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 introduce a generic visual descriptor, termed as distribution awareretinal transform (DART), that encodes the structural context using log-polargrids for event cameras. The DART descriptor is applied to four differentproblems, namely object classification, tracking, detection and featurematching: (1) The DART features are directly employed as local descriptors in abag-of-features classification framework and testing is carried out on fourstandard event-based object datasets (N-MNIST, MNIST-DVS, CIFAR10-DVS,NCaltech-101). (2) Extending the classification system, tracking isdemonstrated using two key novelties: (i) For overcoming the low-sample problemfor the one-shot learning of a binary classifier, statistical bootstrapping isleveraged with online learning; (ii) To achieve tracker robustness, the scaleand rotation equivariance property of the DART descriptors is exploited for theone-shot learning. (3) To solve the long-term object tracking problem, anobject detector is designed using the principle of cluster majority voting. Thedetection scheme is then combined with the tracker to result in a highintersection-over-union score with augmented ground truth annotations on thepublicly available event camera dataset. (4) Finally, the event context encodedby DART greatly simplifies the feature correspondence problem, especially forspatio-temporal slices far apart in time, which has not been explicitly tackledin the event-based vision domain.

DART: Distribution Aware Retinal Transform for Event-based Cameras