Feng Lu
Feng Lu
> By running example.py I faced with error... > > ValueError: Layer #14 (named "net_vlad_1" in the current model) was found to correspond to layer net_vlad_1 in the save file....
> Hi all, > > thanks for reporting this. I will try to figure out what went wrong. Could you please tell me which version of tensorflow you are using?...
> To solve this problem go to netvladlayer.py and comment out the following line > > `self._trainable_weights.append(self.conv.trainable_weights[0])` > > This adds the kernel weights to the model's list of trainable....
是的，384维的全局特征用于直接检索
Thanks for your interest in our work. As the SelaVPR++ paper is still under review, we currently have no plans to release its full code. However, we intend to release...
@ming-gray Hello, we have provided the raw code of deep hashing in SelaVPR++ at https://github.com/Lu-Feng/Deep_Hashing_in_SelaVPRplusplus for reference.
> > [@ming-gray](https://github.com/ming-gray) Hello, we have provided the raw code of deep hashing in SelaVPR++ at https://github.com/Lu-Feng/Deep_Hashing_in_SelaVPRplusplus for reference. > > This Link can not be seen now. This Link...
Hello, I just updated the download link. Please try again using the new code.
Yes, we recommend using the [format_mapillary.py](https://github.com/gmberton/VPR-datasets-downloader/blob/main/format_mapillary.py) in VPR-datasets-downloader repository to format MSLS.