---
title: SmartCity
canonical_url: "https://www.modelscope.cn/datasets/OmniData/SmartCity"
md_url: "https://www.modelscope.cn/datasets/OmniData/SmartCity.md"
repository: OmniData/SmartCity
last_updated: 2024-07-01
license: "[MIT]"
storage_size: "28 MB"
domain:
  - publishDate
  - paperUrl
  - publishUrl
  - mediaTypes
  - publisher
  - taskTypes
  - displayName
tasks:
  - 2018
  - "https://arxiv.org/pdf/1711.04433v4.pdf"
  - "https://github.com/miao0913/SaCNN-CrowdCounting-Tencent_Youtu"
  - Image
  - "Inria Rennes - Bretagne Atlantique Research Centre"
  - "Object Detection"
  - SmartCity
downloads: 403
stars: 0
---

# SmartCity

> SmartCity - OmniData 在 ModelScope 开源的数据集。displayName: SmartCity license: MIT mediaTypes: Image paperUrl: https://arxiv.org/pdf/1711.04433v4.pdf publishDate: "2018" publishUrl: https://github.com/miao0913/SaCNN-CrowdCounting-TencentYoutu publisher: Shanghai…

OmniData/SmartCity 是 ModelScope 魔搭社区上的2018、https://arxiv.org/pdf/1711.04433v4.pdf、https://github.com/miao0913/SaCNN-CrowdCounting-Tencent_Youtu数据集，涉及 publishDate、paperUrl、publishUrl 领域，存储大小 28 MB，采用 [MIT] 许可。

- **Repository**: OmniData/SmartCity
- **License**: [MIT]
- **Tasks**: 2018, https://arxiv.org/pdf/1711.04433v4.pdf, https://github.com/miao0913/SaCNN-CrowdCounting-Tencent_Youtu, Image, Inria Rennes - Bretagne Atlantique Research Centre, Object Detection, SmartCity
- **Domain**: publishDate, paperUrl, publishUrl, mediaTypes, publisher, taskTypes, displayName
- **Storage size**: 28 MB
- **Downloads**: 403
- **Stars**: 0
- **Last updated**: 2024-07-01

Source: https://www.modelscope.cn/datasets/OmniData/SmartCity

---

displayName: SmartCity
license:
- MIT
mediaTypes:
- Image
paperUrl: https://arxiv.org/pdf/1711.04433v4.pdf
publishDate: "2018"
publishUrl: https://github.com/miao0913/SaCNN-CrowdCounting-Tencent_Youtu
publisher:
- Shanghai Jiao Tong University
- Tencent Youtu
- Inria Rennes - Bretagne Atlantique Research Centre
tags:
- People
taskTypes:
- Crowd Counting
- Object Counting
- Object Detection

---
# 数据集介绍
  ## 简介
  SmartCity 共包含 50 张图像，从办公室入口、人行道、中庭、购物中心等 10 个城市场景收集。与现有的人群计数数据集（包含数百/千行人的图像以及几乎所有在户外拍摄的图像）不同，SmartCity 具有图像中的行人很少，由室外和室内场景组成：平均行人数仅为 7.4，最小为 1，最大为 14。
  ## 引文
  ```
@inproceedings{zhang2018crowd,
  title={Crowd counting via scale-adaptive convolutional neural network},
  author={Zhang, Lu and Shi, Miaojing and Chen, Qiaobo},
  booktitle={2018 IEEE Winter Conference on Applications of Computer Vision (WACV)},
  pages={1113--1121},
  year={2018},
  organization={IEEE}
}
```
  
## Download dataset
:modelscope-code[]{type="git"}