Heedless Backbones

RetinaNet

Paper | GitHub

axis

Select an option

Results

Parameters (M)

Images / Second

Publication Date

task

Select an option

---------

Object Detection

dataset

Select an option

---------

COCO (val)

COCO (test)

metric

Select an option

----------

mAP

AP₅₀

AP₇₅

mAP_s

mAP_m

mAP_l

GFLOPs

axis

Select an option

Results

Parameters (M)

Images / Second

GFLOPs

Publication Date

Filters

pretrain dataset

Select an option

---------

JFT-3B

ImageNet-1k

ImageNet-22k

JFT-300M

pretrain method

Select an option

----------

Supervised

FCMAE

MAE

CL

Legend

Legend attribute

Select an option

Family

Pretrain Dataset

Pretrain Method

Instance Head

Instance Training Epochs

Legend attribute (second)

Select an option

----------

Pretrain Dataset

Pretrain Method

Instance Head

Instance Training Epochs

family	model	params (m)	pretrain	head	train	GFLOPs	mAP
MogaNet	MogaNet-XT	3.0	IN-1k : Sup. : 300	RetinaNet	COCO (train) : 12	167.0	39.7
MogaNet	MogaNet-T	5.2	IN-1k : Sup. : 300	RetinaNet	COCO (train) : 12	173.0	41.4
MogaNet	MogaNet-S	25.3	IN-1k : Sup. : 300	RetinaNet	COCO (train) : 12	253.0	45.8

COCO (val)

family	model	pretrain	train	gflops	mAP^b	AP^b₅₀	AP^b₇₅	mAP^b_s	mAP^b_m	mAP^b_l
MogaNet	MogaNet-XT	IN-1k : Sup. : 300	COCO (train) : 12	167.0	39.7	60.0	42.4	23.8	43.6	51.7
MogaNet	MogaNet-T	IN-1k : Sup. : 300	COCO (train) : 12	173.0	41.4	61.5	44.4	25.1	45.7	53.6
MogaNet	MogaNet-S	IN-1k : Sup. : 300	COCO (train) : 12	253.0	45.8	66.6	49.0	29.1	50.1	59.8