yjh0410 2259ef87f2 update 2 жил өмнө
..
README.md ebe1d9e27f train RTCDet-S on COCO 2 жил өмнө
build.py 2259ef87f2 update 2 жил өмнө
loss.py be12f940f2 update 2 жил өмнө
matcher.py 222a4fadd1 debug AlignedSimOTA for RTCDet-M 2 жил өмнө
rtcdet.py 2259ef87f2 update 2 жил өмнө
rtcdet_backbone.py 4e429e1847 train RTCDet-S on COCO 2 жил өмнө
rtcdet_basic.py ae03c10fc1 train RTCDet-M with 320-960 2 жил өмнө
rtcdet_head.py 578dc295f4 redesign RTCDet 2 жил өмнө
rtcdet_neck.py 578dc295f4 redesign RTCDet 2 жил өмнө
rtcdet_pafpn.py 62a5b79e6a update 2 жил өмнө
rtcdet_pred.py f6753b0e9b train RTCDet-S 2 жил өмнө

README.md

RTCDet: My Empirical Study of Real-Time Convolutional Object Detectors.

Model Scale Batch APtest
0.5:0.95
APtest
0.5
APval
0.5:0.95
APval
0.5
FLOPs
(G)
Params
(M)
Weight
RTCDet-N 640 8xb16
RTCDet-T 640 8xb16
RTCDet-S 640 8xb16 30.9 8.5
RTCDet-M 640 8xb16 48.7 67.6 80.3 22.6 ckpt
RTCDet-L 640 8xb16
RTCDet-X 640 8xb16
  • For training, we train my RTCDet series series with 300 epochs on COCO.
  • For data augmentation, we use the large scale jitter (LSJ), Mosaic augmentation and Mixup augmentation, following the setting of YOLOX, but we remove the rotation transformation which is used in YOLOX's strong augmentation.
  • For optimizer, we use AdamW with weight decay 0.05 and base per image lr 0.001 / 64.
  • For learning rate scheduler, we use linear decay scheduler.
  • Due to my limited computing resources, I can not train RTCDet-X with the setting of batch size=128.
Model Scale Batch APval
0.5:0.95
APval
0.5
FLOPs
(G)
Params
(M)
Weight
RTCDet-P 320 8xb16 -
RTCDet-P 416 8xb16 -
RTCDet-P 512 8xb16 -
RTCDet-P 640 8xb16 -