Heedless Backbones

U-HaloNet 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

Legend

Legend attribute

Select an option

Family

Pretrain Dataset

Classification Resolution

Legend attribute (second)

Select an option

----------

Family

Pretrain Method

Classification Resolution

model	params (m)	pretrain	finetune	GFLOPs	Top-1
U-HaloNet-Micro	4.4	IN-1k : Sup. : 300	— : — : —	0.65	75.8
U-HaloNet-Tiny	31.5	IN-1k : Sup. : 300	— : — : —	4.75	83.0
U-HaloNet-Small	52.8	IN-1k : Sup. : 300	— : — : —	8.92	84.0
U-HaloNet-Base	93.3	IN-1k : Sup. : 300	— : — : —	15.84	84.6

model	params (m)	pretrain	finetune	gflops	IN-1k
U-HaloNet-Micro	4.4	IN-1k : Sup. : 300	— : — : —	0.65	75.8/—
U-HaloNet-Tiny	31.5	IN-1k : Sup. : 300	— : — : —	4.75	83.0/—
U-HaloNet-Small	52.8	IN-1k : Sup. : 300	— : — : —	8.92	84.0/—
U-HaloNet-Base	93.3	IN-1k : Sup. : 300	— : — : —	15.84	84.6/—

COCO (val)

model	pretrain	head	train	gflops	mAP^b	AP^b₅₀	AP^b₇₅	mAP^b_s	mAP^b_m	mAP^b_l
U-HaloNet-Micro	IN-1k : Sup. : 300	Mask R-CNN	COCO (train) : 12	None	40.3	—	—	24.5	43.0	53.7
U-HaloNet-Tiny	IN-1k : Sup. : 300	Mask R-CNN	COCO (train) : 12	None	46.9	—	—	30.0	50.2	61.4
U-HaloNet-Small	IN-1k : Sup. : 300	Mask R-CNN	COCO (train) : 12	None	48.2	—	—	31.2	52.5	63.2
U-HaloNet-Base	IN-1k : Sup. : 300	Mask R-CNN	COCO (train) : 12	None	49.0	—	—	31.8	53.0	63.7

COCO (val)

model	pretrain	head	train	gflops	mAP^m	AP^m₅₀	AP^m₇₅	mAP^m_s	mAP^m_m	mAP^m_l
U-HaloNet-Micro	IN-1k : Sup. : 300	Mask R-CNN	COCO (train) : 12	None	37.3	—	—	18.7	39.7	55.2
U-HaloNet-Tiny	IN-1k : Sup. : 300	Mask R-CNN	COCO (train) : 12	None	42.4	—	—	22.6	45.3	60.9
U-HaloNet-Small	IN-1k : Sup. : 300	Mask R-CNN	COCO (train) : 12	None	43.3	—	—	23.4	47.0	62.3
U-HaloNet-Base	IN-1k : Sup. : 300	Mask R-CNN	COCO (train) : 12	None	43.8	—	—	24.4	47.0	62.8