HiProto: Hierarchical Prototype Learning for Interpretable Object Detection Under Low-quality Conditions

2026-04-15

AI Summary

This work addresses the challenge of object detection under low-quality imaging conditions, where visual degradation and prediction uncertainty undermine both interpretability and semantic discriminability. To tackle this issue, the authors propose HiProto, the first approach to introduce hierarchical prototype learning into this domain. HiProto models class semantics through multi-level prototypes and integrates a region-to-prototype contrastive loss (RPC-Loss), a prototype regularization loss (PR-Loss), and a scale-aware pseudo-label generation strategy (SPLGS). Notably, it achieves substantial improvements in detection performance and interpretability without relying on image enhancement or complex architectural modifications. Experiments demonstrate that HiProto excels on the ExDark, RTTS, and VOC2012-FOG benchmarks, with prototype responses offering clear semantic explanations for predictions.

Technology Category

• Computer Vision: Object Detection & Categorization
• Machine Learning: Multi-instance/Multi-view 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
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Interpretability is essential for deploying object detection systems in critical applications, especially under low-quality imaging conditions that degrade visual information and increase prediction uncertainty. Existing methods either enhance image quality or design complex architectures, but often lack interpretability and fail to improve semantic discrimination. In contrast, prototype learning enables interpretable modeling by associating features with class-centered semantics, which can provide more stable and interpretable representations under degradation. Motivated by this, we propose HiProto, a new paradigm for interpretable object detection based on hierarchical prototype learning. By constructing structured prototype representations across multiple feature levels, HiProto effectively models class-specific semantics, thereby enhancing both semantic discrimination and interpretability. Building upon prototype modeling, we first propose a Region-to-Prototype Contrastive Loss (RPC-Loss) to enhance the semantic focus of prototypes on target regions. Then, we propose a Prototype Regularization Loss (PR-Loss) to improve the distinctiveness among class prototypes. Finally, we propose a Scale-aware Pseudo Label Generation Strategy (SPLGS) to suppress mismatched supervision for RPC-Loss, thereby preserving the robustness of low-level prototype representations. Experiments on ExDark, RTTS, and VOC2012-FOG demonstrate that HiProto achieves competitive results while offering clear interpretability through prototype responses, without relying on image enhancement or complex architectures. Our code will be available at https://github.com/xjlDestiny/HiProto.git.

Problem

• interpretable object detection
• low-quality conditions
• prototype learning
• semantic discrimination
• prediction uncertainty

Innovation

• Hierarchical Prototype Learning
• Interpretable Object Detection
• Region-to-Prototype Contrastive Loss
• Prototype Regularization
• Scale-aware Pseudo Labeling