maiomood.blogg.se

Networkįor FCOS training,I use yolov3 resize method,this method resize=667 has same flops as the resize=400 method proposed in the FCOS paper.

Using one RTX2080Ti to test RetinaNet model inference speed.The test is performed COCO2017_val dataset,compute average per image inference time(ms).Testing num_workers=8. Networkįor ResNet50-RetinaNet resize=1000 training,I use ResNet50-RetinaNet resize=667 trained model(mAP=0.293) as a pretrained model parameters to initialize the ResNet50-RetinaNet resize=1000 model.įor RetinaNet training,I use yolov3 resize method,this method resize=1000 has same flops as the resize=600 method proposed in the RetinaNet paper. RetinaNetįor RetinaNet training,I use yolov3 resize method,this method resize=667 has same flops as the resize=400 method proposed in the RetinaNet paper. You can find more model training details in detection_experiments/experiment_folder/. Trained on COCO2017_train dataset, tested on COCO2017_val dataset. You can read the codes in detect_single_image.py to find more details.