Cross attention redistribution with contrastive learning for few shot object detection

Quan, Jianing; Ge, Baozhen; Chen, Lei

doi:10.1016/j.displa.2022.102162

Cited by 15 publications

(13 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(1) LSTD [58] proposes a regularization method based on transfer knowledge and background depression regularizations to enhance the fine-tuning effect; (2) Meta YOLO [20] learns feature representation by reweighting module to reassign feature weights; (3) MetaDet [59] solves the problem of few-shot classification and localization simultaneously through a weight prediction meta-model; (4) CME [21] balances the novel class margins by class margin loss and feature interference; (5) Meta R-CNN [25] obtains the class attention vector through the predictor-head remodeling network (PRN) module to remodel the ROI feature; (6) Viewpoint [26] performs efficient feature similarity calculation through feature subtraction; (7) DCNet [27] introduces adaptive context awareness into the feature aggregation module, to gain better global features and local features; (8) FSCN [60] introduces a novel few-shot classification refinement mechanism to improve the final classification; (9) FsDetView+ISAM+QSAM [61] generated an individual prototype for each sample to extract the unique characteristics of each support sample; (10) CAReD [62] maximizes the inter-class distance and minimizes the intra-class distance through contrastive learning.…”

Section: Baseline Methodsmentioning

confidence: 99%

“…Compared with FsDetView+ISAM+QSAM [61], BFR achieves 5.63% average improvements in mAP. Compared with CAReD [62], BFR achieves 2.11% average improvements in mAP.…”

Section: Voc Datasetmentioning

confidence: 99%

“…However, the proposed BFR approach still outperforms MetaDet [59], Meta R-CNN [25], and Viewpoint [26], which are based on Faster R-CNN. Meanwhile, BFR performs better than DCNet [27], FSCN [60], CAReD [62], etc. under most shots.…”

Section: Voc Datasetmentioning

confidence: 99%

See 2 more Smart Citations

Better Class Feature Representation for Few-Shot Object Detection: Feature Aggregation and Feature Space Redistribution

Zhang¹,

Xu²,

Chen³

et al. 2023

jist

View full text Add to dashboard Cite

Few-shot object detection is a valuable task and is yet to achieve great progress. Most research studies generate class prototypes by support samples to guide the detection. However, there still exists some challenges with this type of approach. On the one hand, the performance is highly sensitive to the quality of class prototype encoding. But limited training data impede the procession of class prototype learning. On the other hand, similar class prototypes distribute densely in the feature space, which easily leads to misjudgment. In this paper, we propose a better class feature representation (BFR) method to obtain more representative and more discriminative class prototypes with a small size of samples. First, BFR obtains the support set feature vector with richer semantic information via different feature aggregation methods. In addition, the distance metric function is used to identify the outliers and reduce the noisy interference with dynamic weights adjustment. Finally, BFR constrains the distribution of feature prototypes through distance loss, so that different prototypes are far away from each other in the feature space to reduce misjudgment. Extensive experiments on public benchmark datasets show that our method achieves superior performance with the main components bringing an 11.3% performance improvement.

show abstract

Section: Baseline Methodsmentioning

confidence: 99%

“…Compared with FsDetView+ISAM+QSAM [61], BFR achieves 5.63% average improvements in mAP. Compared with CAReD [62], BFR achieves 2.11% average improvements in mAP.…”

Section: Voc Datasetmentioning

confidence: 99%

See 1 more Smart Citation

Better Class Feature Representation for Few-Shot Object Detection: Feature Aggregation and Feature Space Redistribution

Zhang¹,

Xu²,

Chen³

et al. 2023

jist

View full text Add to dashboard Cite

show abstract

“…Quan et al [52] (CAReD) followed a similar approach. However, the weight w i is determined by the softmax over the correlation between the support features f S,c i and all other support features { f S,c j } K j=1 of the same category c. Due to the softmax, the weighting factors already sum up to 1 and the factor (1/k) is omitted.…”

Section: ) Aggregation Of Several Support Imagesmentioning

confidence: 99%

“…The outputs of all three matching modules are summed to give the final matching score. Many others [52], [66], [68], [69] adopt or build upon this multirelation detector. The additionally proposed two-way contrastive training strategy is implemented as follows.…”

Section: E Increase Discriminative Powermentioning

confidence: 99%

Few-Shot Object Detection: A Comprehensive Survey

Köhler

Eisenbach

Groß

2024

IEEE Trans. Neural Netw. Learning Syst.

View full text Add to dashboard Cite

Humans are able to learn to recognize new objects even from a few examples. In contrast, training deep-learningbased object detectors requires huge amounts of annotated data. To avoid the need to acquire and annotate these huge amounts of data, few-shot object detection (FSOD) aims to learn from few object instances of new categories in the target domain.In this survey, we provide an overview of the state of the art in FSOD. We categorize approaches according to their training scheme and architectural layout. For each type of approach, we describe the general realization as well as concepts to improve the performance on novel categories. Whenever appropriate, we give short takeaways regarding these concepts in order to highlight the best ideas. Eventually, we introduce commonly used datasets and their evaluation protocols and analyze the reported benchmark results. As a result, we emphasize common challenges in evaluation and identify the most promising current trends in this emerging field of FSOD.

show abstract

Boosting power line inspection in bad weather: Removing weather noise with channel-spatial attention-based UNet

Li,

Qian,

Duan

et al. 2023

Multimed Tools Appl

View full text Add to dashboard Cite

Cross attention redistribution with contrastive learning for few shot object detection

Cited by 15 publications

References 43 publications

Better Class Feature Representation for Few-Shot Object Detection: Feature Aggregation and Feature Space Redistribution

Better Class Feature Representation for Few-Shot Object Detection: Feature Aggregation and Feature Space Redistribution

Few-Shot Object Detection: A Comprehensive Survey

Boosting power line inspection in bad weather: Removing weather noise with channel-spatial attention-based UNet

Contact Info

Product

Resources

About

Cross attention redistribution with contrastive learning for few shot object detection

Cited by 15 publications

References 43 publications

Better Class Feature Representation for Few-Shot Object Detection: Feature Aggregation and Feature Space&#xA0;Redistribution

Better Class Feature Representation for Few-Shot Object Detection: Feature Aggregation and Feature Space&#xA0;Redistribution

Few-Shot Object Detection: A Comprehensive Survey

Boosting power line inspection in bad weather: Removing weather noise with channel-spatial attention-based UNet

Contact Info

Product

Resources

About

Better Class Feature Representation for Few-Shot Object Detection: Feature Aggregation and Feature Space Redistribution

Better Class Feature Representation for Few-Shot Object Detection: Feature Aggregation and Feature Space Redistribution