Group-wise co-salient object detection via multi-view self-labeling novel class discovery

doi:10.1007/s11704-023-3284-5

Front. Comput. Sci.

2024, Vol. 18

Issue (2) : 182709 https://doi.org/10.1007/s11704-023-3284-5

Image and Graphics

Group-wise co-salient object detection via multi-view self-labeling novel class discovery

Yang WU^1,², Gang DONG¹, Lingyan LIANG¹, Yaqian ZHAO¹, Kaihua ZHANG^1,²(

)

¹. Inspur Electronic Information Industry Co., Ltd., Beijing 100080, China
². Jiangsu Key Laboratory of Big Data Analysis Technology and Jiangsu Collaborative Innovation Center of Atmospheric Environment and Equipment Technology, Nanjing University of Information Science and Technology, Nanjing 210044, China

Download: PDF(1545 KB) HTML
Export: BibTeX | EndNote | Reference Manager | ProCite | RefWorks

Corresponding Author(s): Kaihua ZHANG

Just Accepted Date: 19 December 2023 Issue Date: 18 January 2024

Cite this article:

Yang WU,Gang DONG,Lingyan LIANG, et al. Group-wise co-salient object detection via multi-view self-labeling novel class discovery[J]. Front. Comput. Sci., 2024, 18(2): 182709.

URL:

https://academic.hep.com.cn/fcs/EN/10.1007/s11704-023-3284-5
https://academic.hep.com.cn/fcs/EN/Y2024/V18/I2/182709

Fig.1 The pipeline of our method. ‘AP’, ‘L’ and ‘S’ stand for average pooling, linear layer and Softmax layer, respectively

Methods	CoSal2015				CoSOD3k				CoCA
Methods	$M A E ↓$	$S α ↑$	$E ? m a x ↑$	$F β m a x ↑$	$M A E ↓$	$S α ↑$	$E ? m a x ↑$	$F β m a x ↑$	$M A E ↓$	$S α ↑$	$E ? m a x ↑$	$F β m a x ↑$
CoEGNet(TPAMI2021)	0.077	0.836	0.882	0.832	0.092	0.762	0.825	0.736	0.106	0.612	0.717	0.493
GCoNet(CVPR2021)	0.069	0.845	0.887	0.847	0.071	0.802	0.860	0.750	0.105	0.673	0.760	0.524
CADC(ICCV2021)	0.064	0.866	0.906	0.862	0.096	0.801	0.840	0.759	0.132	0.681	0.744	0.548
HrSSMN(TMM2022)	0.062	0.845	0.895	0.841	0.087	0.788	0.842	0.753	0.106	0.671	0.739	0.532
DCFM(CVPR2022)	0.067	0.838	0.892	0.856	0.067	0.810	0.874	0.805	0.085	0.710	0.783	0.598
Ours	0.057	0.862	0.912	0.865	0.063	0.825	0.883	0.809	0.097	0.719	0.795	0.605

Tab.1 Performance comparisons of our model with other state-of-the-arts. Bold indicates the best and underlined indicates the second-best performance

1	Wu Y, Song H, Liu B, Zhang K, Liu D. Co-salient object detection with uncertainty-aware group exchange-masking. In: Proceedings of 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2023, 19639−19648
2	D P, Fan T, Li Z, Lin G P, Ji D, Zhang M M, Cheng H, Fu J Shen . Re-thinking co-salient object detection. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 44( 8): 4339–4354
3	Y, Liu T, Li Y, Wu H, Song K Zhang . Self-supervised image co-saliency detection. Computers and Electrical Engineering, 2023, 105: 108533
4	Wu Y, Liang L, Zhao Y, Zhang K. Object-aware calibrated depth-guided transformer for RGB-D co-salient object detection. In: Proceedings of 2023 IEEE International Conference on Multimedia and Expo. 2023, 1121−1126
5	Wu Y, Zhang H, Liang L, Zhao Y, Zhang K. Group-wise co-salient object detection with Siamese transformers via Brownian distance covariance matching. In: Proceedings of 2023 IEEE International Conference on Acoustics, Speech and Signal Processing. 2023, 1−5
6	Fan Q, Fan D P, Fu H, Tang C K, Shao L, Tai Y W. Group collaborative learning for co-salient object detection. In: Proceedings of 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2021, 12283−12293
7	Fini E, Sangineto E, Lathuilière S, Zhong Z, Nabi M, Ricci E. A unified objective for novel class discovery. In: Proceedings of 2021 IEEE/CVF International Conference on Computer Vision. 2021, 9264−9272
8	Zhang N, Han J, Liu N, Shao L. Summarize and search: learning consensus-aware dynamic convolution for co-saliency detection. In: Proceedings of 2021 IEEE/CVF International Conference on Computer Vision. 2021, 4147−4156
9	K, Zhang Y, Wu M, Dong B, Liu D, Liu Q Liu . Deep object co-segmentation and co-saliency detection via high-order spatial-semantic network modulation. IEEE Transactions on Multimedia, 2023, 25: 5733–5746
10	Yu S, Xiao J, Zhang B, Lim E G. Democracy does matter: comprehensive feature mining for co-salient object detection. In: Proceedings of 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2022, 969−978

[1]

FCS-23284-OF-YW_suppl_1

Download

Viewed

Full text

Abstract

Cited

Shared

Discussed