Heedless Backbones

Vim 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

---------

ImageNet-1k

ImageNet-A

ImageNet-R

ImageNet-Sketch

ImageNet-C

ImageNet-C-bar

ImageNet-V2

ImageNet-ReaL

PASCAL VOC 2007 (val)

PASCAL VOC 2007 (test)

metric

Select an option

----------

Top-1

Top-5

GFLOPs

resolution

Select an option

----------

224x224

384x384

512x512

axis

Select an option

Results

Parameters (M)

Images / Second

GFLOPs

Publication Date

Filters

pretrain dataset

Select an option

---------

MegData73M

JFT-3B

JFT-300M

ImageNet-1k

ImageNet-22k

pretrain method

Select an option

----------

Supervised

Sup. + TL

FCMAE

MAE

CL

Legend

Legend attribute

Select an option

Family

Pretrain Dataset

Pretrain Method

Classification Resolution

Legend attribute (second)

Select an option

----------

Pretrain Dataset

Pretrain Method

Classification Resolution

No Results

model	params (m)	pretrain	finetune	gflops	IN-1k
Vim-Ti	7.0	IN-1k : Sup. : 300	— : — : —	None	76.1/—
Vim-Ti	7.0	IN-1k : Sup. : 300	IN-1k : 30 : 224	None	78.3/—
Vim-S	26.0	IN-1k : Sup. : 300	— : — : —	None	80.3/—
Vim-S	26.0	IN-1k : Sup. : 300	IN-1k : 30 : 224	None	81.4/—
Vim-B	98.0	IN-1k : Sup. : 300	— : — : —	None	81.9/—
Vim-B	98.0	IN-1k : Sup. : 300	IN-1k : 30 : 224	None	83.2/—

COCO (val)

model	pretrain	head	train	gflops	mAP^b	AP^b₅₀	AP^b₇₅	mAP^b_s	mAP^b_m	mAP^b_l
Vim-Ti	IN-1k : Sup. : 300	Cascade Mask R-CNN	COCO (train) : 200	None	45.7	63.9	49.6	26.1	49.0	63.2

COCO (val)

model	pretrain	head	train	gflops	mAP^m	AP^m₅₀	AP^m₇₅	mAP^m_s	mAP^m_m	mAP^m_l
Vim-Ti	IN-1k : Sup. : 300	Cascade Mask R-CNN	COCO (train) : 200	None	39.2	60.9	41.7	18.2	41.8	60.2

ADE20K (val)

model	pretrain	head	train	gflops	mIoU^ms	pAcc^ms	mAcc^ms	mIoU^ss	pAcc^ss	mAcc^ss
Vim-Ti	IN-1k : Sup. : 300	UPerNet	ADE20K (train) : 128 : 512	None	—	—	—	41.0	—	—
Vim-S	IN-1k : Sup. : 300	UPerNet	ADE20K (train) : 128 : 512	None	—	—	—	44.9	—	—