Charlotte
Charlotte
调用search_statuses_high函数的时候返回 array(2) { 'error_code' => int(21409) 'error' => string(21) "illegal sid or no sid" } 这个错误，查了文档，并没有收入这个错误，不知该如何解决？ 这个sdk还会继续更新吗(⊙_⊙)
visualize的时候我额外输出了一些信息，但是deepsort和yolox的代码我没改，我是每秒track一次，视频fps是25，可以看 到以下连续两次track（frame475和frame500），第二次画出来的框其实是前一次那辆车所在的位置，请问为什么会这样呢？有没有什么方法可以修正？谢谢。  
您好，我加载了您提供的i3d features，但是发觉和原本的视频有些对不上。例如 ./save/Crime/UCF_Train_ten_crop_i3d/Normal_Videos586_x264_i3d.npy 原视频有1588帧，1588/16=99.25，上取整之后是100，但是您提供的视频维度是120（其他维度略），请问是怎么回事呢？ 我统计了训练集里所有mismatch的视频如下（第一个视频是您的维度，第二个是我计算的维度）： ``` ./save/Crime/UCF_Train_ten_crop_i3d/Normal_Videos586_x264_i3d.npy 120 100 ./save/Crime/UCF_Train_ten_crop_i3d/Normal_Videos600_x264_i3d.npy 158 132 ./save/Crime/UCF_Train_ten_crop_i3d/Normal_Videos635_x264_i3d.npy 85 71 ./save/Crime/UCF_Train_ten_crop_i3d/Normal_Videos644_x264_i3d.npy 91 76 ./save/Crime/UCF_Train_ten_crop_i3d/Normal_Videos645_x264_i3d.npy 67 56 ./save/Crime/UCF_Train_ten_crop_i3d/Normal_Videos648_x264_i3d.npy 67 56 ./save/Crime/UCF_Train_ten_crop_i3d/Normal_Videos649_x264_i3d.npy 68 56 ./save/Crime/UCF_Train_ten_crop_i3d/Normal_Videos650_x264_i3d.npy 82 69...
你好，关于UCF-Crime原数据集只提供了video级别的标注，请问frame级别的标注去哪里获取呢？我看了`list/make_gt_ucf.py`里的代码，我看读取了一些mat文件，请问这些文件去哪里下载呢？
Thank you for your great work! 我认真看了下你的论文、代码还有这个repo下的issue，和 #23 产生了相同的疑问。看了下论文中的公式和`train.py`中的[代码](https://github.com/tianyu0207/RTFM/blob/main/train.py#L75 )似乎不等价，能否进一步解释一下？另外`alpha`这个参数，请问是基于什么选的呢（i.e.保持和loss_cls相同的量级）？ 另外关于sparsity regularisation ，论文中写的似乎是L1 norm，但是[代码](https://github.com/tianyu0207/RTFM/blob/main/train.py#L11)中用的似乎是L2 norm？另外sparsity regularisation和temporal smoothness 项前的系数是如何选择的呢？还有在计算这两项loss的时候为何只使用了abn_scores而不是全部的scores呢？
I used the codes and the dataset that you have provided and trained 100 epoches. I tried to generate characters with blue eyes and blonde hair and got the following...
Got the following errors when saving the final gif results: ``` MovieWriter imagemagick unavailable; trying to use instead. Traceback (most recent call last): File "C:\ProgramData\Anaconda3\lib\site-packages\matplotlib\animation.py", line 230, in saving yield...
Hi, your results seem very good on ped2 dataset. But my results are very bad. (For other 3 datasets, I can achieve similar results as yours). Would you please share...
I try to tune the hyper parameter for whas dataset and get the following error: ``` Attaching to hyperparamsearch_hp_search_1 hp_search_1 | /usr/local/lib/python2.7/dist-packages/h5py/__init__.py:34: FutureWarning: Conversion of the second argument of issubdtype...
The hyper parameters used for the experiments are already set. May I ask how are the current hyper parameters selected? Are there any tips for selecting hyper parameters to get...