Learning Temporal 3D Semantic Scene Completion via Optical Flow Guidance

TL;DR

FlowScene introduces optical flow guidance into 3D semantic scene completion, effectively capturing motion dynamics and improving temporal consistency for autonomous driving perception.

Contribution

The paper presents a novel framework that leverages optical flow for temporal feature aggregation and occlusion-aware voxel refinement in 3D scene completion.

Findings

Achieves state-of-the-art results on SemanticKITTI and SSCBench-KITTI-360.

Effectively models motion and occlusion for improved scene understanding.

Outperforms existing methods in accuracy and temporal consistency.

Abstract

3D Semantic Scene Completion (SSC) provides comprehensive scene geometry and semantics for autonomous driving perception, which is crucial for enabling accurate and reliable decision-making. However, existing SSC methods are limited to capturing sparse information from the current frame or naively stacking multi-frame temporal features, thereby failing to acquire effective scene context. These approaches ignore critical motion dynamics and struggle to achieve temporal consistency. To address the above challenges, we propose a novel temporal SSC method FlowScene: Learning Temporal 3D Semantic Scene Completion via Optical Flow Guidance. By leveraging optical flow, FlowScene can integrate motion, different viewpoints, occlusions, and other contextual cues, thereby significantly improving the accuracy of 3D scene completion. Specifically, our framework introduces two key components: (1) a Flow-Guided Temporal Aggregation module that aligns and aggregates temporal features using optical flow, capturing motion-aware context and deformable structures; and (2) an Occlusion-Guided Voxel Refinement module that injects occlusion masks and temporally aggregated features into 3D voxel space, adaptively refining voxel representations for explicit geometric modeling. Experimental results demonstrate that FlowScene achieves state-of-the-art performance on the SemanticKITTI and SSCBench-KITTI-360 benchmarks.