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

Self-driving cars must detect vehicles, pedestrians, and other trafficparticipants accurately to operate safely. Small, far-away, or highly occludedobjects are particularly challenging because there is limited information inthe LiDAR point clouds for detecting them. To address this challenge, weleverage valuable information from the past: in particular, data collected inpast traversals of the same scene. We posit that these past data, which aretypically discarded, provide rich contextual information for disambiguating theabove-mentioned challenging cases. To this end, we propose a novel, end-to-endtrainable Hindsight framework to extract this contextual information from pasttraversals and store it in an easy-to-query data structure, which can then beleveraged to aid future 3D object detection of the same scene. We show thatthis framework is compatible with most modern 3D detection architectures andcan substantially improve their average precision on multiple autonomousdriving datasets, most notably by more than 300% on the challenging cases.

Hindsight is 20/20: Leveraging Past Traversals to Aid 3D Perception