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

Object recognition systems are usually trained and evaluated on highresolution images. However, in real world applications, it is common that theimages have low resolutions or have small sizes. In this study, we first trackthe performance of the state-of-the-art deep object recognition network,Faster- RCNN, as a function of image resolution. The results reveals negativeeffects of low resolution images on recognition performance. They also showthat different spatial frequencies convey different information about theobjects in recognition process. It means multi-resolution recognition systemcan provides better insight into optimal selection of features that results inbetter recognition of objects. This is similar to the mechanisms of the humanvisual systems that are able to implement multi-scale representation of avisual scene simultaneously. Then, we propose a multi-resolution objectrecognition framework rather than a single-resolution network. The proposedframework is evaluated on the PASCAL VOC2007 database. The experimental resultsshow the performance of our adapted multi-resolution Faster-RCNN frameworkoutperforms the single-resolution Faster-RCNN on input images with variousresolutions with an increase in the mean Average Precision (mAP) of 9.14%across all resolutions and 1.2% on the full-spectrum images. Furthermore, theproposed model yields robustness of the performance over a wide range ofspatial frequencies.

Improving Performance of Object Detection using the Mechanisms of Visual Recognition in Humans