Pengyu Zhang

I downloaded the dataset via http://human-pose.mpi-inf.mpg.de/#download, and the extension of annotation file is '.mat'. Could you point out the relative path of 'valid.json' Since I did not test on pose...

I found that the error stems from the cupy code 'masked_conv' and it does not always appear. It seems strange. I have no idea for debugging it. Have you met...

Thanks for your attention. The dataset is available at https://zhang-pengyu.github.io/DUT-VTUAV/

Sorry, we cannot provide a full dataset link because of file size limitation. For convenience, we will provide a bash/python file for downloading the full dataset automatically. Thanks for your...

对的

第一句没理解是什么意思（pedestrian_085共945张图片，95个标注），pedestrian_057我这边显示有4039张图片，404行标注

@LY-1 抱歉，这个地方是我写错了，原图片大小是640*512，原图使用PIL库的resize函数（nearest neighbour插值）进行resize，配准，裁剪，得到最终RGB-T图像对

配准是通过手动在两张图上标出特征点对，计算仿射变换矩阵，将一张图warp到另一张图的坐标系下，实现配准，最后将两张图的共同区域裁剪下来，resize到同一分辨率（保证长宽比不变），如果需要可以share给你一份我用的代码

抱歉回复晚了，我上传到网盘（链接: https://pan.baidu.com/s/10uHcGThutMmaI2SpcZKC8Q?pwd=98bf 提取码: 98bf）了，可以自取一下

Sorry for late response. The code cannot conduct multi-step training automatically. You can manually set the corresponding datasets and updated layers in train_ADRNet.py. I will also refine the code to...