DC-Scene: Data-Centric Learning for 3D Scene Understanding

TL;DR

DC-Scene introduces a data-centric approach for 3D scene understanding that improves training efficiency and performance by filtering high-quality data and reducing reliance on large datasets, demonstrated on ScanRefer and Nr3D.

Contribution

The paper proposes a novel CLIP-driven dual-indicator quality filter and curriculum scheduler to enhance data quality and training efficiency in 3D scene understanding.

Findings

Achieves state-of-the-art 86.1 CIDEr score with top-75% data subset.

Reduces training cost by approximately two-thirds.

High-quality data filtering outperforms using full datasets.

Abstract

3D scene understanding plays a fundamental role in vision applications such as robotics, autonomous driving, and augmented reality. However, advancing learning-based 3D scene understanding remains challenging due to two key limitations: (1) the large scale and complexity of 3D scenes lead to higher computational costs and slower training compared to 2D counterparts; and (2) high-quality annotated 3D datasets are significantly scarcer than those available for 2D vision. These challenges underscore the need for more efficient learning paradigms. In this work, we propose DC-Scene, a data-centric framework tailored for 3D scene understanding, which emphasizes enhancing data quality and training efficiency. Specifically, we introduce a CLIP-driven dual-indicator quality (DIQ) filter, combining vision-language alignment scores with caption-loss perplexity, along with a curriculum scheduler that progressively expands the training pool from the top 25% to 75% of scene-caption pairs. This strategy filters out noisy samples and significantly reduces dependence on large-scale labeled 3D data. Extensive experiments on ScanRefer and Nr3D demonstrate that DC-Scene achieves state-of-the-art performance (86.1 CIDEr with the top-75% subset vs. 85.4 with the full dataset) while reducing training cost by approximately two-thirds, confirming that a compact set of high-quality samples can outperform exhaustive training. Code will be available at https://github.com/AIGeeksGroup/DC-Scene.