Weakly Supervised Virus Capsid Detection with Image-Level Annotations in Electron Microscopy Images

2025-08-01

AI Summary

Weakly supervised object detection for viral capsids in electron microscopy images faces high annotation costs due to reliance on expert-drawn bounding boxes. To address this, this paper proposes a virus-particle-oriented weakly supervised detection framework that operates solely with image-level labels. It leverages knowledge distillation from a pre-trained model to generate high-quality pseudo-bounding boxes and progressively refines localization via a receptive field contraction strategy—without modifying the backbone architecture. The core contribution lies in embedding domain-specific prior knowledge into the pseudo-label generation process, substantially improving weakly supervised localization accuracy. Experiments demonstrate that the proposed method achieves state-of-the-art performance across multiple weakly supervised benchmarks. Notably, under stringent annotation time constraints, it even surpasses fully supervised baselines trained with ground-truth bounding boxes in detection accuracy.

Technology Category

• Computer Vision: Object Detection & Categorization
• Machine Learning: Semi-Supervised Learning
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs

Abstract

Current state-of-the-art methods for object detection rely on annotated bounding boxes of large data sets for training. However, obtaining such annotations is expensive and can require up to hundreds of hours of manual labor. This poses a challenge, especially since such annotations can only be provided by experts, as they require knowledge about the scientific domain. To tackle this challenge, we propose a domain-specific weakly supervised object detection algorithm that only relies on image-level annotations, which are significantly easier to acquire. Our method distills the knowledge of a pre-trained model, on the task of predicting the presence or absence of a virus in an image, to obtain a set of pseudo-labels that can be used to later train a state-of-the-art object detection model. To do so, we use an optimization approach with a shrinking receptive field to extract virus particles directly without specific network architectures. Through a set of extensive studies, we show how the proposed pseudo-labels are easier to obtain, and, more importantly, are able to outperform other existing weak labeling methods, and even ground truth labels, in cases where the time to obtain the annotation is limited.

Problem

• Detect virus capsids using weak image-level annotations
• Reduce reliance on costly expert bounding box annotations
• Improve detection accuracy with pseudo-labels from pre-trained models

Innovation

• Weakly supervised detection with image-level annotations
• Knowledge distillation from pre-trained model
• Optimization with shrinking receptive field