---
title: facial_emotions_detection
canonical_url: "https://www.modelscope.cn/models/LangGPT/facial_emotions_detection"
md_url: "https://www.modelscope.cn/models/LangGPT/facial_emotions_detection.md"
repository: LangGPT/facial_emotions_detection
chinese_name: "人脸表情识别模型"
last_updated: 2024-11-27
pipeline_tag: image-classification
tasks:
  - image-classification
model_type:
  - vit
architectures:
  - ViTForImageClassification
parameters: 85.8M
tensor_type:
  - F32
library_name:
  - safetensors
  - pytorch
frameworks:
  - pytorch
downloads: 1042
stars: 3
---

# facial_emotions_detection

> facial_emotions_detection - LangGPT 在 ModelScope 开源的模型。这是一个基于 ViT 的人脸表情识别模型，可以识别7种基本表情： 悲伤 (sad) 厌恶 (disgust) 愤怒 (angry) 中性 (neutral) 恐惧 (fear) 惊讶 (surprise) 快乐 (happy)

- **Repository**: LangGPT/facial_emotions_detection
- **Tasks**: image-classification
- **Parameters**: 85.8M
- **Downloads**: 1042
- **Stars**: 3
- **Last updated**: 2024-11-27

Source: https://www.modelscope.cn/models/LangGPT/facial_emotions_detection

---

# 人脸表情识别模型

这是一个基于 ViT 的人脸表情识别模型，可以识别7种基本表情：
- 悲伤 (sad)
- 厌恶 (disgust)
- 愤怒 (angry)
- 中性 (neutral)
- 恐惧 (fear)
- 惊讶 (surprise)
- 快乐 (happy)

## 模型说明

- 基础模型：ViT-Base
- 训练数据集：FER2013
- 精度：91.2%
- 支持的任务：图像分类、表情识别

## 使用方法

```python
from modelscope.pipelines import pipeline
from modelscope.utils.constant import Tasks

# 加载模型
emotion_pipeline = pipeline(
    Tasks.image_classification,
    model='your_username/facial_emotions_detection'
)

# 进行预测
result = emotion_pipeline(image)
```