SSD

open-mmlab / mmdetection

Last updated on Feb 23, 2021
SSD (VGG16, 120e, caffe, 300)
Memory (M) 10200.0
inference time (s/im) 0.02288
File Size 130.88 MB
Training Data MS COCO
Training Resources 8x NVIDIA V100 GPUs
Training Time
Architecture SSD, Non-Maximum Suppression
lr sched 120e
Memory (M) 10200.0
inference time (s/im) 0.02288
SHOW MORE
SHOW LESS
SSD (VGG16, 120e, caffe, 512)
Memory (M) 9300.0
inference time (s/im) 0.03257
File Size 137.49 MB
Training Data MS COCO
Training Resources 8x NVIDIA V100 GPUs
Training Time
Architecture SSD, Non-Maximum Suppression
lr sched 120e
Memory (M) 9300.0
inference time (s/im) 0.03257
SHOW MORE
SHOW LESS
README.md

SSD: Single Shot MultiBox Detector

Introduction

[ALGORITHM]

@article{Liu_2016,
   title={SSD: Single Shot MultiBox Detector},
   journal={ECCV},
   author={Liu, Wei and Anguelov, Dragomir and Erhan, Dumitru and Szegedy, Christian and Reed, Scott and Fu, Cheng-Yang and Berg, Alexander C.},
   year={2016},
}

Results and models

Backbone Size Style Lr schd Mem (GB) Inf time (fps) box AP Config Download
VGG16 300 caffe 120e 10.2 43.7 25.6 config model | log
VGG16 512 caffe 120e 9.3 30.7 29.4 config model | log

Results

Object Detection on COCO minival

Object Detection
BENCHMARK MODEL METRIC NAME METRIC VALUE GLOBAL RANK
COCO minival SSD (VGG16, 120e, caffe, 512) box AP 29.4 # 118
COCO minival SSD (VGG16, 120e, caffe, 300) box AP 25.6 # 120
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