Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Chu, W. I., Tarroni, G., Li, L.
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2603.10833
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917333065269248
author Chu, W. I.
Tarroni, G.
Li, L.
author_facet Chu, W. I.
Tarroni, G.
Li, L.
contents Adverse drug events are a significant source of preventable harm, which has led to the development of automated pill recognition systems to enhance medication safety. Real-world deployment of these systems is hindered by visually complex conditions, including cluttered scenes, overlapping pills, reflections, and diverse acquisition environments. This study investigates few-shot pill recognition from a deployment-oriented perspective, prioritizing generalization under realistic cross-dataset domain shifts over architectural innovation. A two-stage object detection framework is employed, involving base training followed by few-shot fine-tuning. Models are adapted to novel pill classes using one, five, or ten labeled examples per class and are evaluated on a separate deployment dataset featuring multi-object, cluttered scenes. The evaluation focuses on classification-centric and error-based metrics to address heterogeneous annotation strategies. Findings indicate that semantic pill recognition adapts rapidly with few-shot supervision, with classification performance reaching saturation even with a single labeled example. However, stress testing under overlapping and occluded conditions demonstrates a marked decline in localization and recall, despite robust semantic classification. Models trained on visually realistic, multi-pill data consistently exhibit greater robustness in low-shot scenarios, underscoring the importance of training data realism and the diagnostic utility of few-shot fine-tuning for deployment readiness.
format Preprint
id arxiv_https___arxiv_org_abs_2603_10833
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Evaluating Few-Shot Pill Recognition Under Visual Domain Shift
Chu, W. I.
Tarroni, G.
Li, L.
Computer Vision and Pattern Recognition
Adverse drug events are a significant source of preventable harm, which has led to the development of automated pill recognition systems to enhance medication safety. Real-world deployment of these systems is hindered by visually complex conditions, including cluttered scenes, overlapping pills, reflections, and diverse acquisition environments. This study investigates few-shot pill recognition from a deployment-oriented perspective, prioritizing generalization under realistic cross-dataset domain shifts over architectural innovation. A two-stage object detection framework is employed, involving base training followed by few-shot fine-tuning. Models are adapted to novel pill classes using one, five, or ten labeled examples per class and are evaluated on a separate deployment dataset featuring multi-object, cluttered scenes. The evaluation focuses on classification-centric and error-based metrics to address heterogeneous annotation strategies. Findings indicate that semantic pill recognition adapts rapidly with few-shot supervision, with classification performance reaching saturation even with a single labeled example. However, stress testing under overlapping and occluded conditions demonstrates a marked decline in localization and recall, despite robust semantic classification. Models trained on visually realistic, multi-pill data consistently exhibit greater robustness in low-shot scenarios, underscoring the importance of training data realism and the diagnostic utility of few-shot fine-tuning for deployment readiness.
title Evaluating Few-Shot Pill Recognition Under Visual Domain Shift
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2603.10833