A Survey on Class-Agnostic Counting: Advancements from Reference-Based to Open-World Text-Guided Approaches

2025-01-31

AI Summary

Existing counting methods rely on predefined categories and extensive annotated data, limiting generalization to unseen objects or language-driven scenarios. This paper presents a systematic survey of class-agnostic counting (CAC), introducing the first methodological taxonomy and proposing text-guided counting as a novel open-vocabulary paradigm. We unify modeling across three paradigms—reference-guided, reference-free, and text-guided—by integrating visual-language alignment, few-shot adaptation, intrinsic image pattern mining, and example-guided mechanisms. A standardized evaluation is conducted across multiple benchmarks. Our analysis identifies fundamental bottlenecks: annotation dependency, limited generalizability, and poor scalability. We rigorously characterize performance boundaries across paradigms, revealing trade-offs among accuracy, flexibility, and efficiency. This work establishes the first theoretical framework and practical guideline for open-world intelligent counting, enabling robust, language-grounded, and category-agnostic object enumeration.

Technology Category

• Computer Vision: Interpretability, Explainability, and Transparency
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

Object counting has recently shifted towards class-agnostic counting (CAC), which addresses the challenge of counting objects across arbitrary categories, tackling a critical need in versatile counting systems. While humans effortlessly identify and count objects from diverse categories without prior knowledge, most counting methods remain restricted to enumerating instances of known classes, requiring extensive labeled datasets for training, and struggling under open-vocabulary settings. Conversely, CAC aims to count objects belonging to classes never seen during training, typically operating in a few-shot setting. In this paper, for the first time, we review advancements in CAC methodologies, categorizing them into three paradigms based on how target object classes can be specified: reference-based, reference-less, and open-world text-guided. Reference-based approaches have set performance benchmarks using exemplar-guided mechanisms. Reference-less methods eliminate exemplar dependency by leveraging inherent image patterns. Finally, open-world text-guided methods utilize vision-language models, enabling object class descriptions through textual prompts, representing a flexible and appealing solution. We analyze state-of-the-art techniques and we report their results on existing gold standard benchmarks, comparing their performance and identifying and discussing their strengths and limitations. Persistent challenges -- such as annotation dependency, scalability, and generalization -- are discussed, alongside future directions. We believe this survey serves as a valuable resource for researchers to understand the progressive developments and contributions over time and the current state-of-the-art of CAC, suggesting insights for future directions and challenges to be addressed.

Problem

• Computer Vision
• Object Counting
• Generalization to Novel Categories

Innovation

• Category-Agnostic Counting
• Textual Description Recognition
• Enhanced Generalization