M2Det: A Single-Shot Object Detector based on Multi-Level Feature Pyramid Network @KarolMajek

M2Det: A Single-Shot Object Detector based on Multi-Level Feature Pyramid Network @KarolMajek

Uploaded March 2019 | Updated September 2026, 3 weeks ago

Here's the source code:
github.com/qijiezhao/M2Det

Paper: qijiezhao.github.io/imgs/m2det.pdf

Support me you can:
bit.ly/Coffee4Karol

M2Det: A Single-Shot Object Detector based on Multi-Level Feature Pyramid Network

YOLOv4 1024 FP16 - RTX2070 MaxQ - using ext_output

MMdetection: Cascade RCNN X101 64x4d FPN2x

Faster RCNN X101 64x4d FPN 1x

Faster RCNN HrNetV2p W32 1x

EfficientDet D1 VOC (toandaominh1997/EfficientDet.Pytorch)

MMdetection: Cascade Mask RCNN X101 32x4d FPN 1x

OpenCV YOLO Object Detection at Night - Adrian Rosebrock (pyImageSearch) implementation

F1/10 ESWEEK 2018 - 3min trials

YOLObile best14x 512x320

Cascade RCNN R101 FPN 1x (MMdetection)

Detectron2: RetinaNet R50 FPN 1x

M2Det: A Single-Shot Object Detector based on Multi-Level Feature Pyramid Network