/Lightweight-Segmentation

Lightweight models for real-time semantic segmentation(include mobilenetv1-v3, shufflenetv1-v2, igcv3).

Primary LanguagePythonApache License 2.0Apache-2.0

Lightweight Model for Real-Time Semantic Segmentation

python-image pytorch-image lic-image

This project aims at providing the popular lightweight model implementations for real-time semantic segmentation.

Requisites

  • PyTorch 1.1
  • Python 3.x

Usage

Train

  • Single GPU training
python train.py --model mobilenet --dataset citys --lr 0.0001 --epochs 240
  • Multi-GPU training
# for example, train mobilenet with 4 GPUs:
export NGPUS=4
python -m torch.distributed.launch --nproc_per_node=$NGPUS train.py --model mobilenet --dataset citys --lr 0.0001 --epochs 240

Evaluation

  • Single GPU evaluating
python eval.py --model mobilenet_small --dataset citys
  • Multi-GPU evaluating
# for example, evaluate mobilenet with 4 GPUs:
export NGPUS=4
python -m torch.distributed.launch --nproc_per_node=$NGPUS eval.py --model mobilenet --dataset citys

Result

  • Cityscapes
Backbone OHEM Params(M) FLOPs(G) CPU(fps) GPU(fps) mIoU/pixACC Model
mobilenet ✘ 5.31 4.48 0.81 77.11 0.463/0.901 GoogleDrive
mobilenet ✓ 5.31 4.48 0.81 75.61 0.521/0.907 GoogleDrive
mobilenetv2 ✓ 4.88 4.04 0.49 49.40 0.613/0.930 GoogleDrive
mobilenetv3_small ✓ 1.02 1.64 2.59 104.56 0.529/0.908 GoogleDrive
mobilenetv3_large ✓ 2.68 4.59 1.39 79.43 0.584/0.916 GoogleDrive
shufflenet ✓ 6.89 5.68 0.57 43.79 0.493/0.901 GoogleDrive
shufflenetv2 ✓ 5.24 4.33 0.72 57.71 0.528/0.914 GoogleDrive
igcv3 ✓ 4.86 4.04 0.34 29.70 0.573/0.923 GoogleDrive

Note: crop_size=768, lr=0.01, epochs=80.

Support

To Do

  • provide trained model
  • add squeezenet, condensenet, shiftnet, mnasnet
  • train and eval
  • replace nn.SyncBatchNorm by nn.BatchNorm.convert_sync_batchnorm
  • check find_unused_parameters in nn.parallel.DistributedDataParallel

References