Enhancing Indoor Occupancy Prediction via Sparse Query-Based Multi-Level Consistent Knowledge Distillation

TL;DR

DiScene introduces a sparse query-based framework with multi-level knowledge distillation and teacher-guided initialization, significantly improving efficiency and robustness in indoor occupancy prediction for robotics applications.

Contribution

The paper proposes DiScene, a novel sparse query-based occupancy prediction method utilizing multi-level distillation and optimized warm-up, achieving state-of-the-art performance and faster inference.

Findings

Achieves 23.2 FPS without depth priors, outperforming baseline by 36.1%.

Surpasses existing methods like EmbodiedOcc by 3.7% with faster inference.

Demonstrates versatility across various indoor environments and benchmarks.

Abstract

Occupancy prediction provides critical geometric and semantic understanding for robotics but faces efficiency-accuracy trade-offs. Current dense methods suffer computational waste on empty voxels, while sparse query-based approaches lack robustness in diverse and complex indoor scenes. In this paper, we propose DiScene, a novel sparse query-based framework that leverages multi-level distillation to achieve efficient and robust occupancy prediction. In particular, our method incorporates two key innovations: (1) a Multi-level Consistent Knowledge Distillation strategy, which transfers hierarchical representations from large teacher models to lightweight students through coordinated alignment across four levels, including encoder-level feature alignment, query-level feature matching, prior-level spatial guidance, and anchor-level high-confidence knowledge transfer and (2) a Teacher-Guided Initialization policy, employing optimized parameter warm-up to accelerate model convergence. Validated on the Occ-Scannet benchmark, DiScene achieves 23.2 FPS without depth priors while outperforming our baseline method, OPUS, by 36.1% and even better than the depth-enhanced version, OPUS{\dag}. With depth integration, DiScene{\dag} attains new SOTA performance, surpassing EmbodiedOcc by 3.7% with 1.62$\times$ faster inference speed. Furthermore, experiments on the Occ3D-nuScenes benchmark and in-the-wild scenarios demonstrate the versatility of our approach in various environments. Code and models can be accessed at https://github.com/getterupper/DiScene.