search

sithu31296 / semantic-segmentation

SOTA Semantic Segmentation Models in PyTorch
MIT License
830 stars 149 forks source link

BiSeNetV2 model #51

Closed uyolo-ai closed 1 year ago

uyolo-ai commented 1 year ago

There are some differences between your implementation and others (PaddleSeg, mmsegmentation, BiSeNet)

For example: Your implementation https://github.com/sithu31296/semantic-segmentation/blob/492231c0d77088267ae6ce6b21a4107ccf8e92f8/semseg/models/bisenetv2.py#L139-L181 Others https://github.com/PaddlePaddle/PaddleSeg/blob/13aac35b31b147d2dd9976821ade13dabebff685/paddleseg/models/bisenet.py#L253-L302

https://github.com/CoinCheung/BiSeNet/blob/f2b901599752ce50656d2e50908acecd06f7eb47/lib/models/bisenetv2.py#L224

AggregationLayer only uses one resize op, others have two resize ops