LFD: Enabling Real-World Lensless Face Recognition with a Large-Scale Dataset

2026-07-10

AI Summary

This work addresses the limitations of conventional lens-based facial recognition systems, which rely on bulky, costly hardware and pose significant privacy concerns, while existing datasets fail to capture the realistic artifacts and environmental variations inherent in lensless imaging. To bridge this gap, the authors introduce LFD, a large-scale lensless face dataset comprising 21,080 samples collected across three distinct lensless camera prototypes under diverse conditions—including varying illumination, viewpoints, distances, and natural outdoor scenes—each paired with raw measurements, reconstructed images, and reference ground-truth images. LFD is the first dataset to enable authentic lensless face recognition research in real-world settings, effectively capturing cross-device commonalities in imaging artifacts and facial features. It establishes a reliable benchmark for algorithm development and evaluation, and facilitates exploration of cross-device generalization capabilities in lensless vision systems.

Abstract

Face recognition is a ubiquitously used computer vision task that has a wide range of applications ranging from everyday smartphone biometrics to high-stakes security systems. Most face recognition systems rely on traditional cameras, which often suffer from limitations such as bulky form factors, high costs, and limited privacy protection. To address these limitations, lensless cameras have emerged as an alternative. Lensless cameras use thin optical encoders, enabling smaller size, lower cost, and greater design flexibility. These cameras are typically paired with reconstruction algorithms that convert raw captures into recognizable images. However, reconstructed images often contain artifacts, and the reconstruction methods struggle to generalize well to real-world conditions. Furthermore, existing face datasets do not account for the artifacts present in lensless images. To address this issue, we introduce the Lensless Face Dataset (LFD). LFD comprises 21,080 lensless raw measurements, reconstructions, and standard images of faces captured under diverse lighting, angle, and distance. Our key contributions are: (1) Real-world lensless face data: LFD focuses on capturing a diverse face dataset with varying levels of artifacts introduced under different environments; (2) In-the-wild captures: 4,976 images are captured in outdoor settings with varying intensities of natural light and different background patterns; (3) Multiple lensless devices: LFD includes face images collected from three different types of lensless cameras, each with a unique optical encoder. We use this hardware diversity to demonstrate generalization across different lensless cameras. Through comprehensive evaluations and analysis, we show that LFD effectively captures shared features and artifacts across different lensless imaging devices, making it a valuable dataset for advancing lensless face recognition.

Problem

• lensless imaging
• face recognition
• dataset
• image artifacts
• real-world conditions

Innovation

• lensless imaging
• face recognition
• large-scale dataset
• in-the-wild capture
• optical encoder