This is a SSD experiment reposity, the purpose is to reproduce some related papers based on SSD The code references SSD: Single Shot MultiBox Object Detector, in PyTorch and mmdet. Currently, some experiments are carried out on the VOC dataset, if you want to train your own dataset, you can refer to the part of training-yourself-dataset.
The fold structure as follow:
pip install visdom
python visdom
The pretrained model refer pretrained-models.pytorch,you can download it.
In the SSD_FPN_GIoU fold:
python tools/train.py
Note:
if you want to trainning yourself dataset, there are some steps:
In the SSD_FPN_GIoU fold:
python tools/train.py
python eval.py --trained_model your_weight_address
python test.py -- trained_model your_weight_address
if you want to visual the box, you can add the command --visbox True(default False)
Test | mAP(iou=0.5) | mAP(iou=0.6) | mAP(iou=0.75) |
---|---|---|---|
SSD | 75.49% | 70.87% | 53.44 % |
SSD+Gious | 76.09% | 71.01% | 54.70 % |
SSD+FPN | 78.99% | 73.18% | 55.19 % |
SSD+FPN+Gious | 78.96% | 73.35% | 55.83 % |