Mindspore-image-processing

Preface

• This project is dedicated to the application of mindspore in the field of image processing, using models as classical models in CV, including image classification, key point detection, object detection, instance segmentation and semantic segmentation, as well as some other small cases of mindspore.
• In key point detection, object detection, instance segmentation and semantic segmentation cases, stand-alone single-card and stand-alone multi-card demonstration cases are provided.

The following is the catalog of models used by the project:

Tutorial catalog

• classification

  • LeNet
  • AlexNet
  • VggNet
  • GoogLeNet
  • ResNet
  • ResNeXt
  • MobileNet_V1_V2
  • MobileNet_V3
  • ShuffleNet_V1_V2
  • EfficientNet_V1
  • EfficientNet_V2
  • RepVGG
  • Vision Transformer
  • Swin Transformer
  • ConvNeXt
  • MobileViT

• object-detection

  • Faster-RCNN/FPN
  • SSD
  • Mask R-CNN
  • RetinaNet

• segmentation

  • FCN
  • DeepLabV3
  • LR-ASPP
  • U-Net
  • U2Net

• keypoint

  • HRNet

---

Required environment

• Anaconda3
• python3.8
• VSCode (IDE)
• mindspore2.0
• requirements.txt