RGBD20K: A Large-Scale Benchmark for RGB-D Semantic Segmentation

Hugging Face Daily Papers Papers
rgb-d semantic-segmentation benchmark dataset large-scale computer-vision

Summary

This paper introduces RGBD20K, a large-scale benchmark dataset for RGB-D semantic segmentation with 160 fine-grained categories and 20,000 image pairs, featuring high-quality annotations and a novel score-purified fusion method.

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/.
Original Article
View Cached Full Text

Cached at: 09/25/26, 07:47 PM

Paper page - RGBD20K: A Large-Scale Benchmark for RGB-D Semantic Segmentation

Source: https://huggingface.co/papers/2609.29028

Abstract

Inthispaper,weproposeRGBD20K,anoveldatasetforfacilitatingthedevelopmentofmorerobustandgeneralRGB-Dsemanticsegmentationbyencompassingabundantcategoriesandhigh-qualityannotations.RGBD20Kpossessesseveralattractiveproperties:(1)ExpandedSemanticSpace.Inparticular,itcovers160fine-grainedcategories,largelysurpassingthecategorydiversityofexistingpopularRGB-Dbenchmarks(e.g.,NYUv2with40classesandSUNRGB-Dwith37classes).Withsuchenrichedsemanticcoverage,weexpecttopromotethelearningofmoregeneralizablesegmentationmodels.(2)LargerScale.Comparedwithcurrentbenchmarks,RGBD20Koffers20,000RGB-Dimagepairs,providingasubstantiallylargertrainingresourcethatbenefitsthedevelopmentofmorepowerfuldeepmodels.(3)High-FidelityAnnotation.Weperformrigorousre-evaluationandcorrectionofexistinglabelstoresolvelong-standingannotationnoise,resultinginacleanandreliableground-truthfoundation.Furthermore,weproposeanovelscore-purifiedfusion(SPF)method,whichachievesstate-of-the-artperformanceacrossallevaluatedbenchmarks,demonstratingtheeffectivenessofourapproachinleveraginghigh-qualitymultimodalinformationforRGB-Dsemanticsegmentation.Thedatasetishere:https://github.com/ShaohuaDong2021/RGBD20K/.

View arXiv pageView PDFGitHub4Add to collection

Get this paper in your agent:

hf papers read 2609\.29028

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2609.29028 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2609.29028 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2609.29028 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

CM-EVS: Sparse Panoramic RGB-D-Pose Data for Complete Scene Coverage

Hugging Face Daily Papers

This paper proposes COVER, a training-free method for converting 3D assets into sparse panoramic RGB-D-pose data with complete scene coverage and low redundancy, and introduces the CM-EVS dataset containing 36,373 curated frames from indoor and outdoor scenes.

Toward Robust and 3D-Aware RGB-NIR Imaging in the Dark

Hugging Face Daily Papers

This paper presents a 3D-aware neural approach for RGB-NIR low-light imaging that fuses extremely noisy RGB observations with NIR cues in 3D space, eliminating the need for clean RGB supervision and improving robustness across different noise levels.