---
title: Facial_Keypoints_Detection
canonical_url: "https://www.modelscope.cn/datasets/llmlearningX/Facial_Keypoints_Detection"
md_url: "https://www.modelscope.cn/datasets/llmlearningX/Facial_Keypoints_Detection.md"
repository: llmlearningX/Facial_Keypoints_Detection
chinese_name: "人脸关键点标定竞赛数据集"
last_updated: 2025-12-27
license: "Apache License 2.0"
downloads: 57
stars: 0
---

# Facial_Keypoints_Detection

> Facial_Keypoints_Detection - llmlearningX 在 ModelScope 开源的数据集。本数据集源自Kaggle经典的“Facial Keypoints Detection”竞赛。其核心任务是开发算法自动预测人脸图像中15个关键点（如眼睛中心、鼻尖、嘴角等）的(x, y)坐标。该任务是典型的坐标回归问题，在面部识别、表情分析及AR美颜领域具有重要基础作用。数据集包含：training.zip（7,049张带标注图像）、test.zip（1,783张测试图像）及IdLookupT…

llmlearningX/Facial_Keypoints_Detection 是 ModelScope 魔搭社区上的数据集，采用 Apache License 2.0 许可。

- **Repository**: llmlearningX/Facial_Keypoints_Detection
- **License**: Apache License 2.0
- **Downloads**: 57
- **Stars**: 0
- **Last updated**: 2025-12-27

Source: https://www.modelscope.cn/datasets/llmlearningX/Facial_Keypoints_Detection

---

数据集文件元信息以及数据文件，请浏览“数据集文件”页面获取。

当前数据集卡片使用的是默认模版，数据集的贡献者未提供更加详细的数据集介绍，但是您可以通过如下GIT Clone命令，或者ModelScope SDK来下载数据集

#### 下载方法 
:modelscope-code[]{type="sdk"}
:modelscope-code[]{type="git"}

# # 人脸关键点标定竞赛数据集 (Facial Keypoints Detection)

### ## 1. 数据集简介
本数据集源自 Kaggle 经典的计算机视觉竞赛 "Facial Keypoints Detection"。目标是训练模型在给定的人脸图像中准确标定关键特征点的位置。

### ## 2. 数据组成
- **training.zip**: 训练数据，包含图像及 15 个脸部关键点的坐标。
- **test.zip**: 测试数据，仅包含人脸图像。
- **IdLookupTable.csv**: 提交结果的 ID 映射表，用于规范化预测结果的输出格式。

### ## 3. 任务目标
通过深度学习（如 CNN）或传统回归算法，预测人脸图像中 15 个关键点（如眼心、鼻尖等）的坐标，是学习回归任务与面部识别技术的基准项目。

### ## 4. 技术应用
- 面部姿态评估
- 社交媒体美颜滤镜
- 疲劳驾驶监控
- 表情识别与分析
