Ef-QuantFace: Streamlined Face Recognition with Small Data and Low-Bit Precision

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Gazali, William, Kho, Jocelyn Michelle, Santoso, Joshua, Williem
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866913246023254016
author Gazali, William
Kho, Jocelyn Michelle
Santoso, Joshua
Williem
author_facet Gazali, William
Kho, Jocelyn Michelle
Santoso, Joshua
Williem
contents In recent years, model quantization for face recognition has gained prominence. Traditionally, compressing models involved vast datasets like the 5.8 million-image MS1M dataset as well as extensive training times, raising the question of whether such data enormity is essential. This paper addresses this by introducing an efficiency-driven approach, fine-tuning the model with just up to 14,000 images, 440 times smaller than MS1M. We demonstrate that effective quantization is achievable with a smaller dataset, presenting a new paradigm. Moreover, we incorporate an evaluation-based metric loss and achieve an outstanding 96.15% accuracy on the IJB-C dataset, establishing a new state-of-the-art compressed model training for face recognition. The subsequent analysis delves into potential applications, emphasizing the transformative power of this approach. This paper advances model quantization by highlighting the efficiency and optimal results with small data and training time.
format Preprint
id arxiv_https___arxiv_org_abs_2402_18163
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Ef-QuantFace: Streamlined Face Recognition with Small Data and Low-Bit Precision
Gazali, William
Kho, Jocelyn Michelle
Santoso, Joshua
Williem
Computer Vision and Pattern Recognition
In recent years, model quantization for face recognition has gained prominence. Traditionally, compressing models involved vast datasets like the 5.8 million-image MS1M dataset as well as extensive training times, raising the question of whether such data enormity is essential. This paper addresses this by introducing an efficiency-driven approach, fine-tuning the model with just up to 14,000 images, 440 times smaller than MS1M. We demonstrate that effective quantization is achievable with a smaller dataset, presenting a new paradigm. Moreover, we incorporate an evaluation-based metric loss and achieve an outstanding 96.15% accuracy on the IJB-C dataset, establishing a new state-of-the-art compressed model training for face recognition. The subsequent analysis delves into potential applications, emphasizing the transformative power of this approach. This paper advances model quantization by highlighting the efficiency and optimal results with small data and training time.
title Ef-QuantFace: Streamlined Face Recognition with Small Data and Low-Bit Precision
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2402.18163