Sign in to view source links and access this dataset
Description
9,160 training examples combine the original train and validation portions of the 3D Question Answering project. The dataset is an export for the project by Shuquan Ye, Dongdong Chen, Songfang Han, and Jing Liao, published on Hugging Face. It covers scenes from scene0000_00 through scene0706_00, though some original cloud files were reportedly lost.
Use Cases
Train multimodal VQA models based on 3D scene data and associated questions.
Benchmark 3D scene understanding algorithms using the provided question-answer pairs.
Research on point cloud representation learning for reasoning tasks.
Strengths
9,160 combined training examples provide a substantial base for model development.
Based on the established ScanNet dataset, suggesting a foundation of 3D indoor scene data.
Limitations
Description metadata is limited; actual data quality requires manual inspection after download.
Column-level documentation is absent; field semantics must be inferred after download.
The dataset page notes some original cloud files were lost, which may affect completeness.
Provenance
Source
Hugging Face user 'pleaseconnectwifi', exporting data from the 3D Question Answering project.
Collection Method
Likely an export from the original project's data pipeline, combining train and validation splits.
Freshness
Last updated 2026-05-29 06:07:53; freshness should be verified.
The full description is on the dataset page; users must visit the link for complete details. License information is unknown.