xuanlanxingkongxia

Results 7 comments of xuanlanxingkongxia

Hi, Great work and excellent results. It seems that you release the KITTI results for noc parts? So, does this mean that the results in your article about KITTI are...

Thank you very much! You provide all of the information I need !

The "if_local" choice try to pre-decide the matched pairs from multiple possible ones, and reduce the space cost. It just can provide a trade-off beween time and space.

But our pre-decision algorithm here is not efficient enough and seems to crop some of useful pairs incorrectly, maybe you can write a more efficient one?

May be you don't follow our README to use our code?

I meet similar problem. For the image DSC_0410.JPG and DSC_0411.JPG, it seems that in the case of closing the "flash-attention", Lightglue can be only ~30% faster than superglue? I do...

![LightGlue_benchmark_(cuda)](https://github.com/cvg/LightGlue/assets/43473758/59a1a6c9-6bb9-448a-93e0-240ae9409045) Oh, I find my problem. But with benchmark code, it still seems something wrong? Without falsh-attention and compile, the speed will be similar to superglue?Am I still do something...