TASK	DATASET	MODEL	METRIC NAME	METRIC VALUE	GLOBAL RANK
Face Alignment	300W	HRNet	NME_inter-ocular (%, Full)	3.32	# 24
Face Alignment	300W	HRNet	NME_inter-ocular (%, Common)	2.87	# 21
Face Alignment	300W	HRNet	NME_inter-ocular (%, Challenge)	5.15	# 24
Instance Segmentation	BDD100K val	HRNet	AP	22.5	# 2
Semantic Segmentation	Cityscapes test	HRNetV2 (train+val)	Mean IoU (class)	81.6%	# 38
Semantic Segmentation	Cityscapes val	HRNetV2 (HRNetV2-W48)	mIoU	81.1	# 44
Semantic Segmentation	Cityscapes val	HRNetV2 (HRNetV2-W40)	mIoU	80.2	# 54
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W18)	box AP	41.3	# 164
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W18)	AP50	59.2	# 97
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W18)	AP75	44.9	# 74
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W18)	APS	23.7	# 69
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W18)	APM	44.2	# 69
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W18)	APL	54.1	# 72
Object Detection	COCO minival	HTC (HRNetV2p-W18)	box AP	43.1	# 144
Object Detection	COCO minival	HTC (HRNetV2p-W18)	APS	26.6	# 42
Object Detection	COCO minival	HTC (HRNetV2p-W18)	APM	46.0	# 55
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W32)	box AP	42.3	# 153
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W32)	APS	25.0	# 59
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W32)	APM	45.4	# 62
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W48)	box AP	44.6	# 124
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W48)	AP50	62.7	# 69
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W48)	AP75	48.7	# 42
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W48)	APS	26.3	# 45
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W48)	APM	48.1	# 39
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W48)	APL	58.5	# 52
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W32)	box AP	43.7	# 136
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W32)	AP50	61.7	# 76
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W32)	AP75	47.7	# 49
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W32)	APS	25.6	# 52
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W32)	APM	46.5	# 52
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W32)	APL	57.4	# 59
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W48)	box AP	41.8	# 159
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W48)	AP50	62.8	# 67
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W48)	AP75	45.9	# 67
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W48)	APM	44.7	# 64
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W48)	APL	54.6	# 71
Object Detection	COCO minival	HTC (HRNetV2p-W32)	box AP	45.3	# 113
Object Detection	COCO minival	HTC (HRNetV2p-W32)	APS	27.0	# 36
Object Detection	COCO minival	HTC (HRNetV2p-W32)	APM	48.4	# 32
Object Detection	COCO minival	HTC (HRNetV2p-W32)	APL	59.5	# 44
Instance Segmentation	COCO minival	HTC (HRNetV2p-W48)	mask AP	41.0	# 68
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W18)	box AP	39.2	# 189
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W18)	APM	41.7	# 83
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W18)	APL	51.0	# 85
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W32, cascade)	APS	26.1	# 46
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W32, cascade)	APM	47.9	# 42
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W48, cascade)	box AP	46.0	# 108
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W48, cascade)	APS	27.5	# 33
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W48, cascade)	APL	60.1	# 40
Object Detection	COCO minival	HTC (HRNetV2p-W48)	box AP	47.0	# 98
Object Detection	COCO minival	HTC (HRNetV2p-W48)	APS	28.8	# 26
Object Detection	COCO minival	HTC (HRNetV2p-W48)	APM	50.3	# 25
Object Detection	COCO minival	HTC (HRNetV2p-W48)	APL	62.2	# 31
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W18)	box AP	38.0	# 198
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W18)	AP50	58.9	# 101
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W18)	AP75	41.5	# 94
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W18)	APS	22.6	# 79
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W18)	APM	40.8	# 86
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W18)	APL	49.6	# 90
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W32)	box AP	40.9	# 169
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W32)	AP50	61.8	# 75
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W32)	AP75	44.8	# 75
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W32)	APS	24.4	# 63
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W32)	APM	43.7	# 73
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W32)	APL	53.3	# 75
Object Detection	COCO test-dev	Mask R-CNN (HRNetV2p-W48 + cascade)	box mAP	46.1	# 127
Object Detection	COCO test-dev	Mask R-CNN (HRNetV2p-W48 + cascade)	AP50	64.0	# 96
Object Detection	COCO test-dev	Mask R-CNN (HRNetV2p-W48 + cascade)	AP75	50.3	# 80
Object Detection	COCO test-dev	Mask R-CNN (HRNetV2p-W48 + cascade)	APS	27.1	# 77
Object Detection	COCO test-dev	Mask R-CNN (HRNetV2p-W48 + cascade)	APM	48.6	# 74
Object Detection	COCO test-dev	Mask R-CNN (HRNetV2p-W48 + cascade)	APL	58.3	# 69
Object Detection	COCO test-dev	Mask R-CNN (HRNetV2p-W48 + cascade)	Hardware Burden	15G	# 1
Object Detection	COCO test-dev	Mask R-CNN (HRNetV2p-W48 + cascade)	Operations per network pass	61.8G	# 1
Object Detection	COCO test-dev	Cascade R-CNN (HRNetV2p-W48)	AP75	48.6	# 86
Object Detection	COCO test-dev	Cascade R-CNN (HRNetV2p-W48)	APS	26.0	# 89
Object Detection	COCO test-dev	Cascade R-CNN (HRNetV2p-W48)	APM	47.3	# 85
Object Detection	COCO test-dev	Cascade R-CNN (HRNetV2p-W48)	APL	56.3	# 89
Object Detection	COCO test-dev	Faster R-CNN (HRNetV2p-W48)	box mAP	42.4	# 172
Object Detection	COCO test-dev	Faster R-CNN (HRNetV2p-W48)	AP50	63.6	# 100
Object Detection	COCO test-dev	Faster R-CNN (HRNetV2p-W48)	AP75	46.4	# 113
Object Detection	COCO test-dev	Faster R-CNN (HRNetV2p-W48)	APS	24.9	# 97
Object Detection	COCO test-dev	Faster R-CNN (HRNetV2p-W48)	APM	44.6	# 115
Object Detection	COCO test-dev	Faster R-CNN (HRNetV2p-W48)	APL	53.0	# 120
Object Detection	COCO test-dev	Faster R-CNN (HRNetV2p-W48)	Hardware Burden	16G	# 1
Object Detection	COCO test-dev	Faster R-CNN (HRNetV2p-W48)	Operations per network pass	20.8G	# 1
Object Detection	COCO test-dev	FCOS (HRNetV2p-W48)	box mAP	40.5	# 190
Object Detection	COCO test-dev	FCOS (HRNetV2p-W48)	AP50	59.3	# 140
Object Detection	COCO test-dev	FCOS (HRNetV2p-W48)	APS	23.4	# 113
Object Detection	COCO test-dev	FCOS (HRNetV2p-W48)	APM	42.6	# 130
Object Detection	COCO test-dev	FCOS (HRNetV2p-W48)	APL	51.0	# 133
Object Detection	COCO test-dev	FCOS (HRNetV2p-W48)	Hardware Burden	16G	# 1
Object Detection	COCO test-dev	FCOS (HRNetV2p-W48)	Operations per network pass	27.3G	# 1
Object Detection	COCO test-dev	Mask R-CNN (HRNetV2p-W32 + cascade)	AP50	62.5	# 111
Object Detection	COCO test-dev	Mask R-CNN (HRNetV2p-W32 + cascade)	AP75	48.6	# 86
Object Detection	COCO test-dev	Mask R-CNN (HRNetV2p-W32 + cascade)	APL	56.3	# 89
Object Detection	COCO test-dev	Mask R-CNN (HRNetV2p-W32 + cascade)	Hardware Burden	16G	# 1
Object Detection	COCO test-dev	Mask R-CNN (HRNetV2p-W32 + cascade)	Operations per network pass	50.6G	# 1
Object Detection	COCO test-dev	CenterNet (HRNetV2-W48)	box mAP	43.5	# 154
Object Detection	COCO test-dev	CenterNet (HRNetV2-W48)	AP75	46.5	# 111
Object Detection	COCO test-dev	CenterNet (HRNetV2-W48)	APS	22.2	# 122
Object Detection	COCO test-dev	CenterNet (HRNetV2-W48)	APL	57.8	# 74
Object Detection	COCO test-dev	CenterNet (HRNetV2-W48)	Hardware Burden	16G	# 1
Object Detection	COCO test-dev	CenterNet (HRNetV2-W48)	Operations per network pass	21.7G	# 1
Object Detection	COCO test-dev	HTC (HRNetV2p-W48)	box mAP	47.3	# 117
Object Detection	COCO test-dev	HTC (HRNetV2p-W48)	AP50	65.9	# 73
Object Detection	COCO test-dev	HTC (HRNetV2p-W48)	AP75	51.2	# 67
Object Detection	COCO test-dev	HTC (HRNetV2p-W48)	APS	28.0	# 69
Object Detection	COCO test-dev	HTC (HRNetV2p-W48)	APM	49.7	# 65
Object Detection	COCO test-dev	HTC (HRNetV2p-W48)	APL	59.8	# 63
Object Detection	COCO test-dev	HTC (HRNetV2p-W48)	Hardware Burden	15G	# 1
Object Detection	COCO test-dev	HTC (HRNetV2p-W48)	Operations per network pass	71.7G	# 1
Face Alignment	COFW	HRNet	NME (inter-ocular)	3.45	# 12
Face Alignment	COFW-68	HRNetV2-W18	NME (inter-ocular)	5.06	# 6
Semantic Segmentation	DADA-seg	HRNet (ACDC)	mIoU	27.5	# 9
Dichotomous Image Segmentation	DIS-TE1	HRNet	max F-Measure	0.668	# 11
Dichotomous Image Segmentation	DIS-TE1	HRNet	weighted F-measure	0.579	# 10
Dichotomous Image Segmentation	DIS-TE1	HRNet	MAE	0.088	# 10
Dichotomous Image Segmentation	DIS-TE1	HRNet	S-Measure	0.742	# 10
Dichotomous Image Segmentation	DIS-TE1	HRNet	E-measure	0.797	# 11
Dichotomous Image Segmentation	DIS-TE1	HRNet	HCE	262	# 16
Dichotomous Image Segmentation	DIS-TE2	HRNet	max F-Measure	0.747	# 10
Dichotomous Image Segmentation	DIS-TE2	HRNet	weighted F-measure	0.664	# 10
Dichotomous Image Segmentation	DIS-TE2	HRNet	MAE	0.087	# 10
Dichotomous Image Segmentation	DIS-TE2	HRNet	S-Measure	0.784	# 10
Dichotomous Image Segmentation	DIS-TE2	HRNet	E-measure	0.840	# 7
Dichotomous Image Segmentation	DIS-TE2	HRNet	HCE	555	# 14
Dichotomous Image Segmentation	DIS-TE3	HRNet	max F-Measure	0.784	# 10
Dichotomous Image Segmentation	DIS-TE3	HRNet	weighted F-measure	0.700	# 10
Dichotomous Image Segmentation	DIS-TE3	HRNet	MAE	0.080	# 10
Dichotomous Image Segmentation	DIS-TE3	HRNet	S-Measure	0.805	# 9
Dichotomous Image Segmentation	DIS-TE3	HRNet	E-measure	0.869	# 8
Dichotomous Image Segmentation	DIS-TE3	HRNet	HCE	1049	# 13
Dichotomous Image Segmentation	DIS-TE4	HRNet	max F-Measure	0.772	# 10
Dichotomous Image Segmentation	DIS-TE4	HRNet	weighted F-measure	0.687	# 10
Dichotomous Image Segmentation	DIS-TE4	HRNet	MAE	0.092	# 10
Dichotomous Image Segmentation	DIS-TE4	HRNet	S-Measure	0.792	# 10
Dichotomous Image Segmentation	DIS-TE4	HRNet	E-measure	0.854	# 7
Dichotomous Image Segmentation	DIS-TE4	HRNet	HCE	3864	# 20
Dichotomous Image Segmentation	DIS-VD	HRNet	max F-Measure	0.726	# 12
Dichotomous Image Segmentation	DIS-VD	HRNet	weighted F-measure	0.641	# 10
Dichotomous Image Segmentation	DIS-VD	HRNet	MAE	0.095	# 11
Dichotomous Image Segmentation	DIS-VD	HRNet	S-Measure	0.767	# 12
Dichotomous Image Segmentation	DIS-VD	HRNet	E-measure	0.824	# 9
Dichotomous Image Segmentation	DIS-VD	HRNet	HCE	1560	# 14
Thermal Image Segmentation	MFN Dataset	HRNet	mIOU	51.7	# 40
Semantic Segmentation	PASCAL Context	CFNet (ResNet-101)	mIoU	54.0	# 35
Semantic Segmentation	PASCAL Context	HRNetV2 HRNetV2-W48	mIoU	54	# 35
Semantic Segmentation	Potsdam	HRNet-18	mIoU	84.02	# 2
Semantic Segmentation	Potsdam	HRNet-48	mIoU	84.22	# 1
Semantic Segmentation	US3D	HRNet-48	mIoU	72.66	# 2
Semantic Segmentation	US3D	HRNet-18	mIoU	60.33	# 3
Semantic Segmentation	Vaihingen	HRNet-48	mIoU	76.75	# 8
Semantic Segmentation	Vaihingen	HRNet-18	mIoU	75.90	# 10
Face Alignment	WFLW	HRNet	NME (inter-ocular)	4.60	# 21

Badge	Markdown
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/190807919/object-detection-on-coco)](https://paperswithcode.com/sota/object-detection-on-coco?p=190807919)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/190807919/semantic-segmentation-on-potsdam-1)](https://paperswithcode.com/sota/semantic-segmentation-on-potsdam-1?p=190807919)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/190807919/instance-segmentation-on-bdd100k-val)](https://paperswithcode.com/sota/instance-segmentation-on-bdd100k-val?p=190807919)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/190807919/semantic-segmentation-on-us3d-1)](https://paperswithcode.com/sota/semantic-segmentation-on-us3d-1?p=190807919)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/190807919/face-alignment-on-cofw-68)](https://paperswithcode.com/sota/face-alignment-on-cofw-68?p=190807919)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/190807919/semantic-segmentation-on-vaihingen)](https://paperswithcode.com/sota/semantic-segmentation-on-vaihingen?p=190807919)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/190807919/semantic-segmentation-on-dada-seg)](https://paperswithcode.com/sota/semantic-segmentation-on-dada-seg?p=190807919)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/190807919/dichotomous-image-segmentation-on-dis-te2)](https://paperswithcode.com/sota/dichotomous-image-segmentation-on-dis-te2?p=190807919)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/190807919/dichotomous-image-segmentation-on-dis-te3)](https://paperswithcode.com/sota/dichotomous-image-segmentation-on-dis-te3?p=190807919)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/190807919/dichotomous-image-segmentation-on-dis-te4)](https://paperswithcode.com/sota/dichotomous-image-segmentation-on-dis-te4?p=190807919)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/190807919/dichotomous-image-segmentation-on-dis-te1)](https://paperswithcode.com/sota/dichotomous-image-segmentation-on-dis-te1?p=190807919)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/190807919/face-alignment-on-cofw)](https://paperswithcode.com/sota/face-alignment-on-cofw?p=190807919)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/190807919/dichotomous-image-segmentation-on-dis-vd)](https://paperswithcode.com/sota/dichotomous-image-segmentation-on-dis-vd?p=190807919)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/190807919/face-alignment-on-wflw)](https://paperswithcode.com/sota/face-alignment-on-wflw?p=190807919)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/190807919/face-alignment-on-300w)](https://paperswithcode.com/sota/face-alignment-on-300w?p=190807919)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/190807919/semantic-segmentation-on-pascal-context)](https://paperswithcode.com/sota/semantic-segmentation-on-pascal-context?p=190807919)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/190807919/semantic-segmentation-on-cityscapes)](https://paperswithcode.com/sota/semantic-segmentation-on-cityscapes?p=190807919)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/190807919/thermal-image-segmentation-on-mfn-dataset)](https://paperswithcode.com/sota/thermal-image-segmentation-on-mfn-dataset?p=190807919)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/190807919/object-detection-on-coco-minival)](https://paperswithcode.com/sota/object-detection-on-coco-minival?p=190807919)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/190807919/semantic-segmentation-on-cityscapes-val)](https://paperswithcode.com/sota/semantic-segmentation-on-cityscapes-val?p=190807919)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/190807919/instance-segmentation-on-coco-minival)](https://paperswithcode.com/sota/instance-segmentation-on-coco-minival?p=190807919)`

Deep High-Resolution Representation Learning for Visual Recognition

20 Aug 2019 · Jingdong Wang, Ke Sun, Tianheng Cheng, Borui Jiang, Chaorui Deng, Yang Zhao, Dong Liu, Yadong Mu, Mingkui Tan, Xinggang Wang, Wenyu Liu, Bin Xiao ·

High-resolution representations are essential for position-sensitive vision problems, such as human pose estimation, semantic segmentation, and object detection. Existing state-of-the-art frameworks first encode the input image as a low-resolution representation through a subnetwork that is formed by connecting high-to-low resolution convolutions \emph{in series} (e.g., ResNet, VGGNet), and then recover the high-resolution representation from the encoded low-resolution representation. Instead, our proposed network, named as High-Resolution Network (HRNet), maintains high-resolution representations through the whole process. There are two key characteristics: (i) Connect the high-to-low resolution convolution streams \emph{in parallel}; (ii) Repeatedly exchange the information across resolutions. The benefit is that the resulting representation is semantically richer and spatially more precise. We show the superiority of the proposed HRNet in a wide range of applications, including human pose estimation, semantic segmentation, and object detection, suggesting that the HRNet is a stronger backbone for computer vision problems. All the codes are available at~{\url{https://github.com/HRNet}}.

PDF Abstract

Code

Add Remove Mark official

open-mmlab/mmdetection

30,827

PaddlePaddle/PaddleDetection

13,319

open-mmlab/mmsegmentation

↳ Quickstart in

Colab

8,817

PaddlePaddle/PaddleSeg

8,417

open-mmlab/mmpose

6,349

See all 42 implementations

Tasks

Add Remove

Dichotomous Image Segmentation

Face Alignment

Instance Segmentation

Object Detection

Pose Estimation

Representation Learning

Semantic Segmentation

Thermal Image Segmentation

Vocal Bursts Intensity Prediction

Datasets

MS COCO

Cityscapes

BDD100K

PASCAL Context

300W

COFW

WFLW MFNet

DIS5K

DADA-seg Potsdam

Results from the Paper

Edit

Ranked #1 on Object Detection on COCO test-dev (Hardware Burden metric)

Get a GitHub badge

Task	Dataset	Model	Metric Name	Metric Value	Global Rank	Benchmark
Instance Segmentation	BDD100K val	HRNet	AP	22.5	# 2	Compare
Semantic Segmentation	Cityscapes test	HRNetV2 (train+val)	Mean IoU (class)	81.6%	# 38	Compare
Semantic Segmentation	Cityscapes val	HRNetV2 (HRNetV2-W48)	mIoU	81.1	# 44	Compare
Semantic Segmentation	Cityscapes val	HRNetV2 (HRNetV2-W40)	mIoU	80.2	# 54	Compare
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W18)	box AP	41.3	# 164	Compare
			AP50	59.2	# 97	Compare
			AP75	44.9	# 74	Compare
			APS	23.7	# 69	Compare
			APM	44.2	# 69	Compare
			APL	54.1	# 72	Compare
Object Detection	COCO minival	HTC (HRNetV2p-W18)	box AP	43.1	# 144	Compare
			APS	26.6	# 42	Compare
			APM	46.0	# 55	Compare
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W32)	box AP	42.3	# 153	Compare
			APS	25.0	# 59	Compare
			APM	45.4	# 62	Compare
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W48)	box AP	44.6	# 124	Compare
			AP50	62.7	# 69	Compare
			AP75	48.7	# 42	Compare
			APS	26.3	# 45	Compare
			APM	48.1	# 39	Compare
			APL	58.5	# 52	Compare
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W32)	box AP	43.7	# 136	Compare
			AP50	61.7	# 76	Compare
			AP75	47.7	# 49	Compare
			APS	25.6	# 52	Compare
			APM	46.5	# 52	Compare
			APL	57.4	# 59	Compare
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W48)	box AP	41.8	# 159	Compare
			AP50	62.8	# 67	Compare
			AP75	45.9	# 67	Compare
			APM	44.7	# 64	Compare
			APL	54.6	# 71	Compare
Object Detection	COCO minival	HTC (HRNetV2p-W32)	box AP	45.3	# 113	Compare
			APS	27.0	# 36	Compare
			APM	48.4	# 32	Compare
			APL	59.5	# 44	Compare
Instance Segmentation	COCO minival	HTC (HRNetV2p-W48)	mask AP	41.0	# 68	Compare
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W18)	box AP	39.2	# 189	Compare
			APM	41.7	# 83	Compare
			APL	51.0	# 85	Compare
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W32, cascade)	APS	26.1	# 46	Compare
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W32, cascade)	APM	47.9	# 42	Compare
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W48, cascade)	box AP	46.0	# 108	Compare
			APS	27.5	# 33	Compare
			APL	60.1	# 40	Compare
Object Detection	COCO minival	HTC (HRNetV2p-W48)	box AP	47.0	# 98	Compare
			APS	28.8	# 26	Compare
			APM	50.3	# 25	Compare
			APL	62.2	# 31	Compare
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W18)	box AP	38.0	# 198	Compare
			AP50	58.9	# 101	Compare
			AP75	41.5	# 94	Compare
			APS	22.6	# 79	Compare
			APM	40.8	# 86	Compare
			APL	49.6	# 90	Compare
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W32)	box AP	40.9	# 169	Compare
			AP50	61.8	# 75	Compare
			AP75	44.8	# 75	Compare
			APS	24.4	# 63	Compare
			APM	43.7	# 73	Compare
			APL	53.3	# 75	Compare
Object Detection	COCO test-dev	Mask R-CNN (HRNetV2p-W48 + cascade)	box mAP	46.1	# 127	Compare
			AP50	64.0	# 96	Compare
			AP75	50.3	# 80	Compare
			APS	27.1	# 77	Compare
			APM	48.6	# 74	Compare
			APL	58.3	# 69	Compare
			Hardware Burden	15G	# 1	Compare
			Operations per network pass	61.8G	# 1	Compare
Object Detection	COCO test-dev	Cascade R-CNN (HRNetV2p-W48)	AP75	48.6	# 86	Compare
			APS	26.0	# 89	Compare
			APM	47.3	# 85	Compare
			APL	56.3	# 89	Compare
Object Detection	COCO test-dev	Faster R-CNN (HRNetV2p-W48)	box mAP	42.4	# 172	Compare
			AP50	63.6	# 100	Compare
			AP75	46.4	# 113	Compare
			APS	24.9	# 97	Compare
			APM	44.6	# 115	Compare
			APL	53.0	# 120	Compare
			Hardware Burden	16G	# 1	Compare
			Operations per network pass	20.8G	# 1	Compare
Object Detection	COCO test-dev	FCOS (HRNetV2p-W48)	box mAP	40.5	# 190	Compare
			AP50	59.3	# 140	Compare
			APS	23.4	# 113	Compare
			APM	42.6	# 130	Compare
			APL	51.0	# 133	Compare
			Hardware Burden	16G	# 1	Compare
			Operations per network pass	27.3G	# 1	Compare
Object Detection	COCO test-dev	Mask R-CNN (HRNetV2p-W32 + cascade)	AP50	62.5	# 111	Compare
			AP75	48.6	# 86	Compare
			APL	56.3	# 89	Compare
			Hardware Burden	16G	# 1	Compare
			Operations per network pass	50.6G	# 1	Compare
Object Detection	COCO test-dev	CenterNet (HRNetV2-W48)	box mAP	43.5	# 154	Compare
			AP75	46.5	# 111	Compare
			APS	22.2	# 122	Compare
			APL	57.8	# 74	Compare
			Hardware Burden	16G	# 1	Compare
			Operations per network pass	21.7G	# 1	Compare
Object Detection	COCO test-dev	HTC (HRNetV2p-W48)	box mAP	47.3	# 117	Compare
			AP50	65.9	# 73	Compare
			AP75	51.2	# 67	Compare
			APS	28.0	# 69	Compare
			APM	49.7	# 65	Compare
			APL	59.8	# 63	Compare
			Hardware Burden	15G	# 1	Compare
			Operations per network pass	71.7G	# 1	Compare
Face Alignment	COFW-68	HRNetV2-W18	NME (inter-ocular)	5.06	# 6	Compare
Semantic Segmentation	DADA-seg	HRNet (ACDC)	mIoU	27.5	# 9	Compare
Dichotomous Image Segmentation	DIS-TE1	HRNet	max F-Measure	0.668	# 11	Compare
			weighted F-measure	0.579	# 10	Compare
			MAE	0.088	# 10	Compare
			S-Measure	0.742	# 10	Compare
			E-measure	0.797	# 11	Compare
			HCE	262	# 16	Compare
Dichotomous Image Segmentation	DIS-TE2	HRNet	max F-Measure	0.747	# 10	Compare
			weighted F-measure	0.664	# 10	Compare
			MAE	0.087	# 10	Compare
			S-Measure	0.784	# 10	Compare
			E-measure	0.840	# 7	Compare
			HCE	555	# 14	Compare
Dichotomous Image Segmentation	DIS-TE3	HRNet	max F-Measure	0.784	# 10	Compare
			weighted F-measure	0.700	# 10	Compare
			MAE	0.080	# 10	Compare
			S-Measure	0.805	# 9	Compare
			E-measure	0.869	# 8	Compare
			HCE	1049	# 13	Compare
Dichotomous Image Segmentation	DIS-TE4	HRNet	max F-Measure	0.772	# 10	Compare
			weighted F-measure	0.687	# 10	Compare
			MAE	0.092	# 10	Compare
			S-Measure	0.792	# 10	Compare
			E-measure	0.854	# 7	Compare
			HCE	3864	# 20	Compare
Dichotomous Image Segmentation	DIS-VD	HRNet	max F-Measure	0.726	# 12	Compare
			weighted F-measure	0.641	# 10	Compare
			MAE	0.095	# 11	Compare
			S-Measure	0.767	# 12	Compare
			E-measure	0.824	# 9	Compare
			HCE	1560	# 14	Compare
Thermal Image Segmentation	MFN Dataset	HRNet	mIOU	51.7	# 40	Compare
Semantic Segmentation	PASCAL Context	CFNet (ResNet-101)	mIoU	54.0	# 35	Compare
Semantic Segmentation	PASCAL Context	HRNetV2 HRNetV2-W48	mIoU	54	# 35	Compare
Semantic Segmentation	Potsdam	HRNet-18	mIoU	84.02	# 2	Compare
Semantic Segmentation	Potsdam	HRNet-48	mIoU	84.22	# 1	Compare
Semantic Segmentation	US3D	HRNet-48	mIoU	72.66	# 2	Compare
Semantic Segmentation	US3D	HRNet-18	mIoU	60.33	# 3	Compare
Semantic Segmentation	Vaihingen	HRNet-48	mIoU	76.75	# 8	Compare
Semantic Segmentation	Vaihingen	HRNet-18	mIoU	75.90	# 10	Compare

Results from Other Papers

Task	Dataset	Model	Metric Name	Metric Value	Rank	Compare
Face Alignment	300W	HRNet	NME_inter-ocular (%, Full)	3.32	# 24	See all
			NME_inter-ocular (%, Common)	2.87	# 21	See all
			NME_inter-ocular (%, Challenge)	5.15	# 24	See all
Face Alignment	COFW	HRNet	NME (inter-ocular)	3.45	# 12	See all
Face Alignment	WFLW	HRNet	NME (inter-ocular)	4.60	# 21	See all

Methods

Add Remove

1x1 Convolution • Average Pooling • Batch Normalization • Bottleneck Residual Block • Cascade Corner Pooling • Cascade R-CNN • CenterNet • Center Pooling • Convolution • DLA • Faster R-CNN • Global Average Pooling • HRNet • Kaiming Initialization • Mask R-CNN • Max Pooling • ReLU • Residual Block • Residual Connection • ResNet • RoIAlign • RoIPool • RPN • Softmax

Edit Social Preview

Deep High-Resolution Representation Learning for Visual Recognition

Code Edit Add Remove Mark official

Tasks Edit Add Remove

Datasets Edit

Results from the Paper Edit