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-supervised image networks can be used to address complex 2D tasks (e.g.,semantic segmentation, object discovery) very efficiently and with little or nodownstream supervision. However, self-supervised 3D networks on lidar data donot perform as well for now. A few methods therefore propose to distillhigh-quality self-supervised 2D features into 3D networks. The most recent onesdoing so on autonomous driving data show promising results. Yet, a performancegap persists between these distilled features and fully-supervised ones. Inthis work, we revisit 2D-to-3D distillation. First, we propose, for semanticsegmentation, a simple approach that leads to a significant improvementcompared to prior 3D distillation methods. Second, we show that distillation inhigh capacity 3D networks is key to reach high quality 3D features. Thisactually allows us to significantly close the gap between unsuperviseddistilled 3D features and fully-supervised ones. Last, we show that ourhigh-quality distilled representations can also be used for open-vocabularysegmentation and background/foreground discovery.

Revisiting the Distillation of Image Representations into Point Clouds for Autonomous Driving