Skip to content

Latest commit

 

History

History

mobilenet

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
README.md
README.md
 
 

README.md

Introduction

Official Repo

Code Snippet

MobileNetV2 (CVPR'2018) 
@inproceedings{sandler2018mobilenetv2,
    title={Mobilenetv2: Inverted residuals and linear bottlenecks},
    author={Sandler, Mark and Howard, Andrew and Zhu, Menglong and Zhmoginov, Andrey and Chen, Liang-Chieh},
    booktitle={Proceedings of the IEEE conference on computer vision and pattern recognition},
    pages={4510--4520},
    year={2018}
}
MobileNetV3 (ICCV'2019) 
@inproceedings{Howard_2019_ICCV,
    title={Searching for MobileNetV3},
    author={Howard, Andrew and Sandler, Mark and Chu, Grace and Chen, Liang-Chieh and Chen, Bo and Tan, Mingxing and Wang, Weijun and Zhu, Yukun and Pang, Ruoming and Vasudevan, Vijay and Le, Quoc V. and Adam, Hartwig},
    booktitle={The IEEE International Conference on Computer Vision (ICCV)},
    pages={1314-1324},
    month={October},
    year={2019},
    doi={10.1109/ICCV.2019.00140}}
}

Results

PASCAL VOC

Segmentor Pretrain Backbone Crop Size Schedule Train/Eval Set mIoU Download
FCN ImageNet-1k-224x224 M-V2-D8 512x512 LR/POLICY/BS/EPOCH: 0.01/poly/16/60 trainaug/val 59.89% cfg | model | log
PSPNet ImageNet-1k-224x224 M-V2-D8 512x512 LR/POLICY/BS/EPOCH: 0.01/poly/16/60 trainaug/val 68.40% cfg | model | log
DeepLabV3 ImageNet-1k-224x224 M-V2-D8 512x512 LR/POLICY/BS/EPOCH: 0.01/poly/16/60 trainaug/val 70.08% cfg | model | log
DeepLabV3plus ImageNet-1k-224x224 M-V2-D8 512x512 LR/POLICY/BS/EPOCH: 0.01/poly/16/60 trainaug/val 70.04% cfg | model | log
LRASPPNet - M-V3S-D8 512x512 LR/POLICY/BS/EPOCH: 0.01/poly/16/180 trainaug/val 62.13% cfg | model | log
LRASPPNet - M-V3L-D8 512x512 LR/POLICY/BS/EPOCH: 0.01/poly/16/180 trainaug/val 67.90% cfg | model | log

ADE20k

Segmentor Pretrain Backbone Crop Size Schedule Train/Eval Set mIoU Download
FCN ImageNet-1k-224x224 M-V2-D8 512x512 LR/POLICY/BS/EPOCH: 0.01/poly/16/130 train/val 30.85% cfg | model | log
PSPNet ImageNet-1k-224x224 M-V2-D8 512x512 LR/POLICY/BS/EPOCH: 0.01/poly/16/130 train/val 35.09% cfg | model | log
DeepLabV3 ImageNet-1k-224x224 M-V2-D8 512x512 LR/POLICY/BS/EPOCH: 0.01/poly/16/130 train/val 37.55% cfg | model | log
DeepLabV3plus ImageNet-1k-224x224 M-V2-D8 512x512 LR/POLICY/BS/EPOCH: 0.01/poly/16/130 train/val 37.66% cfg | model | log
LRASPPNet - M-V3S-D8 512x512 LR/POLICY/BS/EPOCH: 0.01/poly/16/390 train/val 26.09% cfg | model | log
LRASPPNet - M-V3L-D8 512x512 LR/POLICY/BS/EPOCH: 0.01/poly/16/390 train/val 30.06% cfg | model | log

CityScapes

Segmentor Pretrain Backbone Crop Size Schedule Train/Eval Set mIoU Download
FCN ImageNet-1k-224x224 M-V2-D8 512x1024 LR/POLICY/BS/EPOCH: 0.01/poly/8/220 train/val 70.77% cfg | model | log
PSPNet ImageNet-1k-224x224 M-V2-D8 512x1024 LR/POLICY/BS/EPOCH: 0.01/poly/8/220 train/val 73.64% cfg | model | log
DeepLabV3 ImageNet-1k-224x224 M-V2-D8 512x1024 LR/POLICY/BS/EPOCH: 0.01/poly/8/220 train/val 76.74% cfg | model | log
DeepLabV3plus ImageNet-1k-224x224 M-V2-D8 512x1024 LR/POLICY/BS/EPOCH: 0.01/poly/8/220 train/val 76.68% cfg | model | log
LRASPPNet - M-V3S-D8 512x1024 LR/POLICY/BS/EPOCH: 0.01/poly/16/660 train/val 65.06% cfg | model | log
LRASPPNet - M-V3L-D8 512x1024 LR/POLICY/BS/EPOCH: 0.01/poly/16/660 train/val 69.98% cfg | model | log

More

You can also download the model weights from following sources: