Cascade R-CNN

open-mmlab / mmdetection

Last updated on Feb 23, 2021
Cascade R-CNN (R-101-FPN, 1x, caffe)
Memory (M) 6200.0
Backbone Layers 101
File Size 337.69 MB
Training Data MS COCO
Training Resources 8x NVIDIA V100 GPUs
Training Time
Architecture RPN, FPN, Cascade R-CNN, ResNet, RoIAlign
lr sched 1x
Memory (M) 6200.0
Backbone Layers 101
SHOW MORE
SHOW LESS
Cascade R-CNN (R-101-FPN, 1x, pytorch)
Memory (M) 6400.0
inference time (s/im) 0.07407
File Size 337.68 MB
Training Data MS COCO
Training Resources 8x NVIDIA V100 GPUs
Training Time
Architecture RPN, FPN, Cascade R-CNN, ResNet, RoIAlign
lr sched 1x
Memory (M) 6400.0
Backbone Layers 101
inference time (s/im) 0.07407
SHOW MORE
SHOW LESS
Cascade R-CNN (R-101-FPN, 20e, pytorch)
lr sched 20e
Backbone Layers 101
File Size 337.69 MB
Training Data MS COCO
Training Resources 8x NVIDIA V100 GPUs
Training Time
Architecture RPN, FPN, Cascade R-CNN, ResNet, RoIAlign
lr sched 20e
Backbone Layers 101
SHOW MORE
SHOW LESS
Cascade R-CNN (R-50-FPN, 1x, caffe)
Memory (M) 4200.0
Backbone Layers 50
File Size 264.99 MB
Training Data MS COCO
Training Resources 8x NVIDIA V100 GPUs
Training Time
Architecture RPN, FPN, Cascade R-CNN, ResNet, RoIAlign
lr sched 1x
Memory (M) 4200.0
Backbone Layers 50
SHOW MORE
SHOW LESS
Cascade R-CNN (R-50-FPN, 1x, pytorch)
Memory (M) 4400.0
inference time (s/im) 0.06211
File Size 264.99 MB
Training Data MS COCO
Training Resources 8x NVIDIA V100 GPUs
Training Time
Architecture RPN, FPN, Cascade R-CNN, ResNet, RoIAlign
lr sched 1x
Memory (M) 4400.0
Backbone Layers 50
inference time (s/im) 0.06211
SHOW MORE
SHOW LESS
Cascade R-CNN (R-50-FPN, 20e, pytorch)
lr sched 20e
Backbone Layers 50
File Size 264.99 MB
Training Data MS COCO
Training Resources 8x NVIDIA V100 GPUs
Training Time
Architecture RPN, FPN, Cascade R-CNN, ResNet, RoIAlign
lr sched 20e
Backbone Layers 50
SHOW MORE
SHOW LESS
Cascade R-CNN (X-101-32x4d-FPN, 1x, pytorch)
Memory (M) 7600.0
inference time (s/im) 0.09174
File Size 336.39 MB
Training Data MS COCO
Training Resources 8x NVIDIA V100 GPUs
Training Time
Architecture RPN, ResNeXt, FPN, Cascade R-CNN, RoIAlign
lr sched 1x
Memory (M) 7600.0
Backbone Layers 101
inference time (s/im) 0.09174
SHOW MORE
SHOW LESS
Cascade R-CNN (X-101-32x4d-FPN, 20e, pytorch)
Memory (M) 7600.0
Backbone Layers 101
File Size 336.50 MB
Training Data MS COCO
Training Resources 8x NVIDIA V100 GPUs
Training Time
Architecture RPN, ResNeXt, FPN, Cascade R-CNN, RoIAlign
lr sched 20e
Memory (M) 7600.0
Backbone Layers 101
SHOW MORE
SHOW LESS
Cascade R-CNN (X-101-64x4d-FPN, 1x, pytorch)
Memory (M) 10700.0
Backbone Layers 101
File Size 486.48 MB
Training Data MS COCO
Training Resources 8x NVIDIA V100 GPUs
Training Time
Architecture RPN, ResNeXt, FPN, Cascade R-CNN, RoIAlign
lr sched 1x
Memory (M) 10700.0
Backbone Layers 101
SHOW MORE
SHOW LESS
Cascade R-CNN (X-101-64x4d-FPN, 20e, pytorch)
Memory (M) 10700.0
Backbone Layers 101
File Size 486.48 MB
Training Data MS COCO
Training Resources 8x NVIDIA V100 GPUs
Training Time
Architecture RPN, ResNeXt, FPN, Cascade R-CNN, RoIAlign
lr sched 20e
Memory (M) 10700.0
Backbone Layers 101
SHOW MORE
SHOW LESS
README.md

Cascade R-CNN: High Quality Object Detection and Instance Segmentation

Introduction

[ALGORITHM]

@article{Cai_2019,
   title={Cascade R-CNN: High Quality Object Detection and Instance Segmentation},
   ISSN={1939-3539},
   url={http://dx.doi.org/10.1109/tpami.2019.2956516},
   DOI={10.1109/tpami.2019.2956516},
   journal={IEEE Transactions on Pattern Analysis and Machine Intelligence},
   publisher={Institute of Electrical and Electronics Engineers (IEEE)},
   author={Cai, Zhaowei and Vasconcelos, Nuno},
   year={2019},
   pages={1–1}
}

Results and models

Cascade R-CNN

Backbone Style Lr schd Mem (GB) Inf time (fps) box AP Config Download
R-50-FPN caffe 1x 4.2 40.4 config model | log
R-50-FPN pytorch 1x 4.4 16.1 40.3 config model | log
R-50-FPN pytorch 20e - - 41.0 config model | log
R-101-FPN caffe 1x 6.2 42.3 config model | log
R-101-FPN pytorch 1x 6.4 13.5 42.0 config model | log
R-101-FPN pytorch 20e - - 42.5 config model | log
X-101-32x4d-FPN pytorch 1x 7.6 10.9 43.7 config model | log
X-101-32x4d-FPN pytorch 20e 7.6 43.7 config model | log
X-101-64x4d-FPN pytorch 1x 10.7 44.7 config model | log
X-101-64x4d-FPN pytorch 20e 10.7 44.5 config model | log

Cascade Mask R-CNN

Backbone Style Lr schd Mem (GB) Inf time (fps) box AP mask AP Config Download
R-50-FPN caffe 1x 5.9 41.2 36.0 config model | log
R-50-FPN pytorch 1x 6.0 11.2 41.2 35.9 config model | log
R-50-FPN pytorch 20e - - 41.9 36.5 config model | log
R-101-FPN caffe 1x 7.8 43.2 37.6 config model | log
R-101-FPN pytorch 1x 7.9 9.8 42.9 37.3 config model | log
R-101-FPN pytorch 20e - - 43.4 37.8 config model | log
X-101-32x4d-FPN pytorch 1x 9.2 8.6 44.3 38.3 config model | log
X-101-32x4d-FPN pytorch 20e 9.2 - 45.0 39.0 config model | log
X-101-64x4d-FPN pytorch 1x 12.2 6.7 45.3 39.2 config model | log
X-101-64x4d-FPN pytorch 20e 12.2 45.6 39.5 config model | log

Notes:

  • The 20e schedule in Cascade (Mask) R-CNN indicates decreasing the lr at 16 and 19 epochs, with a total of 20 epochs.

Results

Object Detection on COCO minival

Object Detection on COCO minival
MODEL BOX AP
Cascade R-CNN (X-101-64x4d-FPN, 1x, pytorch) 44.7
Cascade R-CNN (X-101-64x4d-FPN, 20e, pytorch) 44.5
Cascade R-CNN (X-101-32x4d-FPN, 1x, pytorch) 43.7
Cascade R-CNN (X-101-32x4d-FPN, 20e, pytorch) 43.7
Cascade R-CNN (R-101-FPN, 20e, pytorch) 42.5
Cascade R-CNN (R-101-FPN, 1x, caffe) 42.3
Cascade R-CNN (R-101-FPN, 1x, pytorch) 42.0
Cascade R-CNN (R-50-FPN, 20e, pytorch) 41.0
Cascade R-CNN (R-50-FPN, 1x, caffe) 40.4
Cascade R-CNN (R-50-FPN, 1x, pytorch) 40.3
Contact us on: hello@paperswithcode.com . Papers With Code is a free resource with all data licensed under CC-BY-SA.
Terms Data policy Cookies policy