UNITE: Multitask Learning With Sufficient Feature for Dense Prediction

被引:1
作者
Tian, Yuxin [1 ,2 ]
Lin, Yijie [1 ,2 ]
Ye, Qing [1 ,2 ]
Wang, Jian [1 ,2 ]
Peng, Xi [1 ,2 ]
Lv, Jiancheng [1 ,2 ]
机构
[1] Sichuan Univ, Coll Comp Sci, Chengdu 610065, Peoples R China
[2] Minist Educ, Engn Res Ctr Machine Learning & Ind Intelligence, Chengdu 610065, Peoples R China
来源
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS | 2024年 / 54卷 / 08期
基金
中国国家自然科学基金;
关键词
Consistency and complementarity; dense prediction; multitask learning; sufficiency;
D O I
10.1109/TSMC.2024.3389672
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Existing multitask dense prediction methods typically rely on either global shared neural architecture or cross-task fusion strategy. However, these approaches tend to overlook either potential cross-task complementary or consistent information, resulting in suboptimal results. Motivated by this observation, we propose a novel plug-and-play module to concurrently leverage cross-task consistent and complementary information, thereby capturing a sufficient feature. Specifically, for a given pair of tasks, we compute a cross-task similarity matrix that extracts cross-task consistent features bidirectionally. To integrate the complementary signals from different tasks, we fuse the cross-task consistent features with the corresponding task-specific features using an 1x1 convolution. Extensive experimental results demonstrate the remarkable performance gain of our method on two challenging datasets w.r.t different task sets, compared with seven approaches. Under the two-task setting, our method has achieved 1.63% and 8.32% improvements on NYUD-v2 and PASCAL-Context, respectively. On the three-task setting, we obtain an additional 7.7% multitask performance gain.
引用
收藏
页码:5012 / 5024
页数:13
相关论文
共 64 条
[1]  
Achanta R, 2009, PROC CVPR IEEE, P1597, DOI 10.1109/CVPRW.2009.5206596
[2]  
Bahdanau D, 2016, Arxiv, DOI arXiv:1409.0473
[3]  
Bansal A., 2017, ARXIV
[4]   A model of inductive bias learning [J].
Baxter, J .
JOURNAL OF ARTIFICIAL INTELLIGENCE RESEARCH, 2000, 12 :149-198
[5]  
Bruggemann D., 2020, P BMVC, P1
[6]   Exploring Relational Context for Multi-Task Dense Prediction [J].
Bruggemann, David ;
Kanakis, Menelaos ;
Obukhov, Anton ;
Georgoulis, Stamatios ;
Van Gool, Luc .
2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2021), 2021, :15849-15858
[7]  
Carion N., 2020, EUR C COMP VIS, P213, DOI 10.1007/978-3-030-58452-813
[8]  
Caruana R., 1993, P 10 INT C MACH LEAR, P41
[9]   DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs [J].
Chen, Liang-Chieh ;
Papandreou, George ;
Kokkinos, Iasonas ;
Murphy, Kevin ;
Yuille, Alan L. .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2018, 40 (04) :834-848
[10]  
Chen Z, 2018, PR MACH LEARN RES, V80