Mushroomcat9998 / deep_ensemble_learning

Deep Ensemble Learning for Human Action Recognition in Still Images

deep_ensemble_learning

Deep Ensemble Learning for Human Action Recognition in Still Images

The codes are provided by Xiangchun Yu.

Email:[e-mail]

Data: November 08, 2019

################################################################### Step1. Please extract the compressed file to the current folder. Such as: deep_ensemble_learning/Action_ForCNN/train...

Step2. Train the VGG16, VGG16_NCNN, VGG19, VGG19_NCNN, ResNet50, ResNet50_NCNN models using the following python scripts,

Main1_VGG16.py

Main1_VGG16_NCNN.py

Main1_VGG19.py

Main1_VGG19_NCNN.py

Main3_ResNet50.py

Main3_ResNet50_NCNN.py

And use the Main4_Evaluate_Results.py test the results of the models mentioned above.

Step3. Train the DELWO1-DELWO3 models using the following python scripts,

Main5_DELWO1.py

Main5_DELWO2.py

Main5_DELWO3.py

And use the Main6_Evaluate_Results_for_DELWOs.py test the results of the models mentioned above.

Step4. Evaluate the models' performance using DELVS1-DELVS3 using the following python scripts,

Main7_DELVS1.py

Main7_DELVS2.py

Main7_DELVS3.py

Note: The final results will be obtained directly using the above python scripts.

If you feel the codes are useful, please cite our following published paper,

Xiangchun Yu, Zhe Zhang, Lei Wu, Wei Pang, Hechang Chen, Zhezhou Yu, and Bin Li, "Deep Ensemble Learning for Human Action Recognition in Still Images," Complexity, vol. 2020, Article ID 9428612, 23 pages, 2020. https://doi.org/10.1155/2020/9428612.