SSC-RS: Elevate LiDAR Semantic Scene Completion with Representation Separation and BEV Fusion

TL;DR

This paper introduces SSC-RS, a novel approach for outdoor semantic scene completion that uses representation separation and BEV fusion to improve accuracy and efficiency in autonomous driving applications.

Contribution

The paper proposes a new network architecture with separate branches for semantic and geometric representations and an adaptive BEV fusion module, advancing SSC performance.

Findings

Achieves state-of-the-art results on SemanticKITTI

Operates in real-time with low computational cost

Effectively disentangles semantic and geometric features

Abstract

Semantic scene completion (SSC) jointly predicts the semantics and geometry of the entire 3D scene, which plays an essential role in 3D scene understanding for autonomous driving systems. SSC has achieved rapid progress with the help of semantic context in segmentation. However, how to effectively exploit the relationships between the semantic context in semantic segmentation and geometric structure in scene completion remains under exploration. In this paper, we propose to solve outdoor SSC from the perspective of representation separation and BEV fusion. Specifically, we present the network, named SSC-RS, which uses separate branches with deep supervision to explicitly disentangle the learning procedure of the semantic and geometric representations. And a BEV fusion network equipped with the proposed Adaptive Representation Fusion (ARF) module is presented to aggregate the multi-scale features effectively and efficiently. Due to the low computational burden and powerful representation ability, our model has good generality while running in real-time. Extensive experiments on SemanticKITTI demonstrate our SSC-RS achieves state-of-the-art performance.