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

Image space feature detection is the act of selecting points or parts of animage that are easy to distinguish from the surrounding image region. Bycombining a repeatable point detection with a descriptor, parts of an image canbe matched with one another, which is useful in applications like estimatingpose from camera input or rectifying images. Recently, precise indoor trackinghas started to become important for Augmented and Virtual reality as it isnecessary to allow positioning of a headset in 3D space without the need forexternal tracking devices. Several modern feature detectors use homographies tosimulate different viewpoints, not only to train feature detection anddescription, but test them as well. The problem is that, often, views of indoorspaces contain high depth disparity. This makes the approximation that ahomography applied to an image represents a viewpoint change inaccurate. Weclaim that in order to train detectors to work well in indoor environments,they must be robust to this type of geometry, and repeatable under trueviewpoint change instead of homographies. Here we focus on the problem ofdetecting repeatable feature locations under true viewpoint change. To thisend, we generate labeled 2D images from a photo-realistic 3D dataset. Theseimages are used for training a neural network based feature detector. Wefurther present an algorithm for automatically generating labels of repeatable2D features, and present a fast, easy to use test algorithm for evaluating adetector in an 3D environment.

Training Deep Neural Networks to Detect Repeatable 2D Features Using Large Amounts of 3D World Capture Data