Groma: Grounded Multimodal Assistant

> [**Groma: Localized Visual Tokenization for Grounding Multimodal Large Language Models**](https://arxiv.org/abs/2404.13013) > **Chuofan Ma, Yi Jiang, Jiannan Wu, Zehuan Yuan, Xiaojuan Qi** > >

Groma is an MLLM with exceptional region understanding and visual grounding capabilities. It can take user-defined region inputs (boxes) as well as generate long-form responses that are grounded to visual context.

Groma presents a novel paradigm of grounded MLLMs. (a) LLM for localization (e.g., Kosmos-2, Shikra); (b) External modules for localization (e.g., Lisa); and (c) Visual tokenier for localization (Groma).

## Contents - [Install](#installation) - [Model](#model-weights) - [Data](#prepare-data) - [Training](#training) - [Inference](#inference) - [Evaluation](#evaluation) ## Performance State-of-the-art performance on referring expression comprehension (REC) benchmarks among multimodal large language models.

Method	RefCOCO			RefCOCO+			RefCOCOg		Avergae
Method	val	testA	testB	val	testA	testB	val	test	Avergae
Shikra	87.01	90.61	80.24	81.60	87.36	72.12	82.27	82.19	82.93
Ferret	87.49	91.35	82.45	80.78	87.38	73.14	83.93	84.76	83.91
MiniGPT-v2	88.69	91.65	85.33	79.97	85.12	74.45	84.44	84.66	84.29
Qwen-VL	89.36	92.26	85.34	83.12	88.25	77.21	85.58	85.48	85.83
Groma	89.53	92.09	86.26	83.90	88.91	78.05	86.37	87.01	86.52

## Installation Clone the repository ~~~ git clone https://github.com/FoundationVision/Groma.git cd Groma ~~~ Create the conda environment and install dependencies ~~~ conda create -n groma python=3.9 -y conda activate groma conda install pytorch==2.1.0 torchvision==0.16.0 torchaudio==2.1.0 pytorch-cuda=11.8 -c pytorch -c nvidia pip install --upgrade pip # enable PEP 660 support pip install -e . cd mmcv MMCV_WITH_OPS=1 pip install -e . cd .. ~~~ Install falsh-attention for training ~~~ pip install ninja pip install flash-attn --no-build-isolation ~~~ ## Model Weights To play with Groma, please download the [model weights](https://huggingface.co/FoundationVision/groma-7b-finetune) from huggingface. We additionally provide pretrained checkpoints from intermediate training stages. You can start from any point to customize training. | Training stage | Required checkpoints | |:--------------:|:--------------------:| | Detection pretraining | [DINOv2-L](https://huggingface.co/facebook/dinov2-large) | | Alignment pretraining | [Vicuna-7b-v1.5](https://huggingface.co/lmsys/vicuna-7b-v1.5), [Groma-det-pretrain](https://huggingface.co/FoundationVision/groma-det-pretrain) | | Instruction finetuning | [Groma-7b-pretrain](https://huggingface.co/FoundationVision/groma-7b-pretrain) | ## Prepare Data We provide instructions to download datasets used at different training stages of Groma, including [Groma Instruct](https://huggingface.co/datasets/FoundationVision/groma_instruct/), a 30k viusally grounded conversation dataset constructed with GPT-4V. You don't have to download all of them unless you want to train Groma from scratch. Please follow instructions in [DATA.md](docs/DATA.md) to prepare datasets.

Training stage	Data types	Datasets
Detection pretraining	Detection	COCO, Objects365, OpenImages, V3Det, SA1B
Alignment pretraining	Image caption	ShareGPT-4V-PT
	Grounded caption	Flickr30k Entities
	Region caption	Visual Genome, RefCOCOg
	REC	COCO, RefCOCO/g/+, Grit-20m
Instruction finetuning	Grounded caption	Flickr30k Entities
	Region caption	Visual Genome, RefCOCOg
	REC	COCO, RefCOCO/g/+
	Instruction following	Groma Instruct, LLaVA Instruct, ShareGPT-4V

## Training For detection pretraining, please run ~~~ bash scripts/det_pretrain.sh {path_to_dinov2_ckpt} {output_dir} ~~~ For alignment pretraining, please run ~~~ bash scripts/vl_pretrain.sh {path_to_vicuna_ckpt} {path_to_groma_det_pretrain_ckpt} {output_dir} ~~~ For instruction finetuning, please run ~~~ bash scripts/vl_finetune.sh {path_to_groma_7b_pretrain_ckpt} {output_dir} ~~~ ## Inference To test on single image, you can run ~~~ python -m groma.eval.run_groma \ --model-name {path_to_groma_7b_finetune} \ --image-file {path_to_img} \ --query {user_query} \ --quant_type 'none' # support ['none', 'fp16', '8bit', '4bit'] for inference ~~~ ## Evaluation For evaluation, please refer to [EVAL.md](docs/EVAL.md) for more details. ## Citation If you find this repo useful for your research, feel free to give us a star ⭐ or cite our paper: ``` @article{ma2024groma, title={Groma: Localized Visual Tokenization for Grounding Multimodal Large Language Models}, author={Ma, Chuofan and Jiang, Yi and Wu, Jiannan and Yuan, Zehuan and Qi, Xiaojuan}, journal={arXiv preprint arXiv:2404.13013}, year={2024} } ``` ## Acknowledgement Groma is built upon the awesome works [LLaVA](https://github.com/haotian-liu/LLaVA/) and [GPT4ROI](https://github.com/jshilong/GPT4RoI). ## LICENSE This project is licensed under the Apache License 2.0 - see the [LICENSE](LICENSE) file for details.