NUS-WIDE
收藏魔搭社区2026-01-08 更新2024-08-31 收录
下载链接:
https://modelscope.cn/datasets/OmniData/NUS-WIDE
下载链接
链接失效反馈官方服务:
资源简介:
displayName: NUS-WIDE
labelTypes:
- Classification
license:
- NUS-WIDE Custom
mediaTypes:
- Image
paperUrl: https://lms.comp.nus.edu.sg/wp-content/uploads/2019/research/nuswide/nuswide-civr2009.pdf
publishDate: "2009"
publishUrl: https://lms.comp.nus.edu.sg/wp-content/uploads/2019/research/nuswide/NUS-WIDE.html
publisher:
- Lab for Media Search in National University of Singapore
tags:
- Airplane
- Zebra
taskTypes:
- Image Retrieval
- Multi Label Classification
- Cross Modal Retrieval
---
# 数据集介绍
## 简介
NUS-WIDE 数据集包含从 Flickr 收集的 269,648 张图像,总共 5,018 个标签。这些图像是用 81 个概念手动注释的,包括对象和场景。
## 引文
```
"@inproceedings{10.1145/1646396.1646452,
author = {Chua, Tat-Seng and Tang, Jinhui and Hong, Richang and Li, Haojie and Luo, Zhiping and Zheng, Yantao},
title = {NUS-WIDE: A Real-World Web Image Database from National University of Singapore},
year = {2009}"
```
## Download dataset
:modelscope-code[]{type="git"}
displayName: NUS-WIDE
labelTypes:
- 分类(Classification)
license:
- NUS-WIDE 自定义许可(NUS-WIDE Custom)
mediaTypes:
- 图像(Image)
paperUrl: https://lms.comp.nus.edu.sg/wp-content/uploads/2019/research/nuswide/nuswide-civr2009.pdf
publishDate: "2009"
publishUrl: https://lms.comp.nus.edu.sg/wp-content/uploads/2019/research/nuswide/NUS-WIDE.html
publisher:
- 新加坡国立大学媒体搜索实验室(Lab for Media Search in National University of Singapore)
tags:
- 飞机(Airplane)
- 斑马(Zebra)
taskTypes:
- 图像检索(Image Retrieval)
- 多标签分类(Multi Label Classification)
- 跨模态检索(Cross Modal Retrieval)
---
# 数据集简介
## 概述
NUS-WIDE 数据集收录了从 Flickr 平台采集的269,648张图像,总计涵盖5018个标签。所有图像均通过81个概念(涵盖物体与场景)完成人工标注。
## 引文
"@inproceedings{10.1145/1646396.1646452,
author = {Chua, Tat-Seng and Tang, Jinhui and Hong, Richang and Li, Haojie and Luo, Zhiping and Zheng, Yantao},
title = {NUS-WIDE: A Real-World Web Image Database from National University of Singapore},
year = {2009}"
## 数据集下载
:modelscope-code[]{type="git"}
提供机构:
maas
创建时间:
2024-07-09
搜集汇总
数据集介绍

背景与挑战
背景概述
NUS-WIDE是一个大规模真实世界图像数据集,包含269,648张从Flickr收集的图像,手动标注了81个概念(如对象和场景),总计5,018个标签。该数据集由新加坡国立大学于2009年发布,主要用于图像检索、多标签分类和跨模态检索等计算机视觉任务,具有标注精细、应用广泛的特点。
以上内容由遇见数据集搜集并总结生成



