---
title: "Listen and Count: Expanding the Frontier of Zero-shot Object Counting to Sound-centric Counting"
authors:
  - "Mingjie Wang"
  - "Zhuohang Li"
  - "Qi Zhang"
  - "Jun Zhou"
  - "Zili Yi"
  - "Minglun Gong"
doi: "10.24963/ijcai.2026/194"
url: "https://ijcai.org/proceedings/2026/194"
pdf: "https://ijcai.org/proceedings/2026/0194.pdf"
venue: "Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"
year: 2026
pages: "1740-1748"
track: "Main Track"
keywords:
  - "Computer Vision: Scene analysis and understanding"
  - "Computer Vision: Applications and Systems"
  - "Computer Vision: Multimodal learning"
---

# Listen and Count: Expanding the Frontier of Zero-shot Object Counting to Sound-centric Counting

Mingjie Wang, Zhuohang Li, Qi Zhang, Jun Zhou, Zili Yi, Minglun Gong

*Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence*, pages 1740-1748. DOI: [10.24963/ijcai.2026/194](https://doi.org/10.24963/ijcai.2026/194)

## Abstract

While class-agnostic object counting has recently evolved from image-exemplar to language-guided paradigms, existing methods are limited by text polysemy and the lack of prompts in audio-sensing scenarios. To overcome these challenges, we introduce a sound-centric counting paradigm, enabling models to "listen and count" using the vivid and discriminative signatures of sound cues. We propose S2ICount, which achieves seamless, fine-grained audio-visual alignment for sound-guided counting. At its core, SoundMamba leverages linear-complexity long-range semantic modelling to align cross-modal features, while the Sound Calibration Module designs a circular scanning mechanism to progressively reconcile modality discrepancies using multi-scale cues from a Stable Diffusion backbone. To enhance saliency awareness, a Multi-tier Hybrid Optimization strategy enforces consistency across magnitude, semantics, and distribution. In addition, we introduce SoundCount, the first large-scale sound-oriented counting dataset, comprising over 2,000 scenes and 1,650 audio sources across 66 object categories. Extensive experiments demonstrate that S2ICount achieves state-of-the-art performance, validating the effectiveness of sound as a robust guidance modality for counting.  The dataset, demo and implementation code are made publicly accessible at https://github.com/ZSTU-CV-Lab/S2ICount.