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

A thorough and holistic scene understanding is crucial for autonomousvehicles, where LiDAR semantic segmentation plays an indispensable role.However, most existing methods focus on the network design while neglecting theinherent difficulty, imbalanced data distribution in the realistic dataset(also named long-tailed distribution), which narrows down the capability ofstate-of-the-art methods. In this paper, we propose an input-output balancedframework to handle the issue of long-tailed distribution. Specifically, forthe input space, we synthesize these tailed instances from mesh models and wellsimulate the position and density distribution of LiDAR scan, which enhancesthe input data balance and improves the data diversity. For the output space, amulti-head block is proposed to group different categories based on theirshapes and instance amounts, which alleviates the biased representation ofdominating category during the feature learning. We evaluate the proposed modelon two large-scale datasets, SemanticKITTI and nuScenes, where state-of-the-artresults demonstrate its effectiveness. The proposed new modules can also beused as a plug-and-play, and we apply them on various backbones and datasets,showing its good generalization ability.

Input-Output Balanced Framework for Long-tailed LiDAR Semantic Segmentation