💥 GiT: the first successful general vision model unifies various vision tasks only with a vanilla ViT, fully aligned with the architecture of LLM

This repo is the official implementation of paper: GiT: Towards Generalist Vision Transformer through Universal Language Interface as well as the follow-ups. We have made every effort to ensure that the codebase is clean, concise, easily readable, state-of-the-art, and relies only on minimal dependencies.

GiT: Towards Generalist Vision Transformer through Universal Language Interface

Haiyang Wang*, Hao Tang*, Li Jiang $^\dagger$, Shaoshuai Shi, Muhammad Ferjad Naeem, Hongsheng Li, Bernt Schiele, Liwei Wang $^\dagger$

• Primary contact: Haiyang Wang ( [email protected] ), Hao Tang ( [email protected] )

Overview

• 🤔 Introduction
• 👀 Todo
• 🚀 Main Results
• 🛠️ Quick Start
• 👍 Acknowledgments
• 📘 Citation

🤔 Introduction

Building a universal computation model across all tasks stands as the cornerstone of artificial intelligence, reducing the need for task-specific designs. In this project, we introduce GiT (Generalist Vision Transformer). GiT has the following characteristics:

• 😮 Minimalist architecture design similar to LLM: GiT consists solely of a single transformer, without the inclusion of additional vision encoders and adapters.
• 🚀 Covering all types of visual understanding tasks: GiT addresses a spectrum of visual tasks, including object-level tasks (e.g., object detection), pixel-level tasks (e.g., semantic segmentation), and vision-language tasks (e.g., image captioning).
• 🤗 Achieving multi-task ability by unified language interface: Similar to LLM, GiT observes the task synergy effect in multi-task training. It fosters mutual enhancement across tasks, leading to significant improvements compared to isolated training.
• 🔥 Stong performance on zero-shot and few-shot benchmark: GiT scales well with model size and data, demonstrating remarkable generalizability across diverse scenarios after training on 27 datasets.

📣 News

• [24-3-15] 🚀 Training and inference Code is released.
• [24-3-15] 👀 GiT is released on arXiv.

👀 Todo

• Release the arXiv version.
• SOTA performance of generalist model on multi-tasking benchmark.
• SOTA performance of generalist model on zero- and few-shot benchmark.
• Clean up and release the inference code.
• Clean up and release the training code.
• Engineering Optimization (faster).
• Joint Training including Language (stronger).

🚀 Main Results

Single-Task Benchmark

Model	Params	Metric	Perfomance	ckpt	log	config
GiT-Bdetection	131M	mAP	45.1	ckpt	log	config
GiT-Binsseg	131M	mAP	31.4	ckpt	log	config
GiT-Bsemseg	131M	mIoU	47.7	ckpt	log	config
GiT-Bcaption	131M	BLEU-4	33.7	ckpt	log	config
GiT-Bgrounding	131M	[email protected]	83.3	ckpt	log	config

Multi-Tasking Benchmark

Model	Params	Detection	Ins Seg	Sem Seg	Caption	Grounding	ckpt	log	config
GiT-Bmulti-task	131M	46.7	31.9	47.8	35.3	85.8	ckpt	log	config
GiT-Lmulti-task	387M	51.3	35.1	50.6	35.7	88.4	ckpt	log	config
GiT-Hmulti-task	756M	52.9	35.8	52.4	36.2	89.2	ckpt	log	config

Task Synergy in Multi-Tasking Training

Model	Params	Detection	Ins Seg	Sem Seg	Caption	Grounding
GiT-Bsingle-task	131M	45.1	31.4	47.7	33.7	83.3
Improvement		+1.6	+0.5	+0.1	+1.6	+2.5
GiT-Bmulti-task	131M	46.7	31.9	47.8	35.3	85.8

Zero-shot benchmark

Model	Params	Cityscapes (Det)	Cityscapes (Ins Seg)	Cityscapes (Sem Seg)	SUN RGB-D	nocaps	ckpt	log	config
GiT-Bmulti-task	131M	21.8	14.3	34.4	30.9	9.2	ckpt	log	config
GiT-Buniversal	131M	29.1	17.9	56.2	37.5	10.6	ckpt	log	config
GiT-Luniversal	387M	32.3	20.3	58.0	39.9	11.6	ckpt	log	config
GiT-Huniversal	756M	34.1	18.7	61.8	42.5	12.6	ckpt	log	config

Few-shot benchmark

Model	Params	DRIVE	LoveDA	Potsdam	WIDERFace	DeepFashion	config
GiT-Bmulti-task	131M	34.3	24.9	19.1	17.4	23.0	config
GiT-Buniversal	131M	51.1	30.8	30.6	31.2	38.3	config
GiT-Luniversal	387M	55.4	34.1	37.2	33.4	49.3	config
GiT-Huniversal	756M	57.9	35.1	43.4	34.0	52.2	config

🛠️ Quick Start

Installation

conda create -n GiT python=3.8

conda activate GiT

# We only test in 1.9.1, may be other versions are also worked.
pip install torch==1.9.1+cu111 torchvision==0.10.1+cu111 torchaudio==0.9.1 -f https://download.pytorch.org/whl/torch_stable.html

pip install -U openmim
mim install "mmengine==0.8.3"
mim install "mmcv==2.0.1"
pip install "transformers==4.31.0"

git clone [email protected]:Haiyang-W/GiT.git
cd GiT
pip install -v -e .
pip install -r requirements/optional.txt
pip install -r requirements/runtime.txt

• Please download pretrained text embedding from huggingface and organize the downloaded files as follows:

GiT
|──bert_embed.pt
|——bert_embed_large.pt
|——bert_embed_huge.pt

• (Optional) Install Java manually for image caption evaluation. Without Java, you can train image caption normally, but fail in caption evaluation.
• (Optional) Install lvis api for LVIS dataset.

# current path is ./GiT
cd ..
pip install git+https://github.com/lvis-dataset/lvis-api.git

Dataset Preparation

Multi-Tasking Dataset

Multi-tasking benchmark contains coco2017 for object detection and instance segmentation, ade20k for semantic segmentation, coco caption for image caption, and refcoco series for visual grounding.

GiT
|──data
|  |──ade
|  |  |──ADEChallengeData2016
|  |  |  |──annorations
|  |  |  |  |──training & validation
|  |  |  |──images
|  |  |  |  |──training & validation
|  |  |  |──objectInfo150.txt
|  |  |  |──sceneCategories.txt
|  |──coco
|  |  |──annotations
|  |  |  |──*.json
|  |  |──train2017
|  |  |  |──*.jpg
|  |  |──val2017
|  |  |  |──*.jpg
|  |──coco_2014
|  |  |──annotations
|  |  |  |──*.json
|  |  |  |──coco_karpathy_test.json
|  |  |  |──coco_karpathy_train.json
|  |  |  |──coco_karpathy_val_gt.json
|  |  |  |──coco_karpathy_val.json
|  |  |──train2014
|  |  |  |──*.jpg
|  |  |──val2014
|  |  |  |──*.jpg
|  |  |──refcoco
|  |  |  |──*.p

Universal Dataset

We use 27 datasets in universal training. For more details about dataset preparation, please refer to here.

🚨 We only list part of the commands (GiT-B) below. For more detailed commands, please refer to here.

Training

Single Task

Detection

bash tools/dist_train.sh configs/GiT/single_detection_base.py  ${GPU_NUM} --work-dir ${work_dir}

Multi Task

GiT-B

bash tools/dist_train.sh configs/GiT/multi_fivetask_base.py  ${GPU_NUM} --work-dir ${work_dir}

Universal Training

GiT-B

bash tools/dist_train.sh configs/GiT/universal_base.py  ${GPU_NUM} --work-dir ${work_dir}

Testing

Single Task

Detection

bash tools/dist_test.sh configs/GiT/single_detection_base.py ${ckpt_file} ${GPU_NUM} --work-dir ${work_dir}

Multi Task

GiT-B

bash tools/dist_test.sh configs/GiT/multi_fivetask_base.py ${ckpt_file} ${GPU_NUM} --work-dir ${work_dir}

Zero-shot and few-shot

Please download universal pretrain weight from huggingface and organize files as follows:

GiT
|──universal_base.pth
|——universal_large.pth
|——universal_huge.pth

Zero-shot

bash tools/dist_test.sh configs/GiT/zero-shot/zero_shot_cityscapes_det_base.py ${ckpt_file} ${GPU_NUM} --work-dir ${work_dir}

Few-shot

bash tools/dist_train.sh configs/GiT/few-shot/few_shot_drive_det_base.py ${GPU_NUM} --work-dir ${work_dir}

Customize Dataset

If you want to use GiT on your own dataset, please refer here for more details.

🚀 Lightweight Version

If your GPU memory is insufficient, you can reduce the resolution like here, where we lower the detection resolution to 672. It requires ~20 hours of training and reaches ~41.5 mAP.

👍 Acknowledgement

• MMDetection The codebase we built upon. Thanks for providing such a convenient framework.
• BLIP We extract text embedding from BLIP pretrain models and use the web caption filtered by BLIP. Thanks for their efforts in open source and cleaning the dataset.

📘 Citation

Please consider citing our work as follows if it is helpful.

@article{wang2024git,
    title={GiT: Towards Generalist Vision Transformer through Universal Language Interface},
    author={Haiyang Wang and Hao Tang and Li Jiang and Shaoshuai Shi and Muhammad Ferjad Naeem and Hongsheng Li and Bernt Schiele and Liwei Wang},
    journal={arXiv preprint arXiv:2403.09394},
    year={2024}
}

✨ Star History