ABSTRACT
Recent deep learning methods rely on a large amount of labeled data to achieve high performance. These methods may be impractical in some scenarios, where manual data annotation is costly or the samples of certain categories are scarce (e.g., tumor lesions, endangered animals and rare individual activities). When only limited annotated samples are available, these methods usually suffer from the overfitting problem severely, which degrades the performance significantly. In contrast, humans can recognize the objects in the images rapidly and correctly with their prior knowledge after exposed to only a few annotated samples. To simulate the learning schema of humans and relieve the reliance on the large-scale annotation benchmarks, researchers start shifting towards the few-shot learning problem: they try to learn a model to correctly recognize novel categories with only a few annotated samples.
- Flood Sung, Yongxin Yang, Li Zhang, Tao Xiang, Philip H.S. Torr, Timothy M. Hospedales, Learning to Compare: Relation Network for Few-Shot Learning, CVPR 2018Google Scholar
- Yikai Wang, Li Zhang, Yuan Yao, Yanwei Fu, How to trust unlabeled data? Instance Credibility Inference for Few-Shot Learning, IEEE TPAMI 2021Google Scholar
- Chen Liu, Yanwei Fu, Chengming Xu, Siqian Yang, Jilin Li, Chengjie Wang, Li Zhang, Learning a Few-shot Embedding Model with Contrastive Learning, AAAI, 2021Google Scholar
- Yikai Wang, Chengming Xu, Chen Liu, Li Zhang, Yanwei Fu, Instance Credibility Inference for Few-Shot Learning, CVPR, 2020Google ScholarCross Ref
- Hongguang Zhang, Li Zhang, Xiaojuan Qi, Hongdong Li, Philip H.S. Torr, Piotr Koniusz, Few-shot Action Recognition with Permutation-invariant Attention, ECCV, 2020Google ScholarDigital Library
- Yuqian Fu, Li Zhang, Junke Wang, Yanwei Fu, Yu-Gang Jiang, Depth Guided Adaptive Meta-Fusion Network for Few-shot Video Recognition, ACM MM 2020 Google ScholarDigital Library
- S. Kevin Zhou, H. Greenspan, C. Davatzikos, J.S. Duncan, B. van Ginneken, A. Madabhushi, J.L. Prince, D. Rueckert, and R.M. Summers, ?A review of deep learning in medical imaging: Imaging traits, technology trends, case studies with progress highlights, and future promises," Proceedings of the IEEE, 2021Google Scholar
- Q. Yao, L. Xiao, P. Liu, and S. Kevin Zhou, ?Label-free segmentation of COVID-19 lesions in lung CT," IEEE Trans. on Medical Imaging, 2021.Google Scholar
- G. Shi, L. Xiao, Y. Chen, and S. Kevin Zhou, "Marginal loss and exclusion loss for partially supervised multi-organ segmentation," Medical Image Analysis, 2021.Google Scholar
- H. Li, H. Han, Z. Li, L. Wang, Z. Wu, J. Lu, and S. Kevin Zhou, "High-resolution chest X-ray bone suppression using unpaired CT structural priors," IEEE Trans. on Medical Imaging, 2020.Google Scholar
- C. Huang, H. Han, Q. Yao, S. Zhu, and S. Kevin Zhou, "3D U2Net: A 3D universal u-net for multi-domain medical image segmentation," MICCAI, 2019.Google Scholar
- B. Li, B. Yang, C. Liu, F. Liu, R. Ji, Q. Ye, "Beyond Max-Margin: Class Margin Equilibrium for Few-shot Object Detection," CVPR 2021.Google Scholar
- B. Liu, Y. Ding, J. Jiao, X. Ji, Q. Ye, Anti-aliasing Semantic Reconstruction for Few-Shot Semantic Segmentation, CVPR 2021.Google Scholar
- B. Yang, C Liu, B. Li, J. Jiao, Q. Ye, "Prototype Mixture Models for Few-shot Semantic Segmentation," ECCV 2020.Google Scholar
- B. Liu, J. Jiao, Q. Ye, "Harmonic Feature Activation for Few-Shot Semantic Segmentation," IEEE Trans. Image Process, 30(2):3142 - 3153, 2021Google ScholarCross Ref
- Xiaoyu Tao, Xiaopeng Hong, Xinyuan Chang, Songlin Dong, Xing Wei, Yihong Gong. Few-Shot Class-Incremental Learning. CVPR, 2020.Google ScholarCross Ref
- Songlin Dong, Xiaopeng Hong, Xiaoyu Tao, Xinyuan Chang, Xing Wei, Yinghong Gong. Few-Shot Class-Incremental Learning via Relation Knowledge Distillation. AAAI, 2021.Google Scholar
- Xiaoyu Tao, Xinyuan Chang, Xiaopeng Hong, Songlin Dong, Xing Wei, Yihong Gong. Topology-Preserving Class-Incremental Learning. ECCV, 2020.Google ScholarCross Ref
- Xiaoyu Tao, Xiaopeng Hong, Xinyuan Chang, Yihong Gong. Bi-objective Continual Learning: Learning "New' while Consolidating "Known,' AAAI, 2020.Google Scholar
- Chaofan Chen, Xiaoshan Yang, Changsheng Xu, Xuhui Huang, Zhe Ma: ECKPN: Explicit Class Knowledge Propagation Network for Transductive Few-shot Learning, CVPR 2021Google Scholar
- Yi Huang, Xiaoshan Yang, Junyu Gao, Jitao Sang, Changsheng Xu, Knowledge-driven Egocentric Multimodal Activity Recognition, TOMM, 2020 Google ScholarDigital Library
- Fan Qi, Xiaoshan Yang, Changsheng Xu, Emotion Knowledge Driven Video Highlight Detection, IEEE TMM, 2020Google Scholar
- Li Liu, Wanli Ouyang, Xiaogang Wang, Paul Fieguth, Jie Chen, Xinwang Liu, Matti Pietikäinen; Deep Learning for Generic Object Detection: A Survey, IJCV, 2019.Google Scholar
- Xiawu Zheng, Rongrong Ji_, Qiang Wang, Yuhang Chen, Baochang Zhang, Jie Chen, Qixiang Ye, Feiyue Huang, Yonghong Tian, MIGO-NAS: Towards Fast and Generalizable Neural Architecture Search, IEEE TPAMI, 2021.Google Scholar
- Hongliang He, Yao Ding, Guoli Song, Pengxu Wei, Jie Chen, DGN: Direction Guided Network for Nuclear Instance Segmentation, ICCV, 2021Google Scholar
- Mengjun Cheng, Zishang Kong, Jie Chen, Learnable Oriented-Derivative Network for Polyp Segmentation, MICCAI 2021Google Scholar
Index Terms
- Few-shot Learning for Multi-Modality Tasks
Recommendations
Few-shot partial multi-label learning with synthetic features network
AbstractIn partial multi-label learning (PML) problems, each training sample is partially annotated with a candidate label set, among which only a subset of labels are valid. The major hardship for PML is that its training procedure is prone to be misled ...
Few-shot partial multi-label learning via prototype rectification
AbstractPartial multi-label learning (PML) models the scenario where each training sample is annotated with a candidate label set, among which only a subset corresponds to the ground-truth labels. Existing PML approaches generally promise that there are ...
Few-Shot Classification with Multi-task Self-supervised Learning
Neural Information ProcessingAbstractFew-shot learning aims to mitigate the need for large-scale annotated data in the real world. The focus of few-shot learning is how to quickly adapt to unseen tasks, which heavily depends on outstanding feature extraction ability. Motivated by the ...
Comments