Mask DINO for MapChallenge Instance Segmentation

Installation

See installation instructions.

Pretrained Model

Download the pretrained model: maskdino_swinl_50ep_300q_hid2048_3sd1_instance_maskenhanced_mask52.3ap_box59.0ap.pth

Training

python train_net.py \
     --num-gpus 1 \
     --config-file configs/coco/instance-segmentation/swin/maskdino_R50_bs16_50ep_4s_dowsample1_2048.yaml \
     MODEL.WEIGHTS maskdino_swinl_50ep_300q_hid2048_3sd1_instance_maskenhanced_mask52.3ap_box59.0ap.pth

Evaluation

python train_net.py \
     --eval-only \
     --num-gpus 1 \
     --config-file configs/coco/instance-segmentation/swin/maskdino_R50_bs16_50ep_4s_dowsample1_2048.yaml \
     DATASETS.TEST '("satellite_test",)' \
     MODEL.WEIGHTS "YOU_MODEL_CHECKPOINT_PATH"

MapChallenge Instance Segmentation Results

Model	segm mAP	segm mAP@50	segm mAP@75	segm mAP_s	segm mAP_m	segm mAP_l	segm mAR	segm mAR@50	segm mAR@75
SwinS-Mask2former	0.3110	0.7160	0.2180	0.1690	0.4420	0.0990	0.42	0.833	0.362
SwinL-Mask2former	0.3080	0.7260	0.2360	0.1710	0.4400	0.1310	0.453	0.855	0.464
Rtmdet-X	0.3910	0.7760	0.3440	0.2090	0.5440	0.2930	0.509	0.891	0.5
RTMdet-M	0.3790	0.7540	0.3170	0.2150	0.5210	0.2820	0.485	0.87	0.457
QueryInst-r50	0.2770	0.6520	0.1830	0.1380	0.3890	0.1530	0.432	0.826	0.384
QueryInst-r101	0.2780	0.6340	0.2000	0.1760	0.3910	0.1010	0.458	0.855	0.464
MaskDINO	0.584	0.9022	0.6150	0.3670	0.6911	0.9287	0.6802	0.9569	0.75

Note:

• mAP@50/mAR@50 refers to average precision/recall at IoU>=0.50 and mAP@75/mAR@75 refers to average precision/recall at IOU >= 0.75.

• segm mAP values are usually averaged over multiple IOUs from 0.5 to 0.95 with interval of 0.05

• In metric columns, - s = small objects - m = medium objects - l = large objects

Additional Usage

For full usage and detailed instructions, refer to Detectron2 Getting Started Guide.

Related repositories

• mmdetection training experiments