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 Plus annual

Presents the access of premium content as premium feature

Premium Content

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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

We present a novel method, SALAD, for the challenging vision task of adaptinga pre-trained "source" domain network to a "target" domain, with a small budgetfor annotation in the "target" domain and a shift in the label space. Further,the task assumes that the source data is not available for adaptation, due toprivacy concerns or otherwise. We postulate that such systems need to jointlyoptimize the dual task of (i) selecting fixed number of samples from the targetdomain for annotation and (ii) transfer of knowledge from the pre-trainednetwork to the target domain. To do this, SALAD consists of a novel GuidedAttention Transfer Network (GATN) and an active learning function, HAL. TheGATN enables feature distillation from pre-trained network to the targetnetwork, complemented with the target samples mined by HAL usingtransfer-ability and uncertainty criteria. SALAD has three key benefits: (i) itis task-agnostic, and can be applied across various visual tasks such asclassification, segmentation and detection; (ii) it can handle shifts in outputlabel space from the pre-trained source network to the target domain; (iii) itdoes not require access to source data for adaptation. We conduct extensiveexperiments across 3 visual tasks, viz. digits classification (MNIST, SVHN,VISDA), synthetic (GTA5) to real (CityScapes) image segmentation, and documentlayout detection (PubLayNet to DSSE). We show that our source-free approach,SALAD, results in an improvement of 0.5%-31.3%(across datasets and tasks) overprior adaptation methods that assume access to large amounts of annotatedsource data for adaptation.

SALAD: Source-free Active Label-Agnostic Domain Adaptation for Classification, Segmentation and Detection