OpenScan: A Benchmark for Generalized Open-Vocabulary 3D Scene Understanding
Paper • 2408.11030 • Published
text string |
|---|
5028 |
5028 |
5028 |
5028 |
5028 |
5028 |
5028 |
1 |
1 |
5028 |
5028 |
5028 |
5028 |
1 |
5028 |
5028 |
1 |
1 |
1 |
1 |
5028 |
1 |
5028 |
5028 |
5028 |
5028 |
1 |
1 |
5028 |
5028 |
5028 |
5028 |
5028 |
5028 |
1 |
5028 |
5028 |
5028 |
5028 |
5028 |
5028 |
5028 |
5028 |
5028 |
1 |
1 |
1 |
1 |
1 |
5028 |
105029 |
105029 |
11001 |
5028 |
105029 |
105029 |
105029 |
105026 |
105029 |
105029 |
105026 |
1 |
105029 |
105029 |
105029 |
105029 |
105029 |
105029 |
2022 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
105029 |
105029 |
105029 |
105029 |
105029 |
2022 |
2022 |
2022 |
2022 |
2022 |
2022 |
2022 |
2022 |
1 |
1 |
1 |
1 |
1 |
105026 |
105026 |
105026 |
105026 |
105026 |
2022 |
OpenScan is a benchmark for generalized open-vocabulary 3D scene understanding. It extends ScanNet200-style scene understanding with eight object-attribute aspects: material, affordance, property, type, manner, synonyms, requirement, and element.
This dataset mirror contains the OpenScan validation annotations and evaluation ground-truth files for the AAAI 2026 paper.
| File | Description |
|---|---|
openscan_val_raw.json |
Raw validation annotations with one record per scene object. |
openscan_label_mapping/ |
Expanded TSV label mappings for all eight OpenScan aspects. |
openscan_val_gt/ |
Processed validation ground-truth files for evaluation. |
The raw validation JSON contains 10,388 annotated objects from 312 validation scenes. The processed ground truth contains 312 scene files for each of the eight OpenScan aspects.
Each raw annotation record has this structure:
{
"scene_id": "scene0011_00",
"object_id": "0",
"object_name": "floor",
"material": "stone",
"affordance": "stand",
"property": "other",
"type": "land",
"manner": "other",
"synonyms": "other",
"requirement": "other",
"element": "other"
}
Load the raw validation annotation table with datasets:
from datasets import load_dataset
dataset = load_dataset(
"json",
data_files="hf://datasets/garrying/OpenScan/openscan_val_raw.json",
split="train",
)
Download the processed validation ground truth for evaluation:
hf download garrying/OpenScan --repo-type dataset --local-dir . --include "openscan_val_gt/**"
The openscan_val_gt/ directory contains:
gt_affordance/
gt_element/
gt_manner/
gt_material/
gt_property/
gt_requirement/
gt_synonyms/
gt_type/
@article{zhao2024openscan,
title={OpenScan: A Benchmark for Generalized Open-Vocabulary 3D Scene Understanding},
author={Zhao, Youjun and Lin, Jiaying and Ye, Shuquan and Pang, Qianshi and Lau, Rynson WH},
journal={arXiv preprint arXiv:2408.11030},
year={2024}
}