RetinaNet

open-mmlab / mmdetection

Last updated on Feb 23, 2021
RetinaNet (R-101-FPN, 1x, caffe)
Memory (M) 5500.0
inference time (s/im) 0.06803
File Size 217.80 MB
Training Data MS COCO
Training Resources 8x NVIDIA V100 GPUs
Training Time
Architecture ResNet, FPN, Focal Loss
lr sched 1x
Memory (M) 5500.0
Backbone Layers 101
inference time (s/im) 0.06803
SHOW MORE
SHOW LESS
RetinaNet (R-101-FPN, 1x, pytorch)
Memory (M) 5700.0
inference time (s/im) 0.06667
File Size 217.80 MB
Training Data MS COCO
Training Resources 8x NVIDIA V100 GPUs
Training Time
Architecture ResNet, FPN, Focal Loss
lr sched 1x
Memory (M) 5700.0
Backbone Layers 101
inference time (s/im) 0.06667
SHOW MORE
SHOW LESS
RetinaNet (R-101-FPN, 2x, pytorch)
lr sched 2x
Backbone Layers 101
File Size 217.80 MB
Training Data MS COCO
Training Resources 8x NVIDIA V100 GPUs
Training Time
Architecture ResNet, FPN, Focal Loss
lr sched 2x
Backbone Layers 101
SHOW MORE
SHOW LESS
RetinaNet (R-50-FPN, 1x, caffe)
Memory (M) 3500.0
inference time (s/im) 0.05376
File Size 145.10 MB
Training Data MS COCO
Training Resources 8x NVIDIA V100 GPUs
Training Time
Architecture ResNet, FPN, Focal Loss
lr sched 1x
Memory (M) 3500.0
Backbone Layers 50
inference time (s/im) 0.05376
SHOW MORE
SHOW LESS
RetinaNet (R-50-FPN, 1x, pytorch)
Memory (M) 3800.0
inference time (s/im) 0.05263
File Size 145.10 MB
Training Data MS COCO
Training Resources 8x NVIDIA V100 GPUs
Training Time
Architecture ResNet, FPN, Focal Loss
lr sched 1x
Memory (M) 3800.0
Backbone Layers 50
inference time (s/im) 0.05263
SHOW MORE
SHOW LESS
RetinaNet (R-50-FPN, 2x, pytorch)
lr sched 2x
Backbone Layers 50
File Size 145.10 MB
Training Data MS COCO
Training Resources 8x NVIDIA V100 GPUs
Training Time
Architecture ResNet, FPN, Focal Loss
lr sched 2x
Backbone Layers 50
SHOW MORE
SHOW LESS
RetinaNet (X-101-32x4d-FPN, 1x, pytorch)
Memory (M) 7000.0
inference time (s/im) 0.08264
File Size 216.50 MB
Training Data MS COCO
Training Resources 8x NVIDIA V100 GPUs
Training Time
Architecture ResNeXt, FPN, Focal Loss
lr sched 1x
Memory (M) 7000.0
Backbone Layers 101
inference time (s/im) 0.08264
SHOW MORE
SHOW LESS
RetinaNet (X-101-32x4d-FPN, 2x, pytorch)
lr sched 2x
Backbone Layers 101
File Size 216.50 MB
Training Data MS COCO
Training Resources 8x NVIDIA V100 GPUs
Training Time
Architecture ResNeXt, FPN, Focal Loss
lr sched 2x
Backbone Layers 101
SHOW MORE
SHOW LESS
RetinaNet (X-101-64x4d-FPN, 1x, pytorch)
Memory (M) 10000.0
inference time (s/im) 0.11494
File Size 366.58 MB
Training Data MS COCO
Training Resources 8x NVIDIA V100 GPUs
Training Time
Architecture ResNeXt, FPN, Focal Loss
lr sched 1x
Memory (M) 10000.0
Backbone Layers 101
inference time (s/im) 0.11494
SHOW MORE
SHOW LESS
RetinaNet (X-101-64x4d-FPN, 2x, pytorch)
lr sched 2x
Backbone Layers 101
File Size 366.58 MB
Training Data MS COCO
Training Resources 8x NVIDIA V100 GPUs
Training Time
Architecture ResNeXt, FPN, Focal Loss
lr sched 2x
Backbone Layers 101
SHOW MORE
SHOW LESS
README.md

Focal Loss for Dense Object Detection

Introduction

[ALGORITHM]

@inproceedings{lin2017focal,
  title={Focal loss for dense object detection},
  author={Lin, Tsung-Yi and Goyal, Priya and Girshick, Ross and He, Kaiming and Doll{\'a}r, Piotr},
  booktitle={Proceedings of the IEEE international conference on computer vision},
  year={2017}
}

Results and models

Backbone Style Lr schd Mem (GB) Inf time (fps) box AP Config Download
R-50-FPN caffe 1x 3.5 18.6 36.3 config model | log
R-50-FPN pytorch 1x 3.8 19.0 36.5 config model | log
R-50-FPN pytorch 2x - - 37.4 config model | log
R-101-FPN caffe 1x 5.5 14.7 38.5 config model | log
R-101-FPN pytorch 1x 5.7 15.0 38.5 config model | log
R-101-FPN pytorch 2x - - 38.9 config model | log
X-101-32x4d-FPN pytorch 1x 7.0 12.1 39.9 config model | log
X-101-32x4d-FPN pytorch 2x - - 40.1 config model | log
X-101-64x4d-FPN pytorch 1x 10.0 8.7 41.0 config model | log
X-101-64x4d-FPN pytorch 2x - - 40.8 config model | log

Results

Object Detection on COCO minival

Object Detection on COCO minival
MODEL BOX AP
RetinaNet (X-101-64x4d-FPN, 1x, pytorch) 41.0
RetinaNet (X-101-64x4d-FPN, 2x, pytorch) 40.8
RetinaNet (X-101-32x4d-FPN, 2x, pytorch) 40.1
RetinaNet (X-101-32x4d-FPN, 1x, pytorch) 39.9
RetinaNet (R-101-FPN, 2x, pytorch) 38.9
RetinaNet (R-101-FPN, 1x, pytorch) 38.5
RetinaNet (R-101-FPN, 1x, caffe) 38.5
RetinaNet (R-50-FPN, 2x, pytorch) 37.4
RetinaNet (R-50-FPN, 1x, pytorch) 36.5
RetinaNet (R-50-FPN, 1x, caffe) 36.3
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