Grounding DINO Zira

Official pytorch implementation of ZiRa, a method for incremental vision language object detection (IVLOD)

Install

If you have a CUDA environment, please make sure the environment variable CUDA_HOME is set. We recommend using Python 3.9.

pip install -e .

Dataset

Download the coco dataset and all Odinw sub-datasets in a folder named "dataset". Then download the GroundingDINO pre-trained model from this repo. The Odinw sub-datasets can be downloaded by runing this script.

python download.py

Training

Run the scripts.

sh train_odinw13_zira.sh

Citation

If you find our work helpful for your research, please consider citing the following BibTeX entry.

@article{DBLP:journals/corr/abs-2403-01680,
  author       = {Jieren Deng and
                  Haojian Zhang and
                  Kun Ding and
                  Jianhua Hu and
                  Xingxuan Zhang and
                  Yunkuan Wang},
  title        = {Zero-shot Generalizable Incremental Learning for Vision-Language Object
                  Detection},
  journal      = {CoRR},
  volume       = {abs/2403.01680},
  year         = {2024},
  url          = {https://doi.org/10.48550/arXiv.2403.01680},
  doi          = {10.48550/ARXIV.2403.01680},
  eprinttype    = {arXiv},
  eprint       = {2403.01680},
  timestamp    = {Tue, 02 Apr 2024 16:35:34 +0200},
  biburl       = {https://dblp.org/rec/journals/corr/abs-2403-01680.bib},
  bibsource    = {dblp computer science bibliography, https://dblp.org}
}

@article{liu2023grounding,
  title={Grounding dino: Marrying dino with grounded pre-training for open-set object detection},
  author={Liu, Shilong and Zeng, Zhaoyang and Ren, Tianhe and Li, Feng and Zhang, Hao and Yang, Jie and Li, Chunyuan and Yang, Jianwei and Su, Hang and Zhu, Jun and others},
  journal={arXiv preprint arXiv:2303.05499},
  year={2023}
}