TASK	DATASET	MODEL	METRIC NAME	METRIC VALUE	GLOBAL RANK
RGB Salient Object Detection	DAVIS-S	InSPyReNet	S-measure	0.962	# 7
RGB Salient Object Detection	DAVIS-S	InSPyReNet	F-measure	0.959	# 7
RGB Salient Object Detection	DAVIS-S	InSPyReNet	mBA	0.743	# 2
RGB Salient Object Detection	DAVIS-S	InSPyReNet	MAE	0.009	# 6
RGB Salient Object Detection	DAVIS-S	InSPyReNet (DUTS, HRSOD)	S-measure	0.972	# 5
RGB Salient Object Detection	DAVIS-S	InSPyReNet (DUTS, HRSOD)	F-measure	0.976	# 4
RGB Salient Object Detection	DAVIS-S	InSPyReNet (DUTS, HRSOD)	mBA	0.770	# 1
Dichotomous Image Segmentation	DIS-TE1	InSPyReNet	max F-Measure	0.834	# 5
Dichotomous Image Segmentation	DIS-TE1	InSPyReNet	S-Measure	0.862	# 5
Dichotomous Image Segmentation	DIS-TE1	InSPyReNet	HCE	148	# 4
Dichotomous Image Segmentation	DIS-TE1	InSPyReNet (HR scale)	max F-Measure	0.845	# 4
Dichotomous Image Segmentation	DIS-TE1	InSPyReNet (HR scale)	weighted F-measure	0.788	# 4
Dichotomous Image Segmentation	DIS-TE1	InSPyReNet (HR scale)	MAE	0.045	# 4
Dichotomous Image Segmentation	DIS-TE1	InSPyReNet (HR scale)	S-Measure	0.873	# 4
Dichotomous Image Segmentation	DIS-TE1	InSPyReNet (HR scale)	E-measure	0.894	# 4
Dichotomous Image Segmentation	DIS-TE1	InSPyReNet (HR scale)	HCE	110	# 3
Dichotomous Image Segmentation	DIS-TE2	InSPyReNet	max F-Measure	0.881	# 5
Dichotomous Image Segmentation	DIS-TE2	InSPyReNet	weighted F-measure	0.834	# 4
Dichotomous Image Segmentation	DIS-TE2	InSPyReNet	MAE	0.038	# 4
Dichotomous Image Segmentation	DIS-TE2	InSPyReNet	S-Measure	0.893	# 5
Dichotomous Image Segmentation	DIS-TE2	InSPyReNet	E-measure	0.925	# 4
Dichotomous Image Segmentation	DIS-TE2	InSPyReNet	HCE	316	# 4
Dichotomous Image Segmentation	DIS-TE2	InSPyReNet (HR scale)	max F-Measure	0.894	# 4
Dichotomous Image Segmentation	DIS-TE2	InSPyReNet (HR scale)	S-Measure	0.905	# 3
Dichotomous Image Segmentation	DIS-TE2	InSPyReNet (HR scale)	HCE	255	# 2
Dichotomous Image Segmentation	DIS-TE3	InSPyReNet (HR scale)	max F-Measure	0.919	# 4
Dichotomous Image Segmentation	DIS-TE3	InSPyReNet (HR scale)	weighted F-measure	0.871	# 4
Dichotomous Image Segmentation	DIS-TE3	InSPyReNet (HR scale)	MAE	0.034	# 4
Dichotomous Image Segmentation	DIS-TE3	InSPyReNet (HR scale)	S-Measure	0.918	# 3
Dichotomous Image Segmentation	DIS-TE3	InSPyReNet (HR scale)	E-measure	0.938	# 4
Dichotomous Image Segmentation	DIS-TE3	InSPyReNet (HR scale)	HCE	522	# 1
Dichotomous Image Segmentation	DIS-TE3	InSPyReNet	max F-Measure	0.904	# 5
Dichotomous Image Segmentation	DIS-TE3	InSPyReNet	weighted F-measure	0.856	# 5
Dichotomous Image Segmentation	DIS-TE3	InSPyReNet	MAE	0.038	# 5
Dichotomous Image Segmentation	DIS-TE3	InSPyReNet	S-Measure	0.902	# 5
Dichotomous Image Segmentation	DIS-TE3	InSPyReNet	E-measure	0.938	# 4
Dichotomous Image Segmentation	DIS-TE3	InSPyReNet	HCE	582	# 4
Dichotomous Image Segmentation	DIS-TE4	InSPyReNet (HR scale)	max F-Measure	0.905	# 3
Dichotomous Image Segmentation	DIS-TE4	InSPyReNet (HR scale)	weighted F-measure	0.848	# 4
Dichotomous Image Segmentation	DIS-TE4	InSPyReNet (HR scale)	MAE	0.042	# 4
Dichotomous Image Segmentation	DIS-TE4	InSPyReNet (HR scale)	S-Measure	0.905	# 2
Dichotomous Image Segmentation	DIS-TE4	InSPyReNet (HR scale)	E-measure	0.926	# 4
Dichotomous Image Segmentation	DIS-TE4	InSPyReNet (HR scale)	HCE	2336	# 3
Dichotomous Image Segmentation	DIS-TE4	InSPyReNet	max F-Measure	0.892	# 5
Dichotomous Image Segmentation	DIS-TE4	InSPyReNet	weighted F-measure	0.840	# 5
Dichotomous Image Segmentation	DIS-TE4	InSPyReNet	MAE	0.046	# 5
Dichotomous Image Segmentation	DIS-TE4	InSPyReNet	S-Measure	0.891	# 5
Dichotomous Image Segmentation	DIS-TE4	InSPyReNet	E-measure	0.926	# 4
Dichotomous Image Segmentation	DIS-TE4	InSPyReNet	HCE	2243	# 1
Dichotomous Image Segmentation	DIS-VD	InSPyReNet (HR scale)	max F-Measure	0.889	# 5
Dichotomous Image Segmentation	DIS-VD	InSPyReNet (HR scale)	S-Measure	0.900	# 5
Dichotomous Image Segmentation	DIS-VD	InSPyReNet (HR scale)	HCE	904	# 2
Dichotomous Image Segmentation	DIS-VD	InSPyReNet	max F-Measure	0.876	# 7
Dichotomous Image Segmentation	DIS-VD	InSPyReNet	weighted F-measure	0.826	# 6
Dichotomous Image Segmentation	DIS-VD	InSPyReNet	MAE	0.043	# 6
Dichotomous Image Segmentation	DIS-VD	InSPyReNet	S-Measure	0.887	# 7
Dichotomous Image Segmentation	DIS-VD	InSPyReNet	E-measure	0.921	# 6
Dichotomous Image Segmentation	DIS-VD	InSPyReNet	HCE	905	# 3
RGB Salient Object Detection	DUT-OMRON	InSPyReNet	MAE	0.045	# 6
RGB Salient Object Detection	DUT-OMRON	InSPyReNet	F-measure	0.832	# 5
RGB Salient Object Detection	DUT-OMRON	InSPyReNet	S-Measure	0.875	# 3
RGB Salient Object Detection	DUTS-TE	InSPyReNet	MAE	0.024	# 7
RGB Salient Object Detection	DUTS-TE	InSPyReNet	max F-measure	0.892	# 9
RGB Salient Object Detection	DUTS-TE	InSPyReNet	S-Measure	0.931	# 6
RGB Salient Object Detection	ECSSD	InSPyReNet	MAE	0.031	# 5
RGB Salient Object Detection	ECSSD	InSPyReNet	F-measure	0.96	# 2
RGB Salient Object Detection	ECSSD	InSPyReNet	S-Measure	0.936	# 2
RGB Salient Object Detection	HKU-IS	InSPyReNet	MAE	0.028	# 7
RGB Salient Object Detection	HKU-IS	InSPyReNet	F-measure	0.955	# 1
RGB Salient Object Detection	HKU-IS	InSPyReNet	S-Measure	0.944	# 1
RGB Salient Object Detection	HRSOD	InSPyReNet (DUTS, HRSOD)	S-Measure	0.960	# 4
RGB Salient Object Detection	HRSOD	InSPyReNet (DUTS, HRSOD)	max F-Measure	0.957	# 6
RGB Salient Object Detection	HRSOD	InSPyReNet (DUTS, HRSOD)	MAE	0.014	# 4
RGB Salient Object Detection	HRSOD	InSPyReNet (DUTS, HRSOD)	mBA	0.766	# 2
RGB Salient Object Detection	HRSOD	InSPyReNet (HRSOD, UHRSD)	S-Measure	0.956	# 7
RGB Salient Object Detection	HRSOD	InSPyReNet (HRSOD, UHRSD)	max F-Measure	0.956	# 7
RGB Salient Object Detection	HRSOD	InSPyReNet (HRSOD, UHRSD)	MAE	0.018	# 9
RGB Salient Object Detection	HRSOD	InSPyReNet (HRSOD, UHRSD)	mBA	0.771	# 1
RGB Salient Object Detection	HRSOD	InSPyReNet	S-Measure	0.952	# 9
RGB Salient Object Detection	HRSOD	InSPyReNet	max F-Measure	0.949	# 9
RGB Salient Object Detection	HRSOD	InSPyReNet	MAE	0.016	# 7
RGB Salient Object Detection	HRSOD	InSPyReNet	mBA	0.738	# 3
RGB Salient Object Detection	PASCAL-S	InSPyReNet	MAE	0.048	# 4
RGB Salient Object Detection	PASCAL-S	InSPyReNet	F-measure	0.893	# 2
RGB Salient Object Detection	PASCAL-S	InSPyReNet	S-Measure	0.893	# 2
RGB Salient Object Detection	UHRSD	InSPyReNet (DUTS, HRSOD)	S-Measure	0.936	# 7
RGB Salient Object Detection	UHRSD	InSPyReNet (DUTS, HRSOD)	mBA	0.785	# 2
RGB Salient Object Detection	UHRSD	InSPyReNet	S-Measure	0.932	# 9
RGB Salient Object Detection	UHRSD	InSPyReNet	max F-Measure	0.938	# 7
RGB Salient Object Detection	UHRSD	InSPyReNet	MAE	0.029	# 8
RGB Salient Object Detection	UHRSD	InSPyReNet	mBA	0.741	# 4
RGB Salient Object Detection	UHRSD	InSPyReNet (HRSOD, UHRSD)	S-Measure	0.953	# 2
RGB Salient Object Detection	UHRSD	InSPyReNet (HRSOD, UHRSD)	max F-Measure	0.957	# 5
RGB Salient Object Detection	UHRSD	InSPyReNet (HRSOD, UHRSD)	MAE	0.020	# 5
RGB Salient Object Detection	UHRSD	InSPyReNet (HRSOD, UHRSD)	mBA	0.812	# 1

Badge	Markdown
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/revisiting-image-pyramid-structure-for-high/salient-object-detection-on-hku-is)](https://paperswithcode.com/sota/salient-object-detection-on-hku-is?p=revisiting-image-pyramid-structure-for-high)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/revisiting-image-pyramid-structure-for-high/salient-object-detection-on-ecssd)](https://paperswithcode.com/sota/salient-object-detection-on-ecssd?p=revisiting-image-pyramid-structure-for-high)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/revisiting-image-pyramid-structure-for-high/salient-object-detection-on-pascal-s)](https://paperswithcode.com/sota/salient-object-detection-on-pascal-s?p=revisiting-image-pyramid-structure-for-high)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/revisiting-image-pyramid-structure-for-high/rgb-salient-object-detection-on-uhrsd)](https://paperswithcode.com/sota/rgb-salient-object-detection-on-uhrsd?p=revisiting-image-pyramid-structure-for-high)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/revisiting-image-pyramid-structure-for-high/dichotomous-image-segmentation-on-dis-te4)](https://paperswithcode.com/sota/dichotomous-image-segmentation-on-dis-te4?p=revisiting-image-pyramid-structure-for-high)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/revisiting-image-pyramid-structure-for-high/salient-object-detection-on-dut-omron)](https://paperswithcode.com/sota/salient-object-detection-on-dut-omron?p=revisiting-image-pyramid-structure-for-high)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/revisiting-image-pyramid-structure-for-high/dichotomous-image-segmentation-on-dis-te1)](https://paperswithcode.com/sota/dichotomous-image-segmentation-on-dis-te1?p=revisiting-image-pyramid-structure-for-high)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/revisiting-image-pyramid-structure-for-high/dichotomous-image-segmentation-on-dis-te2)](https://paperswithcode.com/sota/dichotomous-image-segmentation-on-dis-te2?p=revisiting-image-pyramid-structure-for-high)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/revisiting-image-pyramid-structure-for-high/dichotomous-image-segmentation-on-dis-te3)](https://paperswithcode.com/sota/dichotomous-image-segmentation-on-dis-te3?p=revisiting-image-pyramid-structure-for-high)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/revisiting-image-pyramid-structure-for-high/rgb-salient-object-detection-on-hrsod)](https://paperswithcode.com/sota/rgb-salient-object-detection-on-hrsod?p=revisiting-image-pyramid-structure-for-high)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/revisiting-image-pyramid-structure-for-high/rgb-salient-object-detection-on-davis-s)](https://paperswithcode.com/sota/rgb-salient-object-detection-on-davis-s?p=revisiting-image-pyramid-structure-for-high)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/revisiting-image-pyramid-structure-for-high/dichotomous-image-segmentation-on-dis-vd)](https://paperswithcode.com/sota/dichotomous-image-segmentation-on-dis-vd?p=revisiting-image-pyramid-structure-for-high)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/revisiting-image-pyramid-structure-for-high/salient-object-detection-on-duts-te)](https://paperswithcode.com/sota/salient-object-detection-on-duts-te?p=revisiting-image-pyramid-structure-for-high)`

Revisiting Image Pyramid Structure for High Resolution Salient Object Detection

20 Sep 2022 · Taehun Kim, Kunhee Kim, Joonyeong Lee, Dongmin Cha, Jiho Lee, Daijin Kim ·

Salient object detection (SOD) has been in the spotlight recently, yet has been studied less for high-resolution (HR) images. Unfortunately, HR images and their pixel-level annotations are certainly more labor-intensive and time-consuming compared to low-resolution (LR) images and annotations. Therefore, we propose an image pyramid-based SOD framework, Inverse Saliency Pyramid Reconstruction Network (InSPyReNet), for HR prediction without any of HR datasets. We design InSPyReNet to produce a strict image pyramid structure of saliency map, which enables to ensemble multiple results with pyramid-based image blending. For HR prediction, we design a pyramid blending method which synthesizes two different image pyramids from a pair of LR and HR scale from the same image to overcome effective receptive field (ERF) discrepancy. Our extensive evaluations on public LR and HR SOD benchmarks demonstrate that InSPyReNet surpasses the State-of-the-Art (SotA) methods on various SOD metrics and boundary accuracy.

PDF Abstract

Code

Add Remove Mark official

plemeri/transparent-background official

912

plemeri/inspyrenet official

↳ Quickstart in

Colab

Spaces

597

Tasks

Add Remove

Dichotomous Image Segmentation

Object Detection

RGB Salient Object Detection

Salient Object Detection

Vocal Bursts Intensity Prediction

Datasets

DUTS

PASCAL-S

HKU-IS

DUT-OMRON

DIS5K

ECSSD

HRSOD DAVIS-S

UHRSD

Results from the Paper

Edit

Ranked #1 on RGB Salient Object Detection on HKU-IS

Get a GitHub badge

Task	Dataset	Model	Metric Name	Metric Value	Global Rank	Benchmark
RGB Salient Object Detection	DAVIS-S	InSPyReNet	S-measure	0.962	# 7	Compare
			F-measure	0.959	# 7	Compare
			mBA	0.743	# 2	Compare
			MAE	0.009	# 6	Compare
RGB Salient Object Detection	DAVIS-S	InSPyReNet (DUTS, HRSOD)	S-measure	0.972	# 5	Compare
			F-measure	0.976	# 4	Compare
			mBA	0.770	# 1	Compare
Dichotomous Image Segmentation	DIS-TE1	InSPyReNet	max F-Measure	0.834	# 5	Compare
			S-Measure	0.862	# 5	Compare
			HCE	148	# 4	Compare
Dichotomous Image Segmentation	DIS-TE1	InSPyReNet (HR scale)	max F-Measure	0.845	# 4	Compare
			weighted F-measure	0.788	# 4	Compare
			MAE	0.045	# 4	Compare
			S-Measure	0.873	# 4	Compare
			E-measure	0.894	# 4	Compare
			HCE	110	# 3	Compare
Dichotomous Image Segmentation	DIS-TE2	InSPyReNet	max F-Measure	0.881	# 5	Compare
			weighted F-measure	0.834	# 4	Compare
			MAE	0.038	# 4	Compare
			S-Measure	0.893	# 5	Compare
			E-measure	0.925	# 4	Compare
			HCE	316	# 4	Compare
Dichotomous Image Segmentation	DIS-TE2	InSPyReNet (HR scale)	max F-Measure	0.894	# 4	Compare
			S-Measure	0.905	# 3	Compare
			HCE	255	# 2	Compare
Dichotomous Image Segmentation	DIS-TE3	InSPyReNet (HR scale)	max F-Measure	0.919	# 4	Compare
			weighted F-measure	0.871	# 4	Compare
			MAE	0.034	# 4	Compare
			S-Measure	0.918	# 3	Compare
			E-measure	0.938	# 4	Compare
			HCE	522	# 1	Compare
Dichotomous Image Segmentation	DIS-TE3	InSPyReNet	max F-Measure	0.904	# 5	Compare
			weighted F-measure	0.856	# 5	Compare
			MAE	0.038	# 5	Compare
			S-Measure	0.902	# 5	Compare
			E-measure	0.938	# 4	Compare
			HCE	582	# 4	Compare
Dichotomous Image Segmentation	DIS-TE4	InSPyReNet (HR scale)	max F-Measure	0.905	# 3	Compare
			weighted F-measure	0.848	# 4	Compare
			MAE	0.042	# 4	Compare
			S-Measure	0.905	# 2	Compare
			E-measure	0.926	# 4	Compare
			HCE	2336	# 3	Compare
Dichotomous Image Segmentation	DIS-TE4	InSPyReNet	max F-Measure	0.892	# 5	Compare
			weighted F-measure	0.840	# 5	Compare
			MAE	0.046	# 5	Compare
			S-Measure	0.891	# 5	Compare
			E-measure	0.926	# 4	Compare
			HCE	2243	# 1	Compare
Dichotomous Image Segmentation	DIS-VD	InSPyReNet (HR scale)	max F-Measure	0.889	# 5	Compare
			S-Measure	0.900	# 5	Compare
			HCE	904	# 2	Compare
Dichotomous Image Segmentation	DIS-VD	InSPyReNet	max F-Measure	0.876	# 7	Compare
			weighted F-measure	0.826	# 6	Compare
			MAE	0.043	# 6	Compare
			S-Measure	0.887	# 7	Compare
			E-measure	0.921	# 6	Compare
			HCE	905	# 3	Compare
RGB Salient Object Detection	DUT-OMRON	InSPyReNet	MAE	0.045	# 6	Compare
			F-measure	0.832	# 5	Compare
			S-Measure	0.875	# 3	Compare
RGB Salient Object Detection	DUTS-TE	InSPyReNet	MAE	0.024	# 7	Compare
			max F-measure	0.892	# 9	Compare
			S-Measure	0.931	# 6	Compare
RGB Salient Object Detection	ECSSD	InSPyReNet	MAE	0.031	# 5	Compare
			F-measure	0.96	# 2	Compare
			S-Measure	0.936	# 2	Compare
RGB Salient Object Detection	HKU-IS	InSPyReNet	MAE	0.028	# 7	Compare
			F-measure	0.955	# 1	Compare
			S-Measure	0.944	# 1	Compare
RGB Salient Object Detection	HRSOD	InSPyReNet (DUTS, HRSOD)	S-Measure	0.960	# 4	Compare
			max F-Measure	0.957	# 6	Compare
			MAE	0.014	# 4	Compare
			mBA	0.766	# 2	Compare
RGB Salient Object Detection	HRSOD	InSPyReNet (HRSOD, UHRSD)	S-Measure	0.956	# 7	Compare
			max F-Measure	0.956	# 7	Compare
			MAE	0.018	# 9	Compare
			mBA	0.771	# 1	Compare
RGB Salient Object Detection	HRSOD	InSPyReNet	S-Measure	0.952	# 9	Compare
			max F-Measure	0.949	# 9	Compare
			MAE	0.016	# 7	Compare
			mBA	0.738	# 3	Compare
RGB Salient Object Detection	PASCAL-S	InSPyReNet	MAE	0.048	# 4	Compare
			F-measure	0.893	# 2	Compare
			S-Measure	0.893	# 2	Compare
RGB Salient Object Detection	UHRSD	InSPyReNet (DUTS, HRSOD)	S-Measure	0.936	# 7	Compare
RGB Salient Object Detection	UHRSD	InSPyReNet (DUTS, HRSOD)	mBA	0.785	# 2	Compare
RGB Salient Object Detection	UHRSD	InSPyReNet	S-Measure	0.932	# 9	Compare
			max F-Measure	0.938	# 7	Compare
			MAE	0.029	# 8	Compare
			mBA	0.741	# 4	Compare
RGB Salient Object Detection	UHRSD	InSPyReNet (HRSOD, UHRSD)	S-Measure	0.953	# 2	Compare
			max F-Measure	0.957	# 5	Compare
			MAE	0.020	# 5	Compare
			mBA	0.812	# 1	Compare

Methods

Add Remove

Axial Attention • Dense Connections • Laplacian Pyramid • Layer Normalization • Linear Layer • Multi-Head Attention • Residual Connection • Scaled Dot-Product Attention • Softmax • Vision Transformer

Edit Social Preview

Revisiting Image Pyramid Structure for High Resolution Salient Object Detection

Code Edit Add Remove Mark official

Tasks Edit Add Remove

Datasets Edit

Results from the Paper Edit

Methods Edit Add Remove

Code

Add Remove Mark official

Tasks

Add Remove

Datasets

Results from the Paper

Edit

Methods

Add Remove