José M. Fácil

Hi!, For the moment I am involved in other projects and I did not have the time to adapt the code for more updated versions of tensorflow or pytorch. However...

Hi @jsdd25, For training you can find the solver and network defined in the repository in [models/not_fine_tuned](https://github.com/jmfacil/single-view-place-recognition/tree/master/models/not_fine_tuned) or [models/fine_tuned](https://github.com/jmfacil/single-view-place-recognition/tree/master/models/fine_tuned) for the fine tuned version. The triplets are created according to...

Sorry for not being more specific. You have many more negative pairs than positives as you can imagine. We found that the network converge faster if we don't select negative...

Yes. You can also include another 19 images by mirroring them and if you do RGB random data augmentation you would not need to exclude one frame in the same...

Hi, Small edit on @valavanisleonidas comment. ```python if ratio < thresh: if closest_image_retrieved in window5frames: tp += 1 else: fp += 1 else: if closest_image_retrieved in window5frames: fn += 1...

I could not reproduce the problem. I would suggest to do as @jsdd25 proposes and change OpenCV version to 2.4.11. We will write you back if we find out what...

Hi, As you say the angle alpha is the angle between between the rays X-KF and X-CF. That threshold on alpha should not vary much between datasets, you can leave...

> > > Hi, > > > As you say the angle alpha is the angle between between the rays X-KF and X-CF. That threshold on alpha should not vary...

> > Hi, > > As you say the angle alpha is the angle between between the rays X-KF and X-CF. That threshold on alpha should not vary much between...

If the parallax angle is less than 30 (α