Object Detection¶
Pre-requisite¶
All detection modules depend on mmcv-full
, mmdet
and pycocotools
packages.
To install mmcv-full
, run:
mim install mmcv-full
To install mmdet
, run:
pip install mmdet
To install pycocotools
, run:
pip install pycocotools
We suggest to install the version between pycocotools>=2.0.5,<2.0.7
.
Note that Windows users should install pycocotools
by: pip install pycocotools-windows
, but it only supports python 3.6/3.7/3.8.
For additional support, please refer to official instructions for mmdet and mmcv-full
Quick Start¶
How to train high quality object detection model with MultiModalPredictor in under 5 minutes on COCO format dataset.
Data Preparation¶
How to prepare COCO2017 dataset for object detection.
How to prepare Pascal VOC dataset for object detection.
How to prepare Watercolor dataset for object detection.
How to convert a dataset from VOC format to COCO format for object detection.
Finetune¶
How to fast finetune a pretrained model on a dataset in COCO format.
How to finetune a pretrained model on a dataset in COCO format with high performance.
Evaluation¶
How to evaluate the very fast pretrained YOLOv3 model on dataset in COCO format
How to evaluate the pretrained Faster R-CNN model with high performance on dataset in COCO format
How to evaluate the pretrained Deformable DETR model with higher performance on dataset in COCO format
How to evaluate the pretrained Faster R-CNN model on dataset in VOC format