Dynamic R-CNN

open-mmlab / mmdetection

Last updated on Feb 23, 2021
Dynamic R-CNN (R-50, 1x, pytorch)
Memory (M) 3800.0
Backbone Layers 50
File Size 159.54 MB
Training Data MS COCO
Training Resources 8x NVIDIA V100 GPUs
Training Time
Architecture Dynamic SmoothL1 Loss, RPN, Non Maximum Suppression, Softmax, ResNet
lr sched 1x
Memory (M) 3800.0
Backbone Layers 50
SHOW MORE
SHOW LESS
README.md

Dynamic R-CNN: Towards High Quality Object Detection via Dynamic Training

Introduction

[ALGORITHM]

@article{DynamicRCNN,
    author = {Hongkai Zhang and Hong Chang and Bingpeng Ma and Naiyan Wang and Xilin Chen},
    title = {Dynamic {R-CNN}: Towards High Quality Object Detection via Dynamic Training},
    journal = {arXiv preprint arXiv:2004.06002},
    year = {2020}
}

Results and Models

Backbone Style Lr schd Mem (GB) Inf time (fps) box AP Config Download
R-50 pytorch 1x 3.8 38.9 config model | log

Results

Object Detection
BENCHMARK MODEL METRIC NAME METRIC VALUE GLOBAL RANK
COCO minival Dynamic R-CNN (R-50, 1x, pytorch) box AP 38.9 # 85
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