---

You must be logged in to post a comment