Poster
in
Affinity Event: Black in AI

FacEDiM: A Face Embedding Distribution Model for Few-Shot Biometric Authentication of Cattle

Meshia Cédric Oveneke ⋅ Rucha Vaishampayan ⋅ Deogratias Lukamba Nsadisa ⋅ Rucha Vaishampayan

Keywords: Deep Learning Applications of AI to Health Statistical Reasoning Computer Vision

[ Poster]

Abstract

This work proposes to solve the problem of few-shot biometric authentication by computing the Mahalanobis distance between testing embeddings and a multivariate Gaussian distribution of training embeddings obtained using pre-trained CNNs. Experimental results show that models pre-trained on the ImageNet dataset significantly outperform models pre-trained on human faces. With a VGG16 model, we obtain a FRR of 1.18% for a FAR of 1.25% on a dataset of 20 cattle identities.

Chat is not available.