On the Importance of Distractors for Few-Shot Classification

Das, Rajshekhar; Wang, Yu-Xiong; Moura, J.M.F.

doi:10.1109/iccv48922.2021.00890

Cited by 31 publications

(12 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Typically, CD-FSL with only source data is the most strict setting that demands model to recognize totally unseen target dataset without any target information. Flagship works including FWT [44], BSCD-FSL [18], LRP [40], ATA [48], wave-SAN [14], RDC [29], NSAE [32], and ConFT [10]. Though many well-designed techniques e.g.…”

Section: Related Workmentioning

confidence: 99%

“…Though many well-designed techniques e.g. readjusting the batch normalization [44], augmenting the difficult of meta tasks [48], spanning style distributions [14], and even fine-tuning models using few target images during the testing stage [10,18,29,32], the performances of them are still greatly limited due to the huge domain gap. By contrast, STARTUP [33] relaxes this strict setting and uses unlabeled target data for training.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

ME-D2N: Multi-Expert Domain Decompositional Network for Cross-Domain Few-Shot Learning

Xie

et al. 2022

Proceedings of the 30th ACM International Conference on Multimedia

View full text Add to dashboard Cite

Recently, Cross-Domain Few-Shot Learning (CD-FSL) which aims at addressing the Few-Shot Learning (FSL) problem across different domains has attracted rising attention. The core challenge of CD-FSL lies in the domain gap between the source and novel target datasets. Though many attempts have been made for CD-FSL without any target data during model training, the huge domain gap makes it still hard for existing CD-FSL methods to achieve very satisfactory results. Alternatively, learning CD-FSL models with few labeled target domain data which is more realistic and promising is advocated in previous work [13]. Thus, in this paper, we stick to this setting and technically contribute a novel Multi-Expert Domain Decompositional Network (ME-D2N). Concretely, to solve the data imbalance problem between the source data with sufficient examples and the auxiliary target data with limited examples, we build our model under the umbrella of multi-expert * indicates equal contributions, # indicates corresponding author.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

ME-D2N: Multi-Expert Domain Decompositional Network for Cross-Domain Few-Shot Learning

Xie

et al. 2022

Proceedings of the 30th ACM International Conference on Multimedia

View full text Add to dashboard Cite

show abstract

“…According to the way they construct and choose the positive samples, it can be roughly divided into supervised [53,[55][56][57] and self-supervised method [58,59]. Recently, there have been multiple methods, which utilise contrastive learning in few-shot image classification to help improve the performance [18,19,60,61]. While they generally use it to learn a good backbone on base set or task-specific representation, our method utilise it to obtain a triplet network, which projects all the features into a more discriminative space.…”

Section: Contrastive Learningmentioning

confidence: 99%

“…To reduce the overlapping, many methods [18, 19] introduce contrastive learning, which is proved effective to pull features from the same class together and push away that from different classes. However, these methods generally focus on boosting the performance of the backbone network, which helps little when extracting features of novel samples.…”

Section: Introductionmentioning

confidence: 99%

Redefining prior feature space via finetuning a triplet network for few‐shot learning

2022

IET Computer Vision

View full text Add to dashboard Cite

Few‐shot learning is to distinguish novel concepts with few annotated data, which has attracted much attention due to its requirement of limited training data for target classes. Recent few‐shot learning methods usually pretrain a feature extractor with images from the base set to boost the performance of few‐shot tasks and classify novel categories in this prior feature space. However, it is difficult for the pretrained feature extractor to extract accurate representations for novel categories, resulting in large amounts of overlapping areas between new classes. To address these issues, the prior feature space with a triplet network to learn a more discriminative space is refined, where features belonging to same class are pulled together and that from different classes are pushed apart. Specifically, the authors first follow recent paradigm of pretraining to obtain a prior feature space. Then, a triplet network with contrastive learning is trained to project the features from this space into a low‐dimensional one. The main difference lies in that the authors’ model is based on Maximum A Posteriori (MAP) and the triplet network with hallucinated features is finetuned from it to make them generalise well to novel categories. Finally, the authors conduct classification tasks in the finetuned space. The authors’ intuition is that the overlapping areas in novel categories can be separated by finetuning the triplet network pretrained on base set with contrastive learning. Experimental results on four few‐shot benchmarks show that it significantly outperforms the baseline methods, improves around 1.09% ∼ 13.09% than the best results in each dataset on both 1‐ and 5‐shot tasks.

show abstract

“…The basic idea is to mitigate the domain gap through data augmentation. Except for source domain, Das et al [6] and Liang et al [22] further fine-tune their models on unlabeled data in target domain via self-or semi-supervised methods, while paper [8] demonstrates the effectiveness of using very few labeled target data. Considering the acceptable cost of limited labeled data in practice, we advocate this direction.…”

Section: Introductionmentioning

confidence: 99%

TGDM: Target Guided Dynamic Mixup for Cross-Domain Few-Shot Learning

Zhuo

Chen

et al. 2022

Proceedings of the 30th ACM International Conference on Multimedia

View full text Add to dashboard Cite

Given sufficient training data on the source domain, cross-domain few-shot learning (CD-FSL) aims at recognizing new classes with a small number of labeled examples on the target domain. The key to addressing CD-FSL is to narrow the domain gap and transferring knowledge of a network trained on the source domain to the target domain. To help knowledge transfer, this paper introduces an intermediate domain generated by mixing images in the source and the target domain. Specifically, to generate the optimal intermediate domain for different target data, we propose a novel target guided dynamic mixup (TGDM) framework that leverages the target data to guide the generation of mixed images via dynamic mixup. The proposed TGDM framework contains a Mixup-3T network for learning classifiers and a dynamic ratio generation network (DRGN) for learning the optimal mix ratio. To better transfer the knowledge, the proposed Mixup-3T network contains three branches with shared parameters for classifying classes in the source domain, target domain, and intermediate domain. To generate the optimal intermediate domain, the DRGN learns to generate an optimal mix ratio according to the performance on auxiliary target data. Then, the whole TGDM framework is trained via bi-level meta-learning so that TGDM can rectify itself to achieve optimal performance on target data. Extensive experimental results on several benchmark datasets verify the effectiveness of our method.

show abstract

On the Importance of Distractors for Few-Shot Classification

Cited by 31 publications

References 23 publications

ME-D2N: Multi-Expert Domain Decompositional Network for Cross-Domain Few-Shot Learning

ME-D2N: Multi-Expert Domain Decompositional Network for Cross-Domain Few-Shot Learning

Redefining prior feature space via finetuning a triplet network for few‐shot learning

TGDM: Target Guided Dynamic Mixup for Cross-Domain Few-Shot Learning

Contact Info

Product

Resources

About