MultiViewBench / README.md
suijinru's picture
Add diverse task previews
d0beb0a verified
|
Raw
History Blame Contribute Delete
6.89 kB
---
pretty_name: MultiView-Bench
license: mit
task_categories:
- visual-question-answering
language:
- en
tags:
- multi-view
- spatial-reasoning
- vlm-benchmark
- synthetic
size_categories:
- 1K<n<10K
configs:
- config_name: default
drop_labels: true
---
# MultiView-Bench
Evaluation data for **MultiView-Bench: A Diagnostic Benchmark for
World-Centric Multi-View Integration in VLMs**
([arXiv:2607.08970](https://arxiv.org/abs/2607.08970)).
This repository currently contains the synthetic image-and-text portion of the
benchmark. The real-world subset is not included; see the scope note below.
## Contents
- 2,400 evaluation samples across 24 benchmark variants
- 7,900 PNG views (2.94 GiB logical image data)
- One, three, or six images per sample
- English question text and deterministic coordinate answer
- Single-agent and multi-agent question schemas
The `3D Real World` / `exp3D_real_world_50` subset is intentionally excluded.
It uses 3DCoMPaT++ source assets and is not part of this release. This package
also contains no GLTF, BIN, model asset, source code, credentials, or model
outputs.
## Task previews
Each sample asks for a relative 3D position and must be solved by integrating
the evidence across its views.
### Colored-object three-view task
**Question:** What is the relative position of the olive cone to the indigo
cube?
**Answer:** `(-X, -Y, -Z)`
| Front | Side | Top |
|:---:|:---:|:---:|
| ![Front view of colored-object task 17](test/images/exp2D_three_views_2d_colors/task_17/task_17_front.png) | ![Side view of colored-object task 17](test/images/exp2D_three_views_2d_colors/task_17/task_17_side.png) | ![Top view of colored-object task 17](test/images/exp2D_three_views_2d_colors/task_17/task_17_top.png) |
### 3D isometric task
**Question:** What is the relative position of the cylinder to the cone?
**Answer:** `(0X, 0Y, -Z)`
The full sample contains six views; three are shown here.
| View 0 | View 2 | View 4 |
|:---:|:---:|:---:|
| ![Isometric view 0 of 3D task 7](test/images/exp3D_3d_isometric_two_axis_fixed/task_7/task_7_3d_view_00.png) | ![Isometric view 2 of 3D task 7](test/images/exp3D_3d_isometric_two_axis_fixed/task_7/task_7_3d_view_02.png) | ![Isometric view 4 of 3D task 7](test/images/exp3D_3d_isometric_two_axis_fixed/task_7/task_7_3d_view_04.png) |
### Full 3D reasoning (DoF=3)
**Task variant:** `3D DoF=3` (unconstrained placement along X, Y, and Z)
**Question:** What is the relative position of the cube to the cylinder?
**Answer:** `(-X, +Y, +Z)`
The full sample contains six views; three are shown here.
| View 0 | View 2 | View 4 |
|:---:|:---:|:---:|
| ![View 0 of full 3D task 1](test/images/exp3D_3d_isometric_full/task_1/task_1_3d_view_00.png) | ![View 2 of full 3D task 1](test/images/exp3D_3d_isometric_full/task_1/task_1_3d_view_02.png) | ![View 4 of full 3D task 1](test/images/exp3D_3d_isometric_full/task_1/task_1_3d_view_04.png) |
### Single-view 3D task
**Task variant:** `3D Single View`
**Question:** What is the relative position of the cube to the cylinder?
**Answer:** `(-X, +Y, +Z)`
Unlike the multi-view variants, this sample exposes only one isometric view.
![Single isometric view of 3D task 1](test/images/exp3D_3d_isometric_single/task_1/task_1_3d_view_single.png)
### Grid-assisted three-view task
**Task variant:** `2D Three Views Grids`
**Question:** What is the relative position of the cylinder to the sphere?
**Answer:** `(0X, -Y, -Z)`
The projected coordinate grids provide additional alignment cues in each view.
| Front | Side | Top |
|:---:|:---:|:---:|
| ![Front grid view of task 17](test/images/exp2D_three_views_2d_grids/task_17/task_17_front.png) | ![Side grid view of task 17](test/images/exp2D_three_views_2d_grids/task_17/task_17_side.png) | ![Top grid view of task 17](test/images/exp2D_three_views_2d_grids/task_17/task_17_top.png) |
### Rotated-coordinate task (90°)
**Task variant:** `2D Three Views Rotation (90°)`
**Question:** What is the relative position of the cube to the cone?
**Answer:** `(+X, 0Y, -Z)`
The coordinate frame is rotated to test whether the model follows the visible
axes rather than relying on a conventional orientation.
| Front | Side | Top |
|:---:|:---:|:---:|
| ![Front view of rotated task 17](test/images/exp2D_three_views_2d_rotation_90/task_17/task_17_front.png) | ![Side view of rotated task 17](test/images/exp2D_three_views_2d_rotation_90/task_17/task_17_side.png) | ![Top view of rotated task 17](test/images/exp2D_three_views_2d_rotation_90/task_17/task_17_top.png) |
### Multi-agent three-view task
**Task variant:** `2D Three Views multiagent (Rotation 0)`
**Question:** What is the relative position of the sphere to the cone?
**Answer:** `(-X, -Y, +Z)`
Each view has its own axis-pair prompt (`front_question`, `side_question`, or
`top_question`) so three view-specific responses can be combined into the
full XYZ answer.
| Front agent: XZ | Side agent: YZ | Top agent: XY |
|:---:|:---:|:---:|
| ![Front-agent view of task 17](test/images/exp2D_three_views_2d_multiagent/task_17/task_17_front.png) | ![Side-agent view of task 17](test/images/exp2D_three_views_2d_multiagent/task_17/task_17_side.png) | ![Top-agent view of task 17](test/images/exp2D_three_views_2d_multiagent/task_17/task_17_top.png) |
## Load
After upload:
```python
from datasets import load_dataset
dataset = load_dataset("suijinru/MultiViewBench", split="test")
sample = dataset[0]
print(sample["benchmark"], sample["question"], sample["answer"])
print(sample["images"])
```
For a local pre-upload check:
```python
dataset = load_dataset("imagefolder", data_dir="test", split="train")
```
`ImageFolder` derives the `images` column from the `images_file_names` list in
`test/metadata.jsonl`. The Hub recognizes columns ending in `_file_names` as
image lists.
## Fields
- `benchmark`: benchmark variant identifier
- `task_variant`: human-readable task variant label used to group corresponding
results in the paper; it is not a row number. For the 3D variants, DoF
describes the object-placement constraint: DoF=1 restricts objects to the
same axis, DoF=2 restricts them to the same plane, and DoF=3 allows
unconstrained placement in 3D space.
- `task_id`: task identifier within the variant
- `mode`: `single` or `multi`
- `question`: complete prompt for single-agent samples
- `front_question`, `side_question`, `top_question`: per-view prompts for
multi-agent samples
- `answer`: expected coordinate answer
- `images_file_names`: ordered relative paths to all views
- `num_views`: number of views
All rows are evaluation data; do not use them for training when reporting
MultiView-Bench scores.
## License
The released synthetic images, questions, answers, and metadata are available
under the [MIT License](LICENSE). Copyright (c) 2026 MultiView-Bench authors.