Vision Transformers for Dense Prediction——代码实践

相关申明

源代码地址

>       1. @article{Ranftl2021,

>  
>       2.      author    = {Ren\'{e} Ranftl and Alexey Bochkovskiy and Vladlen Koltun},
>  
>       3.      title     = {Vision Transformers for Dense Prediction},
>  
>       4.      journal   = {ArXiv preprint},
>  
>       5.      year      = {2021},
>  
>       6. }
>  
>       7.
>  
>  
> ```
>
>

复制代码

  1. @article{Ranftl2020,

复制代码

  2.    author    = \{Ren\'{e} Ranftl, Katrin Lasinger, David Hafner, Konrad Schindler, Vladlen Koltun\},

  3.    title     = Focusing on Accurate Monocular Depth Estimation: A Comprehensive Approach to Combining Datasets for Zero-shot Cross-dataset Transfer,

  4.    journal   = IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI),

  5.    year      = Year,

  6.

小赵在ubuntu18.04系统下进行论文的代码实践，简单流程如下：

1.下载github项目到本地

    git clone https://github.com/isl-org/DPT.git
    
    python

按照提示获取Monodepth与Segmentation模型参数文件，并将其放入DPT项目的weights文件夹内

3. 通过anaconda3创建相关环境

    conda create -n DPT python=3.7
    
    python

    source activate DPT

    cd DPT

 conda install --yes --file requirements.txt

    
 或
    
 pip3 install -r requirements.txt

Python 3.7, PyTorch 1.8.0, OpenCV 4.5.1, and timm 0.4.5 均通过requirements.txt完成安装

4.

• 将image（单张或多张）放置
• 执行单目深度估计模型

    python run_monodepth.py

• 执行语义分割模型

    python run_segmentation.py

5.结果

结果分别放置在对应的output_monodepth 和 output_semseg

6.拓展

将Monodepth微调后作用在KITTI，下载对应.pt文件，放置到weights

将Monodepth微调后作用在NYUv2，下载对应.pt文件，放置到weights

执行

 python run_monodepth -t [dpt_hybrid_kitti|dpt_hybrid_nyu]

    
  
    
 python run_monodepth.pt -t dpt_hybrid_kitti
    
  
    
 python run_monodepth.pt -t dpt_hybrid_nyu

kitti

nyu