Wen Y.K.
Wen Y.K.
> > The reported results are by DispResNet... > > Thank you for reply. And what are the values of '--num-scales' and '--sequence-length' in your reported results? I have encountered...
And I found that I could not reproduce the results of your depth model，I trained from the beginning according to the code you gave me without any changes. What might...
I am trying again and see what happens my last result is 
> 1. The first problem may be little device difference, as you can see that the results are very similar. You can report your reproduced results in future paper for...
> 评估的话，可以参考evel的代码，或者是SuperPoint的代码 请问您目前的指标情况怎么样了，我在尝试对齐，但是效果很不好
我目前也在复现这篇论文，目前已经基本实现效果，其中参考了您的许多实现，非常感谢，多交流，嘻嘻！
> 按照设定的参数结果detection指标基本能和论文对齐，RS有0.1-0.2的差距， 但是descriptor指标HE和MS差的很多，说明descriptor没训练好，请问有什么建议吗 请问您最近的复现结果怎么样？
> > > > 按照设定的参数结果detection指标基本能和论文对齐，RS有0.1-0.2的差距， 但是descriptor指标HE和MS差的很多，说明descriptor没训练好，请问有什么建议吗 > > > > > > > > > 请问您最近的复现结果怎么样？ > > > > > > RS结果能达到0.640，LE是0.98，HE是0.76，descriptor的指标和论文依然差距很大 > > 你好，能分享一下你的测试代码吗，周末抽空重新调整了描述子监督的地方。看一下能否和你的测试方案对齐结果 https://github.com/eric-yyjau/pytorch-superpoint evaluation.py，export.py 稍作修改即可测试
> Hello. I am trying to implement your code and faced weird behavior of Attention Maps (predMaskC and predMaskN). I am using your pretrained model. I also tried torchvision 0.2.0...
> The version of python should be 2.7 > … I have used python 2.7 （a long time for config env） but it still like this oh on