Generalized-Scale Object Counting with Gradual Query Aggregation

2025-11-11

AI Summary

To address the challenges of multi-scale localization—particularly for dense, small objects—and poor generalization in few-shot object counting, this paper proposes GECO2, an end-to-end framework. Its core innovation is a progressive query aggregation mechanism that generates high-resolution, dense queries to jointly model geometric and semantic features of both large and small objects. By integrating cross-scale feature fusion with a lightweight query generation strategy, GECO2 achieves high detection accuracy while significantly improving computational efficiency. On standard few-shot counting benchmarks, GECO2 outperforms state-of-the-art methods by 10% in both counting and detection accuracy, accelerates inference speed by 3×, and reduces GPU memory consumption by 27%.

Technology Category

• Computer Vision: Object Detection & Categorization
• Search and Optimization: Learning to Search
• Machine Learning: Multi-instance/Multi-view Learning

Application Category

• Security and Privacy: Large-scale security measurements
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs

Abstract

Few-shot detection-based counters estimate the number of instances in the image specified only by a few test-time exemplars. A common approach to localize objects across multiple sizes is to merge backbone features of different resolutions. Furthermore, to enable small object detection in densely populated regions, the input image is commonly upsampled and tiling is applied to cope with the increased computational and memory requirements. Because of these ad-hoc solutions, existing counters struggle with images containing diverse-sized objects and densely populated regions of small objects. We propose GECO2, an end-to-end few-shot counting and detection method that explicitly addresses the object scale issues. A new dense query representation gradually aggregates exemplar-specific feature information across scales that leads to high-resolution dense queries that enable detection of large as well as small objects. GECO2 surpasses state-of-the-art few-shot counters in counting as well as detection accuracy by 10% while running 3x times faster at smaller GPU memory footprint.

Problem

• Addresses object counting across diverse scales and dense small object regions
• Improves detection accuracy for generalized-scale objects in few-shot counting
• Solves computational inefficiency in multi-scale object detection and counting

Innovation

• Gradually aggregates exemplar features across scales
• Generates high-resolution dense queries for detection
• Enables efficient multi-scale object counting and detection