百科.dev
全部条目在线工具AI 编程趋势榜开源项目技术资讯提交条目
登录
< 返回工具列表
S

suds

> 编程语言
开源

可扩展的城市动态场景

200 stars0 点赞0 次浏览
访问官网GitHub

工具介绍

可扩展的城市动态场景

SUDS: Scalable Urban Dynamic Scenes

Haithem Turki, Jason Y. Zhang, Francesco Ferroni, Deva Ramanan

Project Page / Paper

This repository contains the code needed to train SUDS models.

Citation

@misc{turki2023suds,
   title={SUDS: Scalable Urban Dynamic Scenes},
   author={Haithem Turki and Jason Y. Zhang and Francesco Ferroni and Deva Ramanan},
   year={2023},
   eprint={2303.14536},
   archivePrefix={arXiv},
   primaryClass={cs.CV}
}

Setup

bash
conda env create -f environment.yml
conda activate suds
python setup.py install

The codebase has been mainly tested against CUDA >= 11.3 and A100/A6000 GPUs. GPUs with compute capability greater or equal to 7.5 should generally work, although you may need to adjust batch sizes to fit within GPU memory constraints.

Data Preparation

KITTI

  1. Download the following from the KITTI MOT dataset:

    1. Left color images
    2. Right color images
    3. GPS/IMU data
    4. Camera calibration files
    5. Velodyne point clouds
    6. (Optional) Semantic labels
  2. Extract everything to ./data/kitti and keep the data structure

  3. Generate depth maps from the Velodyne point clouds: python scripts/create_kitti_depth_maps.py --kitti_sequence $SEQUENCE

  4. (Optional) Generate sky and static masks from semantic labels: python scripts/create_kitti_masks.py --kitti_sequence $SEQUENCE

  5. Create metadata file: python scripts/create_kitti_metadata.py --config_file scripts/configs/$CONFIG_FILE

  6. Extract DINO features:

    1. python scripts/extract_dino_features.py --metadata_path $METADATA_PATH or python -m torch.distributed.run --standalone --nnodes=1 --nproc_per_node $NUM_GPUS scripts/extract_dino_features.py --metadata_path $METADATA_PATH for multi-GPU extraction
    2. python scripts/run_pca.py --metadata_path $METADATA_PATH
  7. Extract DINO correspondences: python scripts/extract_dino_correspondences.py --metadata_path $METADATA_PATH or python -m torch.distributed.run --standalone --nnodes=1 --nproc_per_node $NUM_GPUS scripts/extract_dino_correspondences.py --metadata_path $METADATA_PATH for multi-GPU extraction

  8. (Optional) Generate feature clusters for visualization: python scripts/create_kitti_feature_clusters.py --metadata_path $METADATA_PATH --output_path $OUTPUT_PATH

VKITTI2

  1. Download the following from the VKITTI2 dataset:

    1. RGB images
    2. Depth images
    3. Camera intrinsics/extrinsics
    4. (Optional) Ground truth forward flow
    5. (Optional) Ground truth backward flow
    6. (Optional) Semantic labels
  2. Extract everything to ./data/vkitti2 and keep the data structure

  3. (Optional) Generate sky and static masks from semantic labels: python scripts/create_vkitti2_masks.py --vkitti2_path $SCENE_PATH

  4. Create metadata file: python scripts/create_vkitti2_metadata.py --config_file scripts/configs/$CONFIG_FILE

  5. Extract DINO features:

    1. python scripts/extract_dino_features.py --metadata_path $METADATA_PATH or python -m torch.distributed.run --standalone --nnodes=1 --nproc_per_node $NUM_GPUS scripts/extract_dino_features.py --metadata_path $METADATA_PATH for multi-GPU extraction
    2. python scripts/run_pca.py --metadata_path $METADATA_PATH
  6. If not using the ground truth flow provided by VKITTI2, extract DINO correspondences: python scripts/extract_dino_correspondences.py --metadata_path $METADATA_PATH or python -m torch.distributed.run --standalone --nnodes=1 --nproc_per_node $NUM_GPUS scripts/extract_dino_correspondences.py --metadata_path $METADATA_PATH for multi-GPU extraction

  7. (Optional) Generate feature clusters for visualization: python scripts/create_vkitti2_feature_clusters.py --metadata_path $METADATA_PATH --vkitti2_path $SCENE_PATH --output_path $OUTPUT_PATH

Training

python suds/train.py suds --experiment-name $EXPERIMENT_NAME --pipeline.datamanager.dataparser.metadata_path $METADATA_PATH [--pipeline.feature_clusters $FEATURE_CLUSTERS]

Evaluation

python suds/eval.py --load_config $SAVED_MODEL_PATH or python -m torch.distributed.run --standalone --nnodes=1 --nproc_per_node $NUM_GPUS suds/eval.py --load_config $SAVED_MODEL_PATH for multi-GPU evaluation

Acknowledgements

This project is built on Nerfstudio and tiny-cuda-nn. The DINO feature extraction scripts are based on ShirAmir's implementation and parts of the KITTI processing code from Neural Scene Graphs.

Issues· 14 开放

查看全部 Issues在 GitHub 打开

暂无开放 Issues,或尚未同步最近议题。

> 标签

Python

暂无评论,来聊聊你的看法吧

> 工具信息

发布日期2026年8月1日
最后更新2026年9月18日
分类编程语言
定价开源

> 相关工具

T
TypeScript
JavaScript 的超集,为前端与全栈提供静态类型
P
Python
通用编程语言,广泛用于 Web、数据与 AI
G
Go
Google 推出的简洁高效系统语言