When Does Self-supervision Improve Few-Shot Learning?

Su, Jen-Liang; Maji, Subhransu; Hariharan, Bharath

doi:10.1007/978-3-030-58571-6_38

Cited by 108 publications

(74 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…nearest neighbor [88]. Overall, the community seems to be reaching a consensus [15], [34], [41], [73], [90], [95]: the key ingredient to high-performing few-shot classification is learning a general representation, rather than sophisticated algorithms for adapting to the new classes. In line with these works, we study what representation is suitable for solving a target task.…”

Section: Related Workmentioning

confidence: 99%

Factors of Influence for Transfer Learning Across Diverse Appearance Domains and Task Types

Mensink¹,

Uijlings

Kuznetsova

et al. 2022

IEEE Trans. Pattern Anal. Mach. Intell.

View full text Add to dashboard Cite

Section: Related Workmentioning

confidence: 99%

Factors of Influence for Transfer Learning Across Diverse Appearance Domains and Task Types

Mensink¹,

Uijlings

Kuznetsova

et al. 2022

IEEE Trans. Pattern Anal. Mach. Intell.

View full text Add to dashboard Cite

“…Most of the prior works [ 61 , 62 ] in computer vision weave self-supervision into few-shot learning by adding pretext tasks loss. Predicting the index of jigsaw puzzles and the angle of rotations are among the most effective pretext task choices.…”

Section: Methodsmentioning

confidence: 99%

RS-SSKD: Self-Supervision Equipped with Knowledge Distillation for Few-Shot Remote Sensing Scene Classification

Zhang

Wang

et al. 2021

Sensors

View full text Add to dashboard Cite

While growing instruments generate more and more airborne or satellite images, the bottleneck in remote sensing (RS) scene classification has shifted from data limits toward a lack of ground truth samples. There are still many challenges when we are facing unknown environments, especially those with insufficient training data. Few-shot classification offers a different picture under the umbrella of meta-learning: digging rich knowledge from a few data are possible. In this work, we propose a method named RS-SSKD for few-shot RS scene classification from a perspective of generating powerful representation for the downstream meta-learner. Firstly, we propose a novel two-branch network that takes three pairs of original-transformed images as inputs and incorporates Class Activation Maps (CAMs) to drive the network mining, the most relevant category-specific region. This strategy ensures that the network generates discriminative embeddings. Secondly, we set a round of self-knowledge distillation to prevent overfitting and boost the performance. Our experiments show that the proposed method surpasses current state-of-the-art approaches on two challenging RS scene datasets: NWPU-RESISC45 and RSD46-WHU. Finally, we conduct various ablation experiments to investigate the effect of each component of the proposed method and analyze the training time of state-of-the-art methods and ours.

show abstract

“…They presented generative algorithms to generate new training examples, and proposed to learn better embeddings from more training examples with larger intra-class diversity. In [56], Su et al studied the efectiveness of utilizing self-supervised learning (SSL) techniques in few-shot setting. SSL utilizes the structure information contained already in an image to facilitate the learning of representations, having been well studied in traditional unsupervised learning with large unlabeled datasets [9,13,27].…”

Section: Towards Improving Feature Embeddingsmentioning

confidence: 99%

Revisiting Local Descriptor for Improved Few-Shot Classification

Hong

Liu

et al. 2022

ACM Trans. Multimedia Comput. Commun. Appl.

View full text Add to dashboard Cite

Few-shot classification studies the problem of quickly adapting a deep learner to understanding novel classes based on few support images. In this context, recent research efforts have been aimed at designing more and more complex classifiers that measure similarities between query and support images, but left the importance of feature embeddings seldom explored. We show that the reliance on sophisticated classifiers is not necessary, and a simple classifier applied directly to improved feature embeddings can instead outperform most of the leading methods in the literature. To this end, we present a new method named DCAP for few-shot classification, in which we investigate how one can improve the quality of embeddings by leveraging D ense C lassification and A ttentive P ooling. Specifically, we propose to train a learner on base classes with abundant samples to solve dense classification problem first and then meta-train the learner on a bunch of randomly sampled few-shot tasks to adapt it to few-shot scenario or the test time scenario. During meta-training, we suggest to pool feature maps by applying attentive pooling instead of the widely used global average pooling (GAP) to prepare embeddings for few-shot classification. Attentive pooling learns to reweight local descriptors, explaining what the learner is looking for as evidence for decision making. Experiments on two benchmark datasets show the proposed method to be superior in multiple few-shot settings while being simpler and more explainable. Code would be made public at: https://github.com/Ukeyboard/dcap/.

show abstract

When Does Self-supervision Improve Few-Shot Learning?

Cited by 108 publications

References 38 publications

Factors of Influence for Transfer Learning Across Diverse Appearance Domains and Task Types

Factors of Influence for Transfer Learning Across Diverse Appearance Domains and Task Types

RS-SSKD: Self-Supervision Equipped with Knowledge Distillation for Few-Shot Remote Sensing Scene Classification

Revisiting Local Descriptor for Improved Few-Shot Classification

Contact Info

Product

Resources

About