AdaCount: Training-Free Similarity-Guided Spatial and Feature Adaptation for Zero-Shot Object Counting

2026-07-02

AI Summary

Zero-shot object counting in dense, small-object scenes often suffers from missed detections and instance fragmentation due to limited resolution and insufficient focus on target regions. This work proposes AdaCount, a training-free framework that introduces, for the first time, a prototype-driven similarity map to guide spatial deformation and modulate encoder features, thereby enhancing SAM³’s representational capacity for target regions while preserving global context. Evaluated across six diverse counting benchmarks, AdaCount significantly outperforms existing zero-shot methods and achieves state-of-the-art performance among all training-free approaches.

Abstract

Zero-shot object counting (ZOC) aims to count instances of arbitrary object categories specified only through textual prompts. Recent training-free approaches leverage foundation models such as SAM to reformulate counting as a prompt-driven segmentation task, eliminating the need for costly counting-specific training data with point-level annotations. More recently, SAM3 introduced promptable concept segmentation, enabling the zero-shot segmentation of all instances corresponding to a text-defined concept. However, SAM3 struggles in densely populated scenes containing numerous small objects, where limited image resolution and insufficient attention to target-relevant regions often lead to missed instances and poor instance separation, hindering accurate object counting. To address this limitation, we propose AdaCount, a training-free framework for ZOC based on similarity-guided spatial and feature adaptation. AdaCount first estimates a prototype-driven similarity map that identifies target-relevant regions. This similarity map subsequently guides two complementary adaptations: (i) similarity-guided spatial warping, which reallocates image resolution toward target instances, and (ii) feature modulation, which amplifies target-relevant encoder representations. Together, these adaptations enable SAM3 to devote greater representational capacity to target-relevant regions while preserving global image context, without requiring any model retraining. Extensive experiments across six diverse counting benchmarks establish AdaCount as a new SOTA among training-free ZOC approaches.

Problem

• zero-shot object counting
• dense scenes
• small objects
• instance separation
• missed detection

Innovation

• zero-shot object counting
• training-free
• similarity-guided adaptation
• spatial warping
• feature modulation