# RGBD20K：大规模 RGB-D 语义分割基准数据集

- 来源：HuggingFace Daily Papers（社区热门论文）
- 发布时间：2026-09-24 08:00
- AIHOT 分数：31
- AIHOT 链接：https://aihot.news/items/cmuhae8wa06hxro3bt8494n9i
- 原文链接：https://arxiv.org/abs/2609.29028

## AI 摘要

RGBD20K 是一个包含 20000 对 RGB-D 图像、覆盖 160 个细粒度类别的 RGB-D 语义分割数据集，类别数远超 NYUv2 的 40 类和 SUN RGB-D 的 37 类。该数据集对已有标注进行了严格重评与修正，以解决长期存在的标注噪声问题。作者同时提出 score-purified fusion（SPF）方法，在所有评测基准上取得 SOTA 性能。

## 正文

In this paper, we propose RGBD20K, a novel dataset for facilitating the development of more robust and general RGB-D semantic segmentation by encompassing abundant categories and high-quality annotations. RGBD20K possesses several attractive properties: (1) Expanded Semantic Space. In particular, it covers 160 fine-grained categories, largely surpassing the category diversity of existing popular RGB-D benchmarks (e.g., NYUv2 with 40 classes and SUN RGB-D with 37 classes). With such enriched semantic coverage, we expect to promote the learning of more generalizable segmentation models. (2) Larger Scale. Compared with current benchmarks, RGBD20K offers 20,000 RGB-D image pairs, providing a substantially larger training resource that benefits the development of more powerful deep models. (3) High-Fidelity Annotation. We perform rigorous re-evaluation and correction of existing labels to resolve long-standing annotation noise, resulting in a clean and reliable ground-truth foundation. Furthermore, we propose a novel score-purified fusion (SPF) method, which achieves state-of-the-art performance across all evaluated benchmarks, demonstrating the effectiveness of our approach in leveraging high-quality multimodal information for RGB-D semantic segmentation. The dataset is here: https://github.com/ShaohuaDong2021/RGBD20K/.