Collections:

- Name: bisenetv2

  Metadata:

    Training Data:

    - Cityscapes

  Paper:

    URL: https://arxiv.org/abs/2004.02147

    Title: 'Bisenet v2: Bilateral Network with Guided Aggregation for Real-time Semantic

      Segmentation'

  README: configs/bisenetv2/README.md

  Code:

    URL: https://github.com/open-mmlab/mmsegmentation/blob/v0.18.0/mmseg/models/backbones/bisenetv2.py#L545

    Version: v0.18.0

Models:

- Name: bisenetv2_fcn_4x4_1024x1024_160k_cityscapes

  In Collection: bisenetv2

  Metadata:

    backbone: BiSeNetV2

    crop size: (1024,1024)

    lr schd: 160000

    inference time (ms/im):

    - value: 31.48

      hardware: V100

      backend: PyTorch

      batch size: 1

      mode: FP32

      resolution: (1024,1024)

    Training Memory (GB): 7.64

  Results:

  - Task: Semantic Segmentation

    Dataset: Cityscapes

    Metrics:

      mIoU: 73.21

      mIoU(ms+flip): 75.74

  Config: configs/bisenetv2/bisenetv2_fcn_4x4_1024x1024_160k_cityscapes.py

  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/bisenetv2/bisenetv2_fcn_4x4_1024x1024_160k_cityscapes/bisenetv2_fcn_4x4_1024x1024_160k_cityscapes_20210902_015551-bcf10f09.pth

- Name: bisenetv2_fcn_ohem_4x4_1024x1024_160k_cityscapes

  In Collection: bisenetv2

  Metadata:

    backbone: BiSeNetV2

    crop size: (1024,1024)

    lr schd: 160000

    Training Memory (GB): 7.64

  Results:

  - Task: Semantic Segmentation

    Dataset: Cityscapes

    Metrics:

      mIoU: 73.57

      mIoU(ms+flip): 75.8

  Config: configs/bisenetv2/bisenetv2_fcn_ohem_4x4_1024x1024_160k_cityscapes.py

  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/bisenetv2/bisenetv2_fcn_ohem_4x4_1024x1024_160k_cityscapes/bisenetv2_fcn_ohem_4x4_1024x1024_160k_cityscapes_20210902_112947-5f8103b4.pth

- Name: bisenetv2_fcn_4x8_1024x1024_160k_cityscapes

  In Collection: bisenetv2

  Metadata:

    backbone: BiSeNetV2

    crop size: (1024,1024)

    lr schd: 160000

    Training Memory (GB): 15.05

  Results:

  - Task: Semantic Segmentation

    Dataset: Cityscapes

    Metrics:

      mIoU: 75.76

      mIoU(ms+flip): 77.79

  Config: configs/bisenetv2/bisenetv2_fcn_4x8_1024x1024_160k_cityscapes.py

  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/bisenetv2/bisenetv2_fcn_4x8_1024x1024_160k_cityscapes/bisenetv2_fcn_4x8_1024x1024_160k_cityscapes_20210903_000032-e1a2eed6.pth

- Name: bisenetv2_fcn_fp16_4x4_1024x1024_160k_cityscapes

  In Collection: bisenetv2

  Metadata:

    backbone: BiSeNetV2

    crop size: (1024,1024)

    lr schd: 160000

    inference time (ms/im):

    - value: 27.29

      hardware: V100

      backend: PyTorch

      batch size: 1

      mode: FP16

      resolution: (1024,1024)

    Training Memory (GB): 5.77

  Results:

  - Task: Semantic Segmentation

    Dataset: Cityscapes

    Metrics:

      mIoU: 73.07

      mIoU(ms+flip): 75.13

  Config: configs/bisenetv2/bisenetv2_fcn_fp16_4x4_1024x1024_160k_cityscapes.py

  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/bisenetv2/bisenetv2_fcn_fp16_4x4_1024x1024_160k_cityscapes/bisenetv2_fcn_fp16_4x4_1024x1024_160k_cityscapes_20210902_045942-b979777b.pth