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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Machine perception applications are increasingly moving toward manipulatingand processing 3D point cloud. This paper focuses on point cloud registration,a key primitive of 3D data processing widely used in high-level tasks such asodometry, simultaneous localization and mapping, and 3D reconstruction. Asthese applications are routinely deployed in energy-constrained environments,real-time and energy-efficient point cloud registration is critical.  We present Tigris, an algorithm-architecture co-designed system specializedfor point cloud registration. Through an extensive exploration of theregistration pipeline design space, we find that, while different design pointsmake vastly different trade-offs between accuracy and performance, KD-treesearch is a common performance bottleneck, and thus is an ideal candidate forarchitectural specialization. While KD-tree search is inherently sequential, wepropose an acceleration-amenable data structure and search algorithm thatexposes different forms of parallelism of KD-tree search in the context ofpoint cloud registration. The co-designed accelerator systematically exploitsthe parallelism while incorporating a set of architectural techniques thatfurther improve the accelerator efficiency. Overall, Tigris achieves77.2$\times$ speedup and 7.4$\times$ power reduction in KD-tree search over anRTX 2080 Ti GPU, which translates to a 41.7% registration performanceimprovements and 3.0$\times$ power reduction.

Tigris: Architecture and Algorithms for 3D Perception in Point Clouds