---
language:
- zh
tags:
- hisilicon
- hispark
- npu
- openharmony
- modelzoo
- pytorch
---

# CrowdCount

CrowdCount是一种基于多尺度卷积神经网络（MSCNN）的高精度人群计数模型。相比传统多列 / 多网络方法，它通过单列网络中的多尺度特征块（MSB）与尺度自适应密度图回归技术，能有效应对透视畸变导致的人物尺度差异问题，兼顾计数精度与模型轻量化，适用于监控图像、公共场所等密集人群计数场景。

## Mirror Metadata

- Hugging Face repo: shadow-cann/hispark-modelzoo-crowdcount
- Portal model id: i9i4tr9hec00
- Created at: 2025-12-25 15:03:50
- Updated at: 2025-12-30 21:27:21
- Category: 计算机视觉

## Framework

- PyTorch

## Supported OS

- OpenHarmony
- Linux

## Computing Power

- Hi3403V100 SVP_NNN
- Hi3403V100 NNN

## Tags

- 人群计数

## Detail Parameters

- 输入: 224x224
- 参数量: 34.738M
- 计算量: 587.807GFLOPs

## Files In This Repo

- mscnn_model.onnx (源模型 / 源模型下载; 源模型 / 源模型元数据)
- SVP_NNN_PC_V1.0.6.0.tgz (附加资源 / 附加资源)

## Upstream Links

- Portal card: https://gitbubble.github.io/hisilicon-developer-portal-mirror/model-detail.html?id=i9i4tr9hec00
- Upstream repository: https://gitee.com/HiSpark/modelzoo/tree/master/samples/built-in/count/CrowdCount
- License reference: https://github.com/zzubqh/CrowdCount/blob/master/LICENSE

## Notes

- This repository was mirrored from the HiSilicon Developer Portal model card and local downloads captured on 2026-03-27.
- File ownership follows the portal card mapping, not just filename similarity.
- Cover image: 1719967754027009_combined_plot.jpg