PyTorch implementation of over 30 realtime semantic segmentations models, e.g. BiSeNetv1, BiSeNetv2, CGNet, ContextNet, DABNet, DDRNet, EDANet, ENet, ERFNet, ESPNet, ESPNetv2, FastSCNN, ICNet, LEDNet, LinkNet, PP-LiteSeg, SegNet, ShelfNet, STDC, SwiftNet, and support knowledge distillation, distributed training, Optuna etc.

TensorRT实现BiSeNetV1与BiSeNetV2部署

Code for BiseNetV1

Code for BiseNetV1

No description available

Implementation of BiseNet neural network as feature extractor and then use it as generator in a domain adversarial neural network framework.

No description available

No description available

Personal implementation of BiseNet v1 Used for Portrait Segmentation

No description available

Combine contrastive learning with the BiSeNetV1 semantic segmentation model, where contrastive learning uses the method mentioned in "Exploring Cross Image Pixel Contrast for Semantic Segmentation".