Multi-scale Adaptive Task Attention Network for Few-Shot Learning

Chen, Haoxing; Li, Huaxiong; Li, Yaohui; Chen, Chunlin

doi:10.48550/arxiv.2011.14479

Cited by 4 publications

(8 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The few-shot learning methods can be roughly classified into two categories: meta-learning based methods [26,7,31] and metric-learning based methods [30,32,20,29,3]. Metric-based few-shot learning methods have achieved remarkable success due to their fewer parameters and effectiveness.…”

Section: Input Imagementioning

confidence: 99%

“…However, due to the scarcity of data, it is not sufficient to measure the relation at the image-level [30,32]. Recently, CovaMNet [21], DN4 [20] and MATANet [3] introduce local representations (LRs) into few-shot learning and utilize these LRs to represent the image features, which can achieve better recognition results.…”

Section: Input Imagementioning

confidence: 99%

“…For example, Relation Nets [32] proposes a network to learn the most suitable image-level similarity metric functions. DN4 [3] proposes a cosine-based image-to-class metric to measure the similarity on pixel-level.…”

Section: Input Imagementioning

confidence: 99%

“…As proved by some recent studies [21,20,3], feature representation based on local descriptor is more abundant than image-level feature representation, which can alleviate the problem of sample scarcity. Therefore, we employ local descriptors to represent each image.…”

Section: Feature Embedding With Local Representationsmentioning

confidence: 99%

“…Humans can learn novel concepts and objects with just a few samples. Recently, many methods were proposed to learn new concepts with limited labeled data, such as semisupervised learning [40,35], zero-shot learning [38,39], and few-shot learning [26,7,30,32,20,29,3]. Facing with the problem of data scarcity, these three paradigms propose solutions from different perspectives.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Multi-level Metric Learning for Few-shot Image Recognition

Chen¹,

Li²,

Li³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Few-shot learning is devoted to training a model on few samples. Recently, the method based on local descriptor metric-learning has achieved great performance. Most of these approaches learn a model based on a pixel-level metric. However, such works can only measure the relations between them on a single level, which is not comprehensive and effective. We argue that if query images can simultaneously be well classified via three distinct level similarity metrics, the query images within a class can be more tightly distributed in a smaller feature space, generating more discriminative feature maps. Motivated by this, we propose a novel Multi-level Metric Learning (MML) method for fewshot learning, which not only calculates the pixel-level similarity but also considers the similarity of part-level features and the similarity of distributions. First, we use a feature extractor to get the feature maps of images. Second, a multi-level metric module is proposed to calculate the part-level, pixel-level, and distribution-level similarities simultaneously. Specifically, the distribution-level similarity metric calculates the distribution distance (i.e., Wasserstein distance, Kullback-Leibler divergence) between query images and the support set, the pixel-level, and the partlevel metric calculates the pixel-level and part-level similarities respectively. Finally, the fusion layer fuses three kinds of relation scores to obtain the final similarity score. Extensive experiments on popular benchmarks demonstrate that the MML method significantly outperforms the current state-of-the-art methods.

show abstract

Section: Input Imagementioning

confidence: 99%

Section: Input Imagementioning

confidence: 99%

Section: Input Imagementioning

confidence: 99%

Section: Feature Embedding With Local Representationsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Multi-level Metric Learning for Few-shot Image Recognition

Chen¹,

Li²,

Li³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

A small sample bearing fault diagnosis method based on ConvGRU relation network

Zhao,

Zhang

2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

Considering that in the fault diagnosis of bearing based on relation network, using the sample mean value as the class prototype for each class is susceptible to outliers, resulting in inaccurate class prototypes, this paper proposes a ConvGRU relation network fault diagnosis model; Firstly, the model utilizes a embedding module to extract bearing fault features, and then uses the Convolutional Gate Recurrent Unit (ConvGRU) as a learnable class prototype generator to generate class prototypes for each class. Secondly, a relation module is utilized to measure the similarity between class prototypes and the sample features of the query set, obtaining relation scores, and ultimately achieving fault diagnosis. In order to test the validity and advantages of the model, experimental verification and analysis were conducted on the Case Western Storage rolling bearing dataset. The results of the experiment show that the class prototypes generated by the ConvGRU class prototype generation module have better discrimination and accuracy compared to the class prototypes generated by the relation network. In the 10-way 5-shot experiment, the accuracy of the model proposed in this paper reaches 99.60%, which increases by 6.63%, 5.10%, 4.80% , and 1.75% compared to KNN, CNN, Prototype Network , and Relation Network. The method proposed in this paper helps to generate more accurate class prototypes and has a certain effect on improving the accuracy of model fault diagnosis.

show abstract

Hierarchical Representation based Query-Specific Prototypical Network for Few-Shot Image Classification

Li,

Chen

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Few-shot image classification aims at recognizing unseen categories with a small number of labeled training data. Recent metric-based frameworks tend to represent a support class by a fixed prototype (e.g., the mean of the support category) and make classification according to the similarities between query instances and support prototypes. However, discriminative dominant regions may locate uncertain areas of images and have various scales, which leads to the misaligned metric. Besides, a fixed prototype for one support category cannot fit for all query instances to accurately reflect their distances with this category, which lowers the efficiency of metric. Therefore, query-specific dominant regions in support samples should be extracted for a high-quality metric. To address these problems, we propose a Hierarchical Representation based Query-Specific Prototypical Network (QPN) to tackle the limitations by generating a region-level prototype for each query sample, which achieves both positional and dimensional semantic alignment simultaneously. Extensive experiments conducted on five benchmark datasets (including three fine-grained datasets) show that our proposed method outperforms the current state-of-the-art methods.

show abstract

Multi-scale Adaptive Task Attention Network for Few-Shot Learning

Cited by 4 publications

References 21 publications

Multi-level Metric Learning for Few-shot Image Recognition

Multi-level Metric Learning for Few-shot Image Recognition

A small sample bearing fault diagnosis method based on ConvGRU relation network

Hierarchical Representation based Query-Specific Prototypical Network for Few-Shot Image Classification

Contact Info

Product

Resources

About