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

Object detection in wide area motion imagery (WAMI) has drawn the attentionof the computer vision research community for a number of years. WAMI proposesa number of unique challenges including extremely small object sizes, bothsparse and densely-packed objects, and extremely large search spaces (largevideo frames). Nearly all state-of-the-art methods in WAMI object detectionreport that appearance-based classifiers fail in this challenging data andinstead rely almost entirely on motion information in the form of backgroundsubtraction or frame-differencing. In this work, we experimentally verify thefailure of appearance-based classifiers in WAMI, such as Faster R-CNN and aheatmap-based fully convolutional neural network (CNN), and propose a noveltwo-stage spatio-temporal CNN which effectively and efficiently combines bothappearance and motion information to significantly surpass the state-of-the-artin WAMI object detection. To reduce the large search space, the first stage(ClusterNet) takes in a set of extremely large video frames, combines themotion and appearance information within the convolutional architecture, andproposes regions of objects of interest (ROOBI). These ROOBI can contain fromone to clusters of several hundred objects due to the large video frame sizeand varying object density in WAMI. The second stage (FoveaNet) then estimatesthe centroid location of all objects in that given ROOBI simultaneously viaheatmap estimation. The proposed method exceeds state-of-the-art results on theWPAFB 2009 dataset by 5-16% for moving objects and nearly 50% for stoppedobjects, as well as being the first proposed method in wide area motion imageryto detect completely stationary objects.

ClusterNet: Detecting Small Objects in Large Scenes by Exploiting Spatio-Temporal Information