Image-Segmentation-PyTorch

Supported networks

• UNet: backbones MobileNetV2 (all aphas and expansions), ResNetV1 (all num_layers)
• DeepLab3+: backbones ResNetV1 (num_layers=18,34,50,101), VGG16_bn
• BiSeNet: backbones ResNetV1 (num_layers=18)
• PSPNet: backbones ResNetV1 (num_layers=18,34,50,101)
• ICNet: backbones ResNetV1 (num_layers=18,34,50,101)

To assess architecture, memory, forward time (in either cpu or gpu), numper of parameters, and number of FLOPs of a network, use this command:

python measure_model.py

Set

• Python3.6.x is used in this repository.
• Clone the repository:

git clone --recursive https://github.com/YuantingMaSC/Image-Segmentation.git
cd Image-Segmentation
git submodule sync
git submodule update --init --recursive

• To install required packages, use pip:

workon humanseg
pip install -r requirements.txt
pip install -e models/pytorch-image-models

Data preparation

• Using "Lableme" tool to prepare and following script:

./original_data> python labelme2voc.py ./imgs_cut ./labelme_dataset --labels ./labels.txt

Training

• For training a network from scratch, for example DeepLab3+, use this command:

python train.py --config config/config_DeepLab.json --device 0

where config/config_DeepLab.json is the configuration file which contains network, dataloader, optimizer, losses, metrics, and visualization configurations.

• For resuming training the network from a checkpoint, use this command:

python train.py --config config/config_DeepLab.json --device 0 --resume path_to_checkpoint/model_best.pth

• One can open tensorboard to monitor the training progress by enabling the visualization mode in the configuration file.

images to Vedio

• using imgs2video.py

python imgs2video.py

Inference

There are two modes of inference:

python inference_video.py --watch --use_cuda --checkpoint path_to_checkpoint/model_best.pth
python inference_webcam.py --use_cuda --checkpoint path_to_checkpoint/model_best.pth

Benchmark

• Networks are trained on a combined dataset from the two mentioned datasets above.
• Input size of model is set to 320.
• The CPU and GPU time is the averaged inference time of 10 runs (there are also 10 warm-up runs before measuring) with batch size 1.
• The mIoU is measured on the testing subset (737 images) from the combined dataset.
• Hardware configuration for benchmarking:

CPU: 12th Gen Intel(R) Core(TM) i9-12900K 3.20 GHz

GPU: NVIDIA GeForce RTX 3090