Heedless Backbones

Mamba-R 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

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
Mamba-R-T	9.0	IN-1k : Sup. : 300	— : — : —	None	77.4/—
Mamba-R-S	28.0	IN-1k : Sup. : 300	IN-1k : 100 : 224 → IN-1k : 20 : 224	None	81.4/—
Mamba-R-B	99.0	IN-1k : Sup. : 300	IN-1k : 100 : 224 → IN-1k : 20 : 224	None	83.0/—
Mamba-R-B	99.0	IN-1k : Sup. : 300	IN-1k : 100 : 224 → IN-1k : — : 384	None	84.3/—
Mamba-R-L	341.0	IN-1k : Sup. : 300	IN-1k : 100 : 224 → IN-1k : 20 : 224	None	83.6/—
Mamba-R-L	341.0	IN-1k : Sup. : 300	IN-1k : 100 : 224 → IN-1k : — : 384	None	84.5/—

ADE20K (val)

model	pretrain	head	train	gflops	mIoU^ms	pAcc^ms	mAcc^ms	mIoU^ss	pAcc^ss	mAcc^ss
Mamba-R-S	IN-1k : Sup. : 300	UPerNet	ADE20K (train) : 128 : 512	None	—	—	—	45.3	—	—
Mamba-R-B	IN-1k : Sup. : 300	UPerNet	ADE20K (train) : 128 : 512	None	—	—	—	47.7	—	—
Mamba-R-L	IN-1k : Sup. : 300	UPerNet	ADE20K (train) : 128 : 512	None	—	—	—	49.1	—	—