---
license: cc-by-4.0
task_categories:
- text-to-image
pretty_name: REVISION_GENERATOR
language:
- en
---
# REVISION: Rendering Tools Enable Spatial Fidelity in Vision-Language Models (ECCV 2024)

<img src="./misc/revision_genenator.png" width="80%" height="80%"/>


This is the official dataset of the REVISION framework with all the corresponding assets (i.e. objects, backgrounds, and floors). 


## ⚒️ Requirements

REVISION requires [blenderproc](https://github.com/DLR-RM/BlenderProc). Simply install it with pip:

```
pip install blenderproc
```

## 👁️ Single Test Run

<img src="./misc/spatial_rel.png" width="80%" height="80%"/>

To generate a two-object reference image deterministically on your own, you may invoke one of the 4 blenderproc scripts in `util/`. E.g., to generate a scene of 'an **apple** *to the left* of a **banana**' in an indoor background, you may use

```
blenderproc run util/blender_left_right_floor.py apple banana background/photo_studio_loft_hall_2k.hdr output/debug/ 0 0

```
The command above is equivalent for generating 'a **banana** *to the right* of an **apple**' in an indoor background.

## 🏃 Batched Test Run

We also provide ``revision_gen_sample_t2i_comp.sh`` or `` revision_gen_sample_mscoco.sh`` to synthesize a sample batch of REVISION reference images in hdf5 format. You may then visualize the reference images with: 

```
blenderproc vis hdf5 <path_to_ref_images>/<image_name>.hdf5
```

## 🖼️ Sample rendered outputs

For convenience, we also have provided rendered outputs in PNG format for all two-object-pairs in MSCOCO or those specified in T2I-CompBench. These images are also the ones used in the RevQA Benchmark. Please find out more under the folder [sample_output/](https://huggingface.co/datasets/revision-t2i/revision-generator/tree/main/sample_output) .

## 🤝🏼 Citation

```bibtex
@misc{chatterjee2024revisionrenderingtoolsenable,
      title={REVISION: Rendering Tools Enable Spatial Fidelity in Vision-Language Models}, 
      author={Agneet Chatterjee and Yiran Luo and Tejas Gokhale and Yezhou Yang and Chitta Baral},
      year={2024},
      eprint={2408.02231},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2408.02231}, 
}
```

## 💖 Acknowledgement

The floor textures and the object models are sourced and modified from [sketchfab.com](https://sketchfab.com). The textured background assets are sourced from [polyhaven.com](http://polyhaven.com). All assets are shared in accordance with [CC-BY-4.0 License](https://creativecommons.org/licenses/by/4.0/deed.en#:~:text=https%3A//creativecommons.org/licenses/by/4.0/).