Characterizing Face Recognition for Resource Efficient Deployment on Edge

Ayan Biswas

Abstract

Deployment of Face Recognition (FR) systems on edge devices has seen significant growth, driven by advances in hardware and neural architectures. However, tailoring State-of-the-art (SOTA) FR solu- tions to specific edge environments remains difficult and largely unexplored. Although benchmark data exists for some combinations of model, hardware, and frameworks, it does not scale to the evaluation of custom or modified architectures. Furthermore, theoretical metrics such as Floating-Point Operations (FLOPs) are often poor predictors of real-world performance, as edge inference throughput is frequently constrained by system-level factors including data movement, memory access, and pipeline overheads.

This limitation arises from a combinatorial design space spanning architectural variants, input res- olutions, deployment frameworks, and hardware configurations, making exhaustive benchmarking in- feasible. An analysis of recent FR literature performed in this thesis reveals that while substantial innovation is concentrated in loss functions, training strategies, and data augmentation, the underly- ing Convolutional Neural Network (CNN) backbone is routinely treated as a fixed abstraction. These backbones are typically inherited from large-scale image classification, where models are optimized to separate visually distinct categories, whereas FR requires learning discriminative embeddings that dis- tinguish between highly similar identities. Consequently, the exploration of lightweight, task-specific architectures for edge deployment remains limited.

To address this gap, this thesis models the relationship between network architecture and inference throughput in an edge deployment setting, demonstrating that throughput follows a predictable structure when grouped by architectural families. Building on this modeling, this work proposes a data-efficient, architecture-aware framework to estimate the throughput of custom FR models from a sparse set of empirical measurements, with predictions that are explainable and exhibit errors small enough to be treated as standard on-device observational noise.

Related Publications

Downloads