tianshuichen/CD-FER-Benchmark

Adversarial Graph Representation Adaptation for Cross-Domain Facial Expression Recognition

Implementation of papers:

• Adversarial Graph Representation Adaptation for Cross-Domain Facial Expression Recognition
  ACM International Conference on Multimedia (ACM MM), 2020.
  Yuan Xie, Tianshui Chen, Tao Pu, Hefeng Wu, Liang Lin.

Environment

Ubuntu 16.04 LTS, Python 3.5, PyTorch 1.3
PS: We also provide docker image for this project, click here. (Tag: py3-pytorch1.3-agra)

Datasets

You can download datasets in Baidu Drive (password: c7xi) and OneDrive (password: xxxx), which includes CK+/SFEW 2.0/FER2013/JAFFE/ExpW/RAF/AFED.

PS : In order to comply with relevant regulations, you need to apply for the image data of the following data sets by yourself, including CK+/SFEW 2.0/FER2013/JAFFE/ExpW/RAF.

Pre-Train Model

You can download pre-train models in Baidu Drive (password: tzrf) and OneDrive (password: xxx).
PS: To replace backbone of each methods, you should modify and run getPreTrainedModel_ResNet.py (or getPreTrainedModel_MobileNet.py) in the folder where you want to use the method.

Usage

Before run these script files, you should download datasets and pre-train model, and run getPreTrainedModel_ResNet.py (or getPreTrainedModel_MobileNet.py).

Run AGRA

cd AGRA
bash TrainOnSourceDomain.sh     # Train Model On Source Domain
bash TransferToTargetDomain.sh  # Then, Transfer Model to Target Domain

Run SAFN

cd AGRA
bash TrainWithSAFN.sh

Run SWD

cd SWD
bash Train.sh

Run LPL

cd LPL
bash Train.sh

Run DETN & ECAN

cd AGRA
bash TrainOnSourceDomain.sh

# Set Lambda = 1, if you want use ECAN.
# Set Lambda = 0, if you want use DETN.
bash TransferToTargetDomain.sh 

Run FTDNN

cd FTDNN
bash Train.sh

Run ICID

cd ICID
bash Train.sh

Run DFA

cd DFA
bash Train.sh

Result

Backbone: ResNet-50

Souce Domain: RAF

Methods	CK+	JAFFE	SFEW2.0	FER2013	ExpW	Mean
ICID	74.42	50.70	48.85	53.70	69.54	59.44
DFA	64.26	44.44	43.07	45.79	56.86	50.88
LPL	74.42	53.05	48.85	55.89	66.90	59.82
DETN	78.22	55.89	49.40	52.29	47.58	56.68
FTDNN	79.07	52.11	47.48	55.98	67.72	60.47
ECAN	79.77	57.28	52.29	56.46	47.37	58.63
CADA	72.09	52.11	53.44	57.61	63.15	59.68
SAFN	75.97	61.03	52.98	55.64	64.91	62.11
SWD	75.19	54.93	52.06	55.84	68.35	61.27
Ours	85.27	61.50	56.43	58.95	68.50	66.13

Souce Domain: AFED

Methods	CK+	JAFFE	SFEW2.0	FER2013	ExpW	Mean
ICID	56.59	57.28	44.27	46.92	52.91	51.59
DFA	51.86	52.70	38.03	41.93	60.12	48.93
LPL	73.64	61.03	49.77	49.54	55.26	57.85
DETN	56.27	52.11	44.72	42.17	59.80	51.01
FTDNN	61.24	57.75	47.25	46.36	52.89	53.10
ECAN	58.14	56.91	46.33	46.30	61.44	53.82
CADA	72.09	49.77	50.92	50.32	61.70	56.96
SAFN	73.64	64.79	49.08	48.89	55.69	58.42
SWD	72.09	61.50	48.85	48.83	56.22	57.50
Ours	78.57	65.43	51.18	51.31	62.71	61.84

---

Backbone: ResNet-18

Souce Domain: RAF

Methods	CK+	JAFFE	SFEW2.0	FER2013	ExpW	Mean
ICID	67.44	48.83	47.02	53.00	68.52	56.96
DFA	54.26	42.25	38.30	47.88	47.42	46.02
LPL	72.87	53.99	49.31	53.61	68.35	59.63
DETN	64.19	52.11	42.25	42.01	43.92	48.90
FTDNN	76.74	50.23	49.54	53.28	68.08	59.57
ECAN	66.51	52.11	48.21	50.76	48.73	53.26
CADA	73.64	55.40	52.29	54.71	63.74	59.96
SAFN	68.99	49.30	50.46	53.31	68.32	58.08
SWD	72.09	53.52	49.31	53.70	65.85	58.89
Ours	77.52	61.03	52.75	54.94	69.70	63.19

Souce Domain: AFED

Methods	CK+	JAFFE	SFEW2.0	FER2013	ExpW	Mean
ICID	54.26	51.17	47.48	46.44	54.85	50.84
DFA	35.66	45.82	34.63	36.88	62.53	43.10
LPL	67.44	62.91	48.39	49.82	54.51	56.61
DETN	44.19	47.23	45.46	45.39	58.41	48.14
FTDNN	58.91	59.15	47.02	48.58	55.29	53.79
ECAN	44.19	60.56	43.26	46.15	62.52	51.34
CADA	72.09	53.99	48.39	48.61	58.50	56.32
SAFN	68.22	61.50	50.46	50.07	55.17	57.08
SWD	77.52	59.15	50.69	51.84	56.56	59.15
Ours	79.84	61.03	51.15	51.95	65.03	61.80

---

Backbone: MobileNet V2

Souce Domain: RAF

Methods	CK+	JAFFE	SFEW2.0	FER2013	ExpW	Mean
ICID	57.36	37.56	38.30	44.47	60.64	47.67
DFA	41.86	35.21	29.36	42.36	43.66	38.49
LPL	59.69	40.38	40.14	50.13	62.26	50.52
DETN	53.49	40.38	35.09	45.88	45.26	44.02
FTDNN	71.32	46.01	45.41	49.96	62.87	55.11
ECAN	53.49	43.08	35.09	45.77	45.09	44.50
CADA	62.79	53.05	43.12	49.34	59.40	53.54
SAFN	66.67	45.07	40.14	49.90	61.40	52.64
SWD	68.22	55.40	43.58	50.30	60.04	55.51
Ours	72.87	55.40	45.64	51.05	63.94	57.78

Souce Domain: AFED

Methods	CK+	JAFFE	SFEW2.0	FER2013	ExpW	Mean
ICID	55.04	42.72	34.86	39.94	44.34	43.38
DFA	44.19	27.70	31.88	35.95	61.55	40.25
LPL	69.77	50.23	43.35	45.57	51.63	52.11
DETN	57.36	54.46	32.80	44.11	64.36	50.62
FTDNN	65.12	46.01	46.10	46.69	53.02	51.39
ECAN	71.32	56.40	37.61	45.34	64.00	54.93
CADA	70.54	45.07	40.14	46.72	54.93	51.48
SAFN	62.79	53.99	42.66	46.61	52.65	51.74
SWD	64.34	53.52	44.72	50.24	55.85	53.73
Ours	75.19	54.46	47.25	47.88	61.10	57.18

---

Mean of All Methods

Souce Domain: RAF

Methods	CK+	JAFFE	SFEW2.0	FER2013	ExpW	Mean
ResNet-50	75.87	54.30	54.49	54.82	62.09	59.51
ResNet-18	69.43	51.88	47.94	51.72	61.26	56.45
MobileNet V2	60.78	45.15	39.59	47.92	56.46	49.98

Souce Domain: AFED

Methods	CK+	JAFFE	SFEW2.0	FER2013	ExpW	Mean
ResNet-50	65.41	57.93	47.04	47.26	57.87	55.10
ResNet-18	60.23	56.25	46.95	47.57	58.34	53.87
MobileNet V2	63.57	48.46	40.14	44.91	56.34	50.68

Citation

To Be Done.

Contributing

For any questions, feel free to open an issue or contact us:

• phoenixsysu@gmail.com
• tianshuichen@gmail.com
• putao537@gmail.com