Back to datasets
Dataset assetOpen Source CommunityScene AnalysisMultimodal Imaging
VDT-2048 Dataset
The dataset comprises 2,048 image sets, each containing three modalities (visible‑light, depth, and thermal images), all at a resolution of 640 × 480. It captures images of 34 household items across seven of the most common home scenarios.
Source
github
Created
Mar 19, 2022
Updated
May 19, 2024
Signals
225 views
Availability
Linked source ready
Overview
Dataset description and usage context
VDT-2048 Dataset Overview
Dataset Content
- Number and Types of Images: 2,048 image sets, each containing three modalities: visible‑light, depth, and thermal images.
- Image Resolution: All images have a resolution of 640 × 480.
- Scenes and Items: The dataset includes 34 household items distributed across seven common home scenes.
Challenging Scenarios
- V Modality: Includes seven challenging scenarios such as V‑SA (Similar Appearance), V‑BSO (Big Salient Object), V‑SSO (Small Salient Object), V‑MSO (Multiple Salient Objects), V‑LI (Low Illumination), V‑SI (Side Illumination), V‑NI (No Illumination).
- D Modality: Includes four challenging scenarios such as D‑BM (Background Clutter), D‑II (Incomplete Information), D‑SSO (Small Salient Object), D‑BI (Background Interference).
- T Modality: Includes three challenging scenarios such as T‑Cr (Cross), T‑RD (Radiation Dispersion), T‑HR (Thermal Reflection).
Dataset Download
- Download Link: The dataset and code are available at the following link: Download Link
Citation
- Paper citation: K. Song, J. Wang, Y. Bao, L. Huang and Y. Yan, "A Novel Visible‑Depth‑Thermal Image Dataset of Salient Object Detection for Robotic Visual Perception," in IEEE/ASME Transactions on Mechatronics, vol. 28, no. 3, pp. 1558‑1569, June 2023, doi: 10.1109/TMECH.2022.3215909.
Need downstream help?
Pair the dataset with AI analysis and content workflows.
Once the source passes your review, move straight into summarization, transformation, report drafting, or presentation generation with the JuheAI toolchain.