Heedless Backbones

HybridNet Family

Paper | GitHub

axis

Select an option

Results

Parameters (M)

Images / Second

Publication Date

task

Select an option

---------

Object Detection

Instance Segmentation

Classification

Semantic Segmentation

Panoptic Segmentation

dataset

Select an option

---------

Cityscapes (val)

Cityscapes (test)

ADE20K (val)

ADE20K (test)

PASCAL VOC 2007 (val)

PASCAL VOC 2007 (test)

metric

Select an option

mIoU^ms

pAcc^ms

mAcc^ms

mIoU^ss

pAcc^ss

mAcc^ss

GFLOPs

head

Select an option

---------

UPerNet

Mask2Former

Panoptic FPN

SETR

resolution

Select an option

----------

512x2048

640x2560

axis

Select an option

Results

Parameters (M)

Images / Second

GFLOPs

Publication Date

Filters

pretrain dataset

Select an option

---------

ImageNet-1k

ImageNet-22k

JFT-300M

JFT-3B

MegData73M

pretrain method

Select an option

----------

Supervised

Sup. + TL

FCMAE

MAE

CL

MAP

Legend

Legend attribute

Select an option

Family

Pretrain Dataset

Semantic Segmentation Head

Semantic Segmentation Resolution

Semantic Segmentation Training Epochs

Legend attribute (second)

Select an option

----------

Family

Pretrain Method

Semantic Segmentation Head

Semantic Segmentation Resolution

Semantic Segmentation Training Epochs

No Results

model	params (m)	pretrain	finetune	gflops	IN-1k
HybridNet-T	12.0	IN-1k : Sup. : 300	— : — : —	7.6	77.7/—
HybridNet-T	12.0	IN-1k : MAP : 1600	IN-1k : 400 : 224	7.6	78.6/—
HybridNet-S	37.0	IN-1k : Sup. : 300	— : — : —	14.6	81.3/—
HybridNet-S	37.0	IN-1k : MAP : 1600	IN-1k : 400 : 224	14.6	82.5/—
HybridNet-B	128.0	IN-1k : Sup. : 300	— : — : —	76.1	84.5/—
HybridNet-B	128.0	IN-1k : MAP : 1600	IN-1k : 400 : 224	30.0	84.9/—
HybridNet-B	128.0	IN-1k : MAP : 1600	IN-1k : 400 : 384	76.1	85.5/—
HybridNet-L	443.0	IN-1k : Sup. : 300	— : — : —	None	84.6/—
HybridNet-L	443.0	IN-1k : MAP : 1600	IN-1k : 400 : 224	78.3	85.0/—
HybridNet-L	443.0	IN-1k : MAP : 1600	IN-1k : 400 : 384	None	86.2/—

COCO (val)

model	pretrain	head	train	gflops	mAP^b	AP^b₅₀	AP^b₇₅	mAP^b_s	mAP^b_m	mAP^b_l
HybridNet-T	IN-1k : MAP : 1600	Cascade Mask R-CNN	COCO (train) : 36	None	46.4	—	—	—	—	—

COCO (val)

model	pretrain	head	train	gflops	mAP^m	AP^m₅₀	AP^m₇₅	mAP^m_s	mAP^m_m	mAP^m_l
HybridNet-T	IN-1k : MAP : 1600	Cascade Mask R-CNN	COCO (train) : 36	None	39.8	—	—	—	—	—

ADE20K (val)

model	pretrain	head	train	gflops	mIoU^ms	pAcc^ms	mAcc^ms	mIoU^ss	pAcc^ss	mAcc^ss
HybridNet-S	IN-1k : Sup. : 300	UPerNet	ADE20K (train) : 128 : 512	None	—	—	—	45.6	—	—
HybridNet-S	IN-1k : MAP : 1600	UPerNet	ADE20K (train) : 128 : 512	None	—	—	—	46.9	—	—