Hugging Face's logo

cyinen
/
S2D

Model card Files
xet
Community
S2D
File size: 3,893 Bytes 
8acda2c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
340ae55
8acda2c
 
 
 
 
 
 
 
 
 
 
 
340ae55
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
 
---
license: apache-2.0
---
# From Static to Dynamic: Adapting Landmark-Aware Image Models for Facial Expression Recognition in Videos

 <img width="1024" alt="image" src="https://github.com/user-attachments/assets/c629e924-cec2-46c9-9e9a-369b4e6d0aef">


[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/from-static-to-dynamic-adapting-landmark-1/dynamic-facial-expression-recognition-on)](https://paperswithcode.com/sota/dynamic-facial-expression-recognition-on?p=from-static-to-dynamic-adapting-landmark-1)<br>
[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/from-static-to-dynamic-adapting-landmark-1/dynamic-facial-expression-recognition-on-dfew)](https://paperswithcode.com/sota/dynamic-facial-expression-recognition-on-dfew?p=from-static-to-dynamic-adapting-landmark-1)<br>
[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/from-static-to-dynamic-adapting-landmark-1/dynamic-facial-expression-recognition-on-mafw)](https://paperswithcode.com/sota/dynamic-facial-expression-recognition-on-mafw?p=from-static-to-dynamic-adapting-landmark-1)<br>

[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/from-static-to-dynamic-adapting-landmark-1/facial-expression-recognition-on-affectnet)](https://paperswithcode.com/sota/facial-expression-recognition-on-affectnet?p=from-static-to-dynamic-adapting-landmark-1)<br>
[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/from-static-to-dynamic-adapting-landmark-1/facial-expression-recognition-on-raf-db)](https://paperswithcode.com/sota/facial-expression-recognition-on-raf-db?p=from-static-to-dynamic-adapting-landmark-1)<br>

>[From Static to Dynamic: Adapting Landmark-Aware Image Models for Facial Expression Recognition in Videos](https://ieeexplore.ieee.org/document/10663980)<br>
>Yin Chen$^{†}$, Jia Li$^{β€ βˆ—}$, Shiguang Shan, Meng Wang, and Richang Hong 


## πŸ“° News
**[2024.9.5]** The fine-tuned checkpoints are available.

**[2024.9.2]** The code and pre-trained models are available.

**[2024.8.28]** The paper is accepted by IEEE Transactions on Affective Computing.

**[2023.12.5]** ~~Code and pre-trained models will be released here~~. 

## πŸš€ Main Results

### Dynamic Facial Expression Recognition
<img width="1024" alt="image" src="https://github.com/user-attachments/assets/2144837d-9fd5-4f88-8447-1f6049b38e9a">

<img width="1024" alt="image" src="https://github.com/user-attachments/assets/4a80731e-666e-4cef-9f74-5f794eea7116">


### Static Facial Expression Recognition
<img width="1024" alt="image" src="https://github.com/user-attachments/assets/89a47ea3-1036-4124-927c-563af8007d1f">

### Visualization 
<img width="1024" alt="image" src="https://github.com/user-attachments/assets/aea1385d-0d1b-4f5e-8775-087a30363751">


## Fine-tune with pre-trained weights
1、 Download the pre-trained weights from [baidu drive](https://pan.baidu.com/s/1J5eCnTn_Wpn0raZTIUCfgw?pwd=dji4) or [google drive](https://drive.google.com/file/d/1Y9zz8z_LwUi-tSFBAwDPZkVoyY6mhZlu/view?usp=drive_link) or [onedrive](https://mailhfuteducn-my.sharepoint.com/:f:/g/personal/2022111029_mail_hfut_edu_cn/EgKQNq8Y2chKl2TSoYf_OA0BQpCwx-FDw2ksPaMxBntZ8A), and move it to the ckpts directory.

2、 Run the following command to fine-tune the model on the target dataset.
```bash
conda create -n s2d python=3.9
conda activate s2d
pip install -r requirements.txt
bash run.sh
```

## πŸ“‹ Reported Results and Fine-tuned Weights
The fine-tuned checkpoints can be downloaded from [baidu driver](https://pan.baidu.com/s/1Xz5j8QW32x7L0bnTEorUbA?pwd=5drk) or [huggingface](https://huggingface.co/cyinen/S2D).
<table border="1" cellspacing="0" cellpadding="5">
[![Star History Chart](https://api.star-history.com/svg?repos=MSA-LMC/S2D&type=Date)](https://star-history.com/#MSA-LMC/S2D&Date)