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

Class-agnostic counting (CAC) aims to count all instances in a query imagegiven few exemplars. A standard pipeline is to extract visual features fromexemplars and match them with query images to infer object counts. Twoessential components in this pipeline are feature representation and similaritymetric. Existing methods either adopt a pretrained network to representfeatures or learn a new one, while applying a naive similarity metric withfixed inner product. We find this paradigm leads to noisy similarity matchingand hence harms counting performance. In this work, we propose asimilarity-aware CAC framework that jointly learns representation andsimilarity metric. We first instantiate our framework with a naive baselinecalled Bilinear Matching Network (BMNet), whose key component is a learnablebilinear similarity metric. To further embody the core of our framework, weextend BMNet to BMNet+ that models similarity from three aspects: 1)representing the instances via their self-similarity to enhance featurerobustness against intra-class variations; 2) comparing the similaritydynamically to focus on the key patterns of each exemplar; 3) learning from asupervision signal to impose explicit constraints on matching results.Extensive experiments on a recent CAC dataset FSC147 show that our modelssignificantly outperform state-of-the-art CAC approaches. In addition, we alsovalidate the cross-dataset generality of BMNet and BMNet+ on a car countingdataset CARPK. Code is at tiny.one/BMNet

Represent, Compare, and Learn: A Similarity-Aware Framework for Class-Agnostic Counting