Skip to content

idejie/3DSyn

Repository files navigation

3D Vision and Language Pretraining with Large-Scale Synthetic Data

Install

  1. Install conda package
conda env create --name 3dsyn --file=environments.yml
  1. install pointnet2
cd vision/pointnet2
python3 setup.py install

Prepare dataset

  1. Follow Vil3dref and download scannet data under data/scanfamily/scan_data, this folder should look like
./data/scanfamily/scan_data/
├── instance_id_to_gmm_color
├── instance_id_to_loc
├── instance_id_to_name
└── pcd_with_global_alignment
  1. Download scanrefer+referit3d, scanqa, and sqa3d, and put them under /data/scanfamily/annotations
data/scanfamily/annotations/
├── meta_data
│   ├── cat2glove42b.json
│   ├── scannetv2-labels.combined.tsv
│   ├── scannetv2_raw_categories.json
│   ├── scanrefer_corpus.pth
│   └── scanrefer_vocab.pth
├── qa
│   ├── ScanQA_v1.0_test_w_obj.json
│   ├── ScanQA_v1.0_test_wo_obj.json
│   ├── ScanQA_v1.0_train.json
│   └── ScanQA_v1.0_val.json
├── refer
│   ├── nr3d.jsonl
│   ├── scanrefer.jsonl
│   ├── sr3d+.jsonl
│   └── sr3d.jsonl
├── splits
│   ├── scannetv2_test.txt
│   ├── scannetv2_train.txt
│   └── scannetv2_val.txt
└── sqa_task
    ├── answer_dict.json
    └── balanced
        ├── v1_balanced_questions_test_scannetv2.json
        ├── v1_balanced_questions_train_scannetv2.json
        ├── v1_balanced_questions_val_scannetv2.json
        ├── v1_balanced_sqa_annotations_test_scannetv2.json
        ├── v1_balanced_sqa_annotations_train_scannetv2.json
        └── v1_balanced_sqa_annotations_val_scannetv2.json

Pretrain

To pretrain the model, use the following command:

python3 run.py --config config/pretrain/pretrained.yml

Fine-tuning

To fine-tune the model, use the following command:

python3 run.py --config config/finetune/{task}_config.yml

Evaluation

Download all checkpoints and put them under project/pretrain_weights

Checkpoint Link Note
Pretrain link
ScanRefer link Fine-tuned ScanRefer from pre-trained checkpoint.
ScanQA link Fine-tined ScanQA from pre-trained checkpoint.
Sr3D link Fine-tuned Sr3D from pre-trained checkpoint.
Nr3D link Fine-tuned Nr3D from pre-trained checkpoint.
Scan2Cap link Fine-tuned Scan2Cap from pre-trained checkpoint.

Run

To run the model, use the following command, task includes scanrefer, scanqa, sr3d, nr3d, and scan2cap.

python3 run.py --config config/eval/{task}_config.yml

Acknowledgement

We would like to thank the authors of Vil3dref, 3D-Vista, 3D-VLP and for their open-source release.

Citation:

@inproceedings{3DSyn,
  title     = {3D Vision and Language Pretraining with Large-Scale Synthetic Data},
  author    = {Dejie Yang, Zhu Xu, Wentao Mo, Qingchao Chen, Siyuan Huang, Yang Liu},
  booktitle = {Proceedings of the Thirty-Second International Joint Conference on
               Artificial Intelligence, {IJCAI-24}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  year      = {2024},
}

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published