|
--- |
|
inference: false |
|
pipeline_tag: image-text-to-text |
|
license: apache-2.0 |
|
datasets: |
|
- Zery/BS-Objaverse |
|
- Lin-Chen/ShareGPT4V |
|
language: |
|
- en |
|
--- |
|
<br> |
|
<br> |
|
|
|
# MV-LLaVA-7B Model Card |
|
|
|
## Model details |
|
|
|
**Model type:** |
|
MV-LLaVA-7B is an open-source chatbot for 3D multi-view images trained by fine-tuning CLIP vision tower and LLaMA/Vicuna on GPT4-Vision-assisted [BS-Objaverse](https://huggingface.co/datasets/Zery/BS-Objaverse) data and [ShareGPT4V](https://huggingface.co/datasets/Lin-Chen/ShareGPT4V) data. |
|
|
|
**Model date:** |
|
MV-LLaVA-7B was trained in Apr, 2024. |
|
|
|
**Paper or resources for more information:** |
|
[[Project](https://sunzey.github.io/Bootstrap3D/)] [[Paper](https://huggingface.co/papers/2406.00093)] [[Code](https://github.com/SunzeY/Bootstrap3D)] |
|
|
|
## Usage |
|
You can directly utilize this model as we provide in our [[repository](https://github.com/SunzeY/Bootstrap3D/tree/main/MV_LLaVA)]. |
|
## License |
|
Llama 2 is licensed under the LLAMA 2 Community License, |
|
Copyright (c) Meta Platforms, Inc. All Rights Reserved. |
|
## Intended use |
|
**Primary intended uses:** |
|
The primary use of ShareGPT4V-7B is research on large multimodal models and chatbots for 3D content. |
|
**Primary intended users:** |
|
The primary intended users of the model are researchers and hobbyists in computer vision, natural language processing, machine learning, and artificial intelligence. |
|
## Training dataset |
|
- 1.2M ShareGPT4V-PT data |
|
- 30K GPT4-Vision-generated multi-view image-text pairs |
|
- LLaVA instruction-tuning data |
