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

Tracking the object 6-DoF pose is crucial for various downstream robot tasksand real-world applications. In this paper, we investigate the real-world robottask of aerial vision guidance for aerial robotics manipulation, utilizingcategory-level 6-DoF pose tracking. Aerial conditions inevitably introducespecial challenges, such as rapid viewpoint changes in pitch and roll andinter-frame differences. To support these challenges in task, we firstlyintroduce a robust category-level 6-DoF pose tracker (Robust6DoF). This trackerleverages shape and temporal prior knowledge to explore optimal inter-framekeypoint pairs, generated under a priori structural adaptive supervision in acoarse-to-fine manner. Notably, our Robust6DoF employs a Spatial-TemporalAugmentation module to deal with the problems of the inter-frame differencesand intra-class shape variations through both temporal dynamic filtering andshape-similarity filtering. We further present a Pose-Aware Discrete Servostrategy (PAD-Servo), serving as a decoupling approach to implement the finalaerial vision guidance task. It contains two servo action policies to betteraccommodate the structural properties of aerial robotics manipulation.Exhaustive experiments on four well-known public benchmarks demonstrate thesuperiority of our Robust6DoF. Real-world tests directly verify that ourRobust6DoF along with PAD-Servo can be readily used in real-world aerialrobotic applications.

Towards Real-World Aerial Vision Guidance with Categorical 6D Pose Tracker