MohsenZand / ObjectBox

(ECCV 22 Oral) ObjectBox: From Centers to Boxes for Anchor-Free Object Detection
GNU General Public License v3.0
131 stars 16 forks source link

On COCO2017 data set, the training is finished and verification is carried out, but the verification results are inconsistent. #19

Closed 1358422654 closed 1 year ago

1358422654 commented 1 year ago

Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.002 Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.003 Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.002 Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.002 Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.002 Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.004 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.070 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.149 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.161 Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.094 Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.168 Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.252

Hope to get your reply, thank you

1358422654 commented 1 year ago

Have been settled