jahongir7174 / DeepLab-tf

DeepLabV3+ Implementation using TensorFlow 2

DeepLabV3+

DeepLabV3+ implementation for Semantic Segmentation using TensorFlow 2

Train

• Run python train.py for final training

Test

• Run python test.py

Dataset structure (similar to CamVid dataset)

├── Dataset folder 
    ├── train
        ├── 1111.png
        ├── 2222.png
    ├── train_labels
        ├── 1111_L.png
        ├── 2222_L.png
    ├── class_dict.csv

Note

• default feature extractor is EfficientNetV2-S
• you can change backbone network in nets\nn.py
• changing configuration of training, change parameters in utils/config.py