English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Few-shot learning in image classification aims to learn a classifier toclassify images when only few training examples are available for each class.Recent work has achieved promising classification performance, where animage-level feature based measure is usually used. In this paper, we argue thata measure at such a level may not be effective enough in light of the scarcityof examples in few-shot learning. Instead, we think a local descriptor basedimage-to-class measure should be taken, inspired by its surprising success inthe heydays of local invariant features. Specifically, building upon the recentepisodic training mechanism, we propose a Deep Nearest Neighbor Neural Network(DN4 in short) and train it in an end-to-end manner. Its key difference fromthe literature is the replacement of the image-level feature based measure inthe final layer by a local descriptor based image-to-class measure. Thismeasure is conducted online via a $k$-nearest neighbor search over the deeplocal descriptors of convolutional feature maps. The proposed DN4 not onlylearns the optimal deep local descriptors for the image-to-class measure, butalso utilizes the higher efficiency of such a measure in the case of examplescarcity, thanks to the exchangeability of visual patterns across the images inthe same class. Our work leads to a simple, effective, and computationallyefficient framework for few-shot learning. Experimental study on benchmarkdatasets consistently shows its superiority over the related state-of-the-art,with the largest absolute improvement of $17\%$ over the next best. The sourcecode can be available from \UrlFont{https://github.com/WenbinLee/DN4.git}.

Revisiting Local Descriptor based Image-to-Class Measure for Few-shot Learning