Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks

Shaoqing RenKaiming HeRoss GirshickJian Sun

   Papers with code   Abstract  PDF

State-of-the-art object detection networks depend on region proposal algorithms to hypothesize object locations. Advances like SPPnet and Fast R-CNN have reduced the running time of these detection networks, exposing region proposal computation as a bottleneck... (read more)

Benchmarked Models

RANK
MODEL
REPO
CODE RESULT
PAPER RESULT
ε-REPRODUCED
BUILD
1
Faster R-CNN
(ResNeXt-101 64x4d-FPN…)
0.413
--
2
Faster R-CNN
(ResNeXt-101 64x4d-FPN…)
0.407
--
3
Faster R-CNN
(ResNeXt-101 32x4d-FPN…)
0.404
--
4
Faster R-CNN
(ResNeXt-101 32x4d-FPN…)
0.401
--
5
Faster R-CNN
(ResNet-101-FPN, 2x LR)
0.394
--
6
Faster R-CNN
(ResNet-101-FPN, 1x LR)
0.385
--
7
Faster R-CNN
(ResNet-50-FPN, 2x LR)
0.377
--
8
Faster R-CNN
(ResNet-50-FPN, 1x LR)
0.364
--