Sample-Centric Feature Generation for Semi-Supervised Few-Shot Learning

Zhang, Bo; Ye, Hancheng; Yu, Gang; Wang, Bin; Wu, Yike; Fan, Jianchao; Chen, Tao

doi:10.1109/tip.2022.3154938

Cited by 21 publications

(15 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We found ViT performed the best accuracy for both 1-shot and 5-shot learning tasks. Note that other one-shot learning methods in previous study can obtain the same degree of accuracy without using such an excellent backbone DNN like ViT (e.g., 82% [40] and 87% [41] with ResNet-12), and thus DONE is not the best method for accuracy. However, these results suggest that DONE with ViT is already at a level of practical uses.…”

Section: Evaluation Of Dnns In Donementioning

confidence: 86%

“…This approach includes various types such as semi-supervised approaches and example generation using Generative Adversarial Networks [10]. Meta learning approaches train abilities of learning systems to learn [17,22,27,31,36,40]. The purpose of meta learning is to aim to increase the learning efficiency itself, and this is a powerful approach for learning from a small amount of training data, typically one-shot learning task [38].…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

It's DONE: Direct ONE-shot learning without training optimization

Hosoda¹,

Nishida²,

Seno³

et al. 2022

Preprint

View full text Add to dashboard Cite

Learning a new concept from one example is a superior function of human brain and it is drawing attention in the field of machine learning as one-shot learning task. In this paper, we propose the simplest method for this task, named Direct ONE-shot learning (DONE). DONE adds a new class to a pretrained deep neural network (DNN) classifier with neither training optimization nor other-classes modification. DONE is inspired by Hebbian theory and directly uses the neural activity input of the final dense layer obtained from a data that belongs to the new additional class as the connectivity weight (synaptic strength) with a newly-provided-output neuron for the new class. DONE requires just one inference for obtaining the output of the final dense layer and its procedure is simple, deterministic, not requiring parameter tuning and hyperparameters. The performance of DONE depends entirely on the pretrained DNN model used as a backbone model, and we confirmed that DONE with a well-trained backbone model performs a practical-level accuracy. DONE has some advantages including a DNN's practical use that is difficult to spend high cost for a training, an evaluation of existing DNN models, and the understanding of the brain. DONE might be telling us one-shot learning is an easy task that can be achieved by a simple principle not only for humans but also for current well-trained DNN models.

show abstract

Section: Evaluation Of Dnns In Donementioning

confidence: 86%

Section: Related Workmentioning

confidence: 99%

It's DONE: Direct ONE-shot learning without training optimization

Hosoda¹,

Nishida²,

Seno³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…As the study progresses, FSOD [27]- [34] gradually achieves an increasing detection precision. The constantly emerging FSOD methods employ various techniques, but they can be roughly divided into three main categories: transfer-learningbased [27]- [28], meta-learning-based [29]- [30] and metriclearning-based [31]- [32].…”

Section: Introductionmentioning

confidence: 92%

Few-Shot Object Detection With Self-Adaptive Global Similarity and Two-Way Foreground Stimulator in Remote Sensing Images

Zhang

Wang

2022

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

Self Cite

View full text Add to dashboard Cite

Few-shot object detection aims to localize and recognize potential objects of interest only by using a few annotated data, and it is beneficial for remote sensing images (RSIs) based applications such as urban monitoring. Previous RSIs-based few-shot object detection works often try to convert the support images from class-agnostic features to class-specific vectors, and then perform feature attention operations on query image features to be tested. However, such methods still face two critical challenges: 1) They ignore the spatial similarity of support-query features, which is indispensable for RSIs detection; 2) They perform the feature attention operation in a unidirectional manner, which means that the learned supportquery relations are asymmetric. In this paper, to address the challenges above, we design a few-shot object detector, which can quickly and accurately generalize to unseen categories with only a small amount of data. The proposed approach contains two components: 1) the self-adaptive global similarity module that preserves the internal context information to calculate the similarity map between the objects in support and query images, and 2) the two-way foreground stimulator module that can apply the similarity map to the detailed embeddings of support and query images at the same time to make full use of support information, further strengthening the foreground objects and weakening the unconcerned samples. Experiments are conducted on DIOR and NWPU VHR-10 datasets and their results demonstrate the superiority of the proposed method compared with several state-of-the-art methods.

show abstract

“…Few-Shot Fine-Grained Learning (FSFGL). Recent FSL works [1,23,25,37,63,65,68] can be roughly categorized into three types: 1) Optimization-based methods [11,38] that focus on learning good initialization parameters in order to quickly adapt the few-shot model to novel classes; 2) Metric-based methods [35,41,43,48] that aim to design a distance metric, so that the few-shot model can learn the semantic relation between different input images; 3) Data augmentation-based methods [6,14,26,36,54,70] that produce new samples to enlarge the training set for model training. Recently, inspired by the rapid development of meta-learning, researchers [9, 12, 19, 27-29, 46, 50, 57, 58, 71] start to explore the generalization ability of FSL model on novel fine-grained subclasses where only a few training examples are given.…”

Section: Related Workmentioning

confidence: 99%

Learning Cross-Image Object Semantic Relation in Transformer for Few-Shot Fine-Grained Image Classification

Zhang¹,

Yuan²,

Li³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Few-shot fine-grained learning aims to classify a query image into one of a set of support categories with fine-grained differences. Although learning different objects' local differences via Deep Neural Networks has achieved success, how to exploit the query-support cross-image object semantic relations in Transformer-based architecture remains under-explored in the few-shot fine-grained scenario. In this work, we propose a Transformer-based doublehelix model, namely HelixFormer, to achieve the cross-image object semantic relation mining in a bidirectional and symmetrical manner. The HelixFormer consists of two steps: 1) Relation Mining Process (RMP) across different branches, and 2) Representation Enhancement Process (REP) within each individual branch. By the designed RMP, each branch can extract fine-grained object-level Cross-image Semantic Relation Maps (CSRMs) using information from the other branch, ensuring better cross-image interaction in semantically related local object regions. Further, with the aid of CSRMs, the developed REP can strengthen the extracted features for those discovered semantically-related local regions in each branch, boosting the model's ability to distinguish subtle feature differences of fine-grained objects. Extensive experiments conducted on five public fine-grained benchmarks demonstrate that HelixFormer can effectively enhance the cross-image object semantic relation matching for recognizing fine-grained objects, achieving much better performance over most state-of-the-art methods under 1-shot and 5-shot scenarios. Our code is available at: https:// github.com/ JiakangYuan/ HelixFormer. CCS CONCEPTS• Computing methodologies → Visual content-based indexing and retrieval; Matching.

show abstract

Sample-Centric Feature Generation for Semi-Supervised Few-Shot Learning

Cited by 21 publications

References 45 publications

It's DONE: Direct ONE-shot learning without training optimization

It's DONE: Direct ONE-shot learning without training optimization

Few-Shot Object Detection With Self-Adaptive Global Similarity and Two-Way Foreground Stimulator in Remote Sensing Images

Learning Cross-Image Object Semantic Relation in Transformer for Few-Shot Fine-Grained Image Classification

Contact Info

Product

Resources

About