SSD is simply relative to methods that require object proposals because it eliminates proposal generation and subsequent pixel or feature resampling stages and encapsulates all computation in a single network
SSD discretizes the output space of bounding boxes into a set of default boxes over different aspect ratios and scales per future map location
Experiment Table
Backbone
Dataset
Training dataset
Valid dataset
Image size
mAP
mAP_50
mAP_75
Original paper
PASCAL VOC
trainval 2007+2012
test2007
300x300
--
74.3
--
Our implementation
PASCAL VOC
trainval 2007+2012
test2007
300x300
3.87
66.9
39.9
Dataset
Download Pascal VOC train+val 2012+2007
Download Pascal VOC test 2007
s
Put all images, annotations, and .txt files in dataset/VOC folder as follows: