TASK	DATASET	MODEL	METRIC NAME	METRIC VALUE	GLOBAL RANK
Semantic Segmentation	CamVid	HRNetV2 (HRNetV2-W48)	Mean IoU	78.47%	# 2
Semantic Segmentation	Cityscapes val	HRNetV2 (HRNetV2-W48)	mIoU	81.1%	# 3
Semantic Segmentation	Cityscapes val	HRNetV2 (HRNetV2-W40)	mIoU	80.2%	# 8
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W48)	box AP	44.6	# 26
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W48)	AP50	62.7	# 22
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W48)	AP75	48.7	# 16
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W48)	APS	26.3	# 17
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W48)	APM	48.1	# 12
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W48)	APL	58.5	# 15
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W32)	box AP	40.9	# 46
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W32)	AP50	61.8	# 26
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W32)	AP75	44.8	# 31
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W32)	APS	24.4	# 26
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W32)	APM	43.7	# 31
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W32)	APL	53.3	# 29
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W18)	box AP	38.0	# 65
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W18)	AP50	58.9	# 41
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W18)	AP75	41.5	# 45
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W18)	APS	22.6	# 33
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W18)	APM	40.8	# 40
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W18)	APL	49.6	# 42
Object Detection	COCO minival	HTC (HRNetV2p-W48)	box AP	47.0	# 15
Object Detection	COCO minival	HTC (HRNetV2p-W48)	APS	28.8	# 6
Object Detection	COCO minival	HTC (HRNetV2p-W48)	APM	50.3	# 6
Object Detection	COCO minival	HTC (HRNetV2p-W48)	APL	62.2	# 6
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W48, cascade)	box AP	46.0	# 19
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W48, cascade)	APS	27.5	# 11
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W48, cascade)	APM	48.9	# 8
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W48, cascade)	APL	60.1	# 9
Object Detection	COCO minival	HTC (HRNetV2p-W32)	box AP	45.3	# 22
Object Detection	COCO minival	HTC (HRNetV2p-W32)	APS	27	# 13
Object Detection	COCO minival	HTC (HRNetV2p-W32)	APM	48.4	# 10
Object Detection	COCO minival	HTC (HRNetV2p-W32)	APL	59.5	# 11
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W32, cascade)	box AP	44.5	# 27
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W32, cascade)	APS	26.1	# 18
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W32, cascade)	APM	47.9	# 13
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W32, cascade)	APL	58.5	# 15
Object Detection	COCO minival	HTC (HRNetV2p-W18)	box AP	43.1	# 34
Object Detection	COCO minival	HTC (HRNetV2p-W18)	APS	26.6	# 16
Object Detection	COCO minival	HTC (HRNetV2p-W18)	APM	46	# 21
Object Detection	COCO minival	HTC (HRNetV2p-W18)	APL	56.9	# 21
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W32)	box AP	42.3	# 38
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W32)	APS	25.0	# 24
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W32)	APM	45.4	# 23
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W32)	APL	54.9	# 26
Instance Segmentation	COCO minival	HTC (HRNetV2p-W48)	mask AP	41.0	# 12
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W18)	box AP	39.2	# 58
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W18)	APS	23.7	# 29
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W18)	APM	41.7	# 37
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W18)	APL	51.0	# 38
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W48)	box AP	41.8	# 40
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W48)	AP50	62.8	# 21
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W48)	AP75	45.9	# 26
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W48)	APS	25.0	# 24
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W48)	APM	44.7	# 24
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W48)	APL	54.6	# 27
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W18)	box AP	41.3	# 44
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W18)	AP50	59.2	# 40
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W18)	AP75	44.9	# 30
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W18)	APS	23.7	# 29
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W18)	APM	44.2	# 28
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W18)	APL	54.1	# 28
Object Detection	COCO test-dev	HTC (HRNetV2p-W48)	box AP	47.3	# 32
Object Detection	COCO test-dev	HTC (HRNetV2p-W48)	AP50	65.9	# 36
Object Detection	COCO test-dev	HTC (HRNetV2p-W48)	AP75	51.2	# 38
Object Detection	COCO test-dev	HTC (HRNetV2p-W48)	APS	28.0	# 41
Object Detection	COCO test-dev	HTC (HRNetV2p-W48)	APM	49.7	# 40
Object Detection	COCO test-dev	HTC (HRNetV2p-W48)	APL	59.8	# 33
Object Detection	COCO test-dev	Mask R-CNN (HRNetV2p-W32 + cascade)	box AP	44.7	# 49
Object Detection	COCO test-dev	Mask R-CNN (HRNetV2p-W32 + cascade)	AP50	62.5	# 59
Object Detection	COCO test-dev	Mask R-CNN (HRNetV2p-W32 + cascade)	AP75	48.6	# 52
Object Detection	COCO test-dev	Mask R-CNN (HRNetV2p-W32 + cascade)	APS	25.8	# 55
Object Detection	COCO test-dev	Mask R-CNN (HRNetV2p-W32 + cascade)	APM	47.1	# 56
Object Detection	COCO test-dev	Mask R-CNN (HRNetV2p-W32 + cascade)	APL	56.3	# 51
Object Detection	COCO test-dev	Cascade R-CNN (HRNetV2p-W48)	box AP	44.8	# 48
Object Detection	COCO test-dev	Cascade R-CNN (HRNetV2p-W48)	AP50	63.1	# 55
Object Detection	COCO test-dev	Cascade R-CNN (HRNetV2p-W48)	AP75	48.6	# 52
Object Detection	COCO test-dev	Cascade R-CNN (HRNetV2p-W48)	APS	26.0	# 53
Object Detection	COCO test-dev	Cascade R-CNN (HRNetV2p-W48)	APM	47.3	# 54
Object Detection	COCO test-dev	Cascade R-CNN (HRNetV2p-W48)	APL	56.3	# 51
Object Detection	COCO test-dev	CenterNet (HRNetV2-W48)	box AP	43.5	# 55
Object Detection	COCO test-dev	CenterNet (HRNetV2-W48)	AP50	62.1	# 62
Object Detection	COCO test-dev	CenterNet (HRNetV2-W48)	AP75	46.5	# 64
Object Detection	COCO test-dev	CenterNet (HRNetV2-W48)	APS	22.2	# 75
Object Detection	COCO test-dev	CenterNet (HRNetV2-W48)	APM	46.5	# 61
Object Detection	COCO test-dev	CenterNet (HRNetV2-W48)	APL	57.8	# 41
Object Detection	COCO test-dev	FCOS (HRNetV2p-W48)	box AP	40.5	# 73
Object Detection	COCO test-dev	FCOS (HRNetV2p-W48)	AP50	59.3	# 75
Object Detection	COCO test-dev	FCOS (HRNetV2p-W48)	AP75	43.3	# 83
Object Detection	COCO test-dev	FCOS (HRNetV2p-W48)	APS	23.4	# 69
Object Detection	COCO test-dev	FCOS (HRNetV2p-W48)	APM	42.6	# 78
Object Detection	COCO test-dev	FCOS (HRNetV2p-W48)	APL	51.0	# 77
Object Detection	COCO test-dev	Mask R-CNN (HRNetV2p-W48 + cascade)	box AP	46.1	# 40
Object Detection	COCO test-dev	Mask R-CNN (HRNetV2p-W48 + cascade)	AP50	64.0	# 49
Object Detection	COCO test-dev	Mask R-CNN (HRNetV2p-W48 + cascade)	AP75	50.3	# 45
Object Detection	COCO test-dev	Mask R-CNN (HRNetV2p-W48 + cascade)	APS	27.1	# 45
Object Detection	COCO test-dev	Mask R-CNN (HRNetV2p-W48 + cascade)	APM	48.6	# 46
Object Detection	COCO test-dev	Mask R-CNN (HRNetV2p-W48 + cascade)	APL	58.3	# 38
Object Detection	COCO test-dev	Faster R-CNN (HRNetV2p-W48)	box AP	42.4	# 62
Object Detection	COCO test-dev	Faster R-CNN (HRNetV2p-W48)	AP50	63.6	# 52
Object Detection	COCO test-dev	Faster R-CNN (HRNetV2p-W48)	AP75	46.4	# 65
Object Detection	COCO test-dev	Faster R-CNN (HRNetV2p-W48)	APS	24.9	# 60
Object Detection	COCO test-dev	Faster R-CNN (HRNetV2p-W48)	APM	44.6	# 70
Object Detection	COCO test-dev	Faster R-CNN (HRNetV2p-W48)	APL	53.0	# 67
Semantic Segmentation	PASCAL Context	CFNet (ResNet-101)	mIoU	54.0	# 10
Semantic Segmentation	PASCAL Context	HRNetV2 (HRNetV2-W48)	mIoU	54.0	# 10

Badge	Markdown
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/190807919/semantic-segmentation-on-camvid)](https://paperswithcode.com/sota/semantic-segmentation-on-camvid?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/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/instance-segmentation-on-coco-minival)](https://paperswithcode.com/sota/instance-segmentation-on-coco-minival?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/object-detection-on-coco)](https://paperswithcode.com/sota/object-detection-on-coco?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... (read more)

PDF Abstract

Code

Add Remove Mark official

open-mmlab/mmdetection

13,710

CSAILVision/semantic-segmentation-pytorch

↳ Quickstart in

Colab

3,735

leoxiaobin/deep-high-resolution-net.pytorch

3,016

HRNet/HRNet-Semantic-Segmentation

1,828

open-mmlab/mmsegmentation

↳ Quickstart in

Colab

1,381

See all 20 implementations

Tasks

Add Remove

INSTANCE SEGMENTATION

OBJECT DETECTION

POSE ESTIMATION

REPRESENTATION LEARNING

SEMANTIC SEGMENTATION

Datasets

COCO

Cityscapes

WIDER FACE

Helen AFW

300W

LFPW

AFLW

PASCAL Context

COFW

WFLW

Results from the Paper

Edit

Ranked #2 on Semantic Segmentation on CamVid

Get a GitHub badge

TASK	DATASET	MODEL	METRIC NAME	METRIC VALUE	GLOBAL RANK	BENCHMARK
Semantic Segmentation	CamVid	HRNetV2 (HRNetV2-W48)	Mean IoU	78.47%	# 2	Compare
Semantic Segmentation	Cityscapes val	HRNetV2 (HRNetV2-W48)	mIoU	81.1%	# 3	Compare
Semantic Segmentation	Cityscapes val	HRNetV2 (HRNetV2-W40)	mIoU	80.2%	# 8	Compare
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W48)	box AP	44.6	# 26	Compare
			AP50	62.7	# 22	Compare
			AP75	48.7	# 16	Compare
			APS	26.3	# 17	Compare
			APM	48.1	# 12	Compare
			APL	58.5	# 15	Compare
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W32)	box AP	40.9	# 46	Compare
			AP50	61.8	# 26	Compare
			AP75	44.8	# 31	Compare
			APS	24.4	# 26	Compare
			APM	43.7	# 31	Compare
			APL	53.3	# 29	Compare
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W18)	box AP	38.0	# 65	Compare
			AP50	58.9	# 41	Compare
			AP75	41.5	# 45	Compare
			APS	22.6	# 33	Compare
			APM	40.8	# 40	Compare
			APL	49.6	# 42	Compare
Object Detection	COCO minival	HTC (HRNetV2p-W48)	box AP	47.0	# 15	Compare
			APS	28.8	# 6	Compare
			APM	50.3	# 6	Compare
			APL	62.2	# 6	Compare
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W48, cascade)	box AP	46.0	# 19	Compare
			APS	27.5	# 11	Compare
			APM	48.9	# 8	Compare
			APL	60.1	# 9	Compare
Object Detection	COCO minival	HTC (HRNetV2p-W32)	box AP	45.3	# 22	Compare
			APS	27	# 13	Compare
			APM	48.4	# 10	Compare
			APL	59.5	# 11	Compare
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W32, cascade)	box AP	44.5	# 27	Compare
			APS	26.1	# 18	Compare
			APM	47.9	# 13	Compare
			APL	58.5	# 15	Compare
Object Detection	COCO minival	HTC (HRNetV2p-W18)	box AP	43.1	# 34	Compare
			APS	26.6	# 16	Compare
			APM	46	# 21	Compare
			APL	56.9	# 21	Compare
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W32)	box AP	42.3	# 38	Compare
			APS	25.0	# 24	Compare
			APM	45.4	# 23	Compare
			APL	54.9	# 26	Compare
Instance Segmentation	COCO minival	HTC (HRNetV2p-W48)	mask AP	41.0	# 12	Compare
Object Detection	COCO minival	Mask R-CNN (HRNetV2p-W18)	box AP	39.2	# 58	Compare
			APS	23.7	# 29	Compare
			APM	41.7	# 37	Compare
			APL	51.0	# 38	Compare
Object Detection	COCO minival	Faster R-CNN (HRNetV2p-W48)	box AP	41.8	# 40	Compare
			AP50	62.8	# 21	Compare
			AP75	45.9	# 26	Compare
			APS	25.0	# 24	Compare
			APM	44.7	# 24	Compare
			APL	54.6	# 27	Compare
Object Detection	COCO minival	Cascade R-CNN (HRNetV2p-W18)	box AP	41.3	# 44	Compare
			AP50	59.2	# 40	Compare
			AP75	44.9	# 30	Compare
			APS	23.7	# 29	Compare
			APM	44.2	# 28	Compare
			APL	54.1	# 28	Compare
Object Detection	COCO test-dev	HTC (HRNetV2p-W48)	box AP	47.3	# 32	Compare
			AP50	65.9	# 36	Compare
			AP75	51.2	# 38	Compare
			APS	28.0	# 41	Compare
			APM	49.7	# 40	Compare
			APL	59.8	# 33	Compare
Object Detection	COCO test-dev	Mask R-CNN (HRNetV2p-W32 + cascade)	box AP	44.7	# 49	Compare
			AP50	62.5	# 59	Compare
			AP75	48.6	# 52	Compare
			APS	25.8	# 55	Compare
			APM	47.1	# 56	Compare
			APL	56.3	# 51	Compare
Object Detection	COCO test-dev	Cascade R-CNN (HRNetV2p-W48)	box AP	44.8	# 48	Compare
			AP50	63.1	# 55	Compare
			AP75	48.6	# 52	Compare
			APS	26.0	# 53	Compare
			APM	47.3	# 54	Compare
			APL	56.3	# 51	Compare
Object Detection	COCO test-dev	CenterNet (HRNetV2-W48)	box AP	43.5	# 55	Compare
			AP50	62.1	# 62	Compare
			AP75	46.5	# 64	Compare
			APS	22.2	# 75	Compare
			APM	46.5	# 61	Compare
			APL	57.8	# 41	Compare
Object Detection	COCO test-dev	FCOS (HRNetV2p-W48)	box AP	40.5	# 73	Compare
			AP50	59.3	# 75	Compare
			AP75	43.3	# 83	Compare
			APS	23.4	# 69	Compare
			APM	42.6	# 78	Compare
			APL	51.0	# 77	Compare
Object Detection	COCO test-dev	Mask R-CNN (HRNetV2p-W48 + cascade)	box AP	46.1	# 40	Compare
			AP50	64.0	# 49	Compare
			AP75	50.3	# 45	Compare
			APS	27.1	# 45	Compare
			APM	48.6	# 46	Compare
			APL	58.3	# 38	Compare
Object Detection	COCO test-dev	Faster R-CNN (HRNetV2p-W48)	box AP	42.4	# 62	Compare
			AP50	63.6	# 52	Compare
			AP75	46.4	# 65	Compare
			APS	24.9	# 60	Compare
			APM	44.6	# 70	Compare
			APL	53.0	# 67	Compare
Semantic Segmentation	PASCAL Context	CFNet (ResNet-101)	mIoU	54.0	# 10	Compare
Semantic Segmentation	PASCAL Context	HRNetV2 (HRNetV2-W48)	mIoU	54.0	# 10	Compare

Methods used in the Paper

Add Remove

METHOD	TYPE
Average Pooling	Pooling Operations
Center Pooling	Pooling Operations
Cascade Corner Pooling	Pooling Operations
Cascade R-CNN	Object Detection Models
CenterNet	Object Detection Models
RPN	Region Proposal
RoIPool	RoI Feature Extractors
Faster R-CNN	Object Detection Models
Softmax	Output Functions
RoIAlign	RoI Feature Extractors
Mask R-CNN	Instance Segmentation Models
HRNet	Convolutional Neural Networks
Residual Connection	Skip Connections
ReLU	Activation Functions
1x1 Convolution	Convolutions
Batch Normalization	Normalization
Bottleneck Residual Block	Skip Connection Blocks
Global Average Pooling	Pooling Operations
Residual Block	Skip Connection Blocks
Kaiming Initialization	Initialization
Max Pooling	Pooling Operations
ResNet	Convolutional Neural Networks
Convolution	Convolutions

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

Methods used in the Paper Edit Add Remove

Code

Add Remove Mark official

Tasks

Add Remove

Datasets

Results from the Paper

Edit

Methods used in the Paper

Add Remove