Spaces:
Paused
LISA: Reasoning Segmentation via Large Language Model
News
- [2023.8.4] Online Demo is released!
- [2023.8.4] ReasonSeg Dataset and the LISA-13B-llama2-v0-explainatory model are released!
- [2023.8.3] Inference code and the LISA-13B-llama2-v0 model are released. Welcome to check out!
- [2023.8.2] Paper is released and GitHub repo is created.
TODO
- Training Code Release
LISA: Reasoning Segmentation Via Large Language Model [Paper]
Xin Lai,
Zhuotao Tian,
Yukang Chen,
Yanwei Li,
Yuhui Yuan,
Shu Liu,
Jiaya Jia
Abstract
In this work, we propose a new segmentation task --- reasoning segmentation. The task is designed to output a segmentation mask given a complex and implicit query text. We establish a benchmark comprising over one thousand image-instruction pairs, incorporating intricate reasoning and world knowledge for evaluation purposes. Finally, we present LISA: Large-language Instructed Segmentation Assistant, which inherits the language generation capabilities of the multi-modal Large Language Model (LLM) while also possessing the ability to produce segmentation masks. For more details, please refer to the paper.
Highlights
LISA unlocks the new segmentation capabilities of multi-modal LLMs, and can handle cases involving:
- complex reasoning;
- world knowledge;
- explanatory answers;
- multi-turn conversation.
LISA also demonstrates robust zero-shot capability when trained exclusively on reasoning-free datasets. In addition, fine-tuning the model with merely 239 reasoning segmentation image-instruction pairs results in further performance enhancement.
Experimental results
Installation
pip install -r requirements.txt
Inference
To chat with LISA-13B-llama2-v0 or LISA-13B-llama2-v0-explainatory: (Note that LISA-13B-llama2-v0 currently does not support explanatory answers.)
CUDA_VISIBLE_DEVICES=0 python3 chat.py --version='xinlai/LISA-13B-llama2-v0'
To use bf16 or fp16 data type for inference:
CUDA_VISIBLE_DEVICES=0 python3 chat.py --version='xinlai/LISA-13B-llama2-v0' --precision='bf16'
To use 8bit or 4bit data type for inference (this enables running 13B model on a single 24G or 12G GPU at some cost of generation quality):
CUDA_VISIBLE_DEVICES=0 python3 chat.py --version='xinlai/LISA-13B-llama2-v0' --precision='fp16' --load_in_8bit
CUDA_VISIBLE_DEVICES=0 python3 chat.py --version='xinlai/LISA-13B-llama2-v0' --precision='fp16' --load_in_4bit
After that, input the text prompt and then the image path. For example，
- Please input your prompt: Where can the driver see the car speed in this image? Please output segmentation mask.
- Please input the image path: imgs/example1.jpg
- Please input your prompt: Can you segment the food that tastes spicy and hot?
- Please input the image path: imgs/example2.jpg
The results should be like:
Dataset
In ReasonSeg, we have collected 1218 images (239 train, 200 val, and 779 test). The training and validation sets can be download from this link.
Each image is provided with an annotation JSON file:
image_1.jpg, image_1.json
image_2.jpg, image_2.json
...
image_n.jpg, image_n.json
Important keys contained in JSON files:
- "text": text instructions.
- "is_sentence": whether the text instructions are long sentences.
- "shapes": target polygons.
The elements of the "shapes" exhibit two categories, namely "target" and "ignore". The former category is indispensable for evaluation, while the latter category denotes the ambiguous region and hence disregarded during the evaluation process.
We provide a script that demonstrates how to process the annotations:
python3 utils/data_proc_demo.py
Besides, we leveraged GPT-3.5 for rephrasing instructions, so images in the training set may have more than one instructions (but fewer than six) in the "text" field. During training, users may randomly select one as the text query to obtain a better model.
Citation
If you find this project useful in your research, please consider citing:
@article{reason-seg,
title={LISA: Reasoning Segmentation Via Large Language Model},
author={Xin Lai and Zhuotao Tian and Yukang Chen and Yanwei Li and Yuhui Yuan and Shu Liu and Jiaya Jia},
journal={arXiv:2308.00692},
year={2023}
}