Towards Billion-scale Multi-modal Biometric Search

TL;DR

This paper introduces Bharat ABIS, a scalable multi-modal biometric search system capable of efficiently processing and searching over a billion records using fingerprint, face, and iris data, with high accuracy and speed.

Contribution

It presents the first large-scale open-source architecture for multi-modal biometric search at a billion-record scale, including detailed analysis and performance evaluation.

Findings

Achieves 0.3% FNIR at 0.5% FPIR on 220 million identities

Processes 100 searches per second on 40 million gallery with 8 Nvidia H100 GPUs

Outperforms three state-of-the-art commercial off-the-shelf systems

Abstract

Searching a multi-biometric database of a billion records for a country-level identity system requires pushing the limits of all aspects of a biometric system, including acquisition, preprocessing, feature extraction, accuracy, matching speed, presentation attack detection, and handling of special cases (e.g., missing finger digits). This is the first paper that gives insights into such a large-scale multimodal biometric search system, called Bharat ABIS, based on open-source architectures. The end-to-end pipeline of Bharat ABIS processes fingerprint, face and iris modalities through modality-specific stages of preprocessing (segmentation), quality assessment, presentation attack detection, and learning an embedding (feature extraction), producing a concatenated template of 13.5KB per person. We present a detailed analysis of the modalities and how they are integrated to create an efficient and effective solution for 1:N search (de-duplication). Evaluations on a demographically stratified gallery of 220 million identities, randomly sampled from 1.55 billion records in India's Aadhaar database, yield an FNIR of 0.3% at an FPIR of 0.5%, for adult probes (over 18 years). We also compare the performance of Bharat ABIS against three state-of-the-art COTS systems on a 20M gallery. Our system achieves a throughput of 100 searches per second on a gallery of 40M on a single server (8xNvidia H100 GPUs, 2TB RAM).