SINet-伪装物体检测(COD) 配置过程

https://github.com/DengPingFan/SINet

1. Configuring your environment (Prerequisites):

   Note that SINet is only tested on Ubuntu OS with the following environments. It may work on other operating systems as well but we do not guarantee that it will.

   • Creating a virtual environment in terminal: conda create -n SINet python=3.6.
     conda create --name SINet-pytorch pyhton=3.7.4   (这是我的环境)

   • Installing necessary packages: pip install -r requirements.txt.

   • (Optional: only for training) Installing NVIDIA-Apex for accelerate training process with mixed precision. (Instructions) (Under CUDA-10.0 and Cudnn-7.4).

2. Downloading Training and Testing Sets:
    downloading NEW testing dataset (COD10K-test + CAMO-test + CHAMELEON) and move it into ./Dataset/TestDataset/, which can be found in this Google Drive link or Baidu Pan link with the fetch code: z83z.
    测试集(以COD10K为例)路径:   ./Dataset/TestDataset/COD10K

    

3.   download NEW training dataset (COD10K-train) which can be found in this Google Drive link or Baidu Pan link with the fetch code:djq2.  Please refer to our original paper for other training data.

4. Testing Configuration:
   After you download all the pre-trained model and testing data,  just run MyTest.py to generate the final prediction map: replace your trained model directory (--model_path) and assign your the save directory of the inferred mask (--test_save)
   Note that we re-trained our model (marked as $\diamondsuit$ in the following figure) equipped with mixed training strategy of Apex lib (mode=O1) and get better performance in 40 epoch. Here we provide a new pre-trained model (Baidu Drive [fetch code:2pp2]/Google Drive) here. Later, We will try different backbones based SINet to improve performance and provide more comprehensive comparison.

   下载所有预先训练的模型和测试数据后，只需运行MyTest.py要生成最终的预测图：替换训练的模型目录（--model_path），并分配推断掩码的保存目录（--test_save）

   请注意，我们重新训练了我们的模型（下图中标记为$\diamondsuit$），采用Apex-lib（模式=O1）的混合训练策略，在40个历元中获得了更好的性能。这里我们提供了一个新的预训练模型（Baidu Drive[fetch代码：2pp2]/谷歌硬盘）这里。稍后，我们将尝试基于不同的主干网来提高性能，并提供更全面的比较。