Skip to content

2 classes of lightweight object detection. train on VOC 、SAR

Notifications You must be signed in to change notification settings

237014845/MobilenetV2-Retina-Pytorch

Repository files navigation

Lightweight Object Detection(One-Stage)

Introduction

The code is based on the mmdetection.

mmdetection is an open source object detection toolbox based on PyTorch. It is a part of the open-mmlab project developed by Multimedia Laboratory, CUHK.

Currently, it contains these features:

  • Multiple Base Network: Mobilenet V2, ShuffleNet V2
  • One-Stage Lightweight Detector: MobileV2-SSD, MobileV2-RetinaNet

Performance

VOC2007 SSD RetinaNet
MobilenetV2 81.9%
ShufflenetV2
SAR(SSDD) SSD RetinaNet
MobilenetV2 90.4% 91.7%
ShufflenetV2
COCO2017 SSD RetinaNet
MobilenetV2 31.7
ShufflenetV2

Demo

demo image demo image demo image

TODO

About

2 classes of lightweight object detection. train on VOC 、SAR

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published