Issues
- 1
DeDoDe
#74 opened by terrance-wang - 0
The Required Datasets for Training.
#88 opened by zhenpengju - 0
Performance
#87 opened by junwei0812 - 0
the two places name the gamma differently?
#86 opened by Devoe-97 - 1
simpler GT line matching
#83 opened by Golbstein - 3
gluestick training time
#81 opened by fatih4d - 2
question about homography transformations
#66 opened by noahzn - 0
what is add_scale_ori and how to use it?
#78 opened by noahzn - 0
- 0
mixed precision training
#76 opened by Golbstein - 0
memory not release
#75 opened by junwei0812 - 2
How to train lightglue with my extractor?
#71 opened by Adolfhill - 1
Why duplicating the log_assignment calculation?
#65 opened by 013292 - 1
- 1
pycolmap error
#73 opened by zj12345648 - 6
Training SuperGlue on custom data
#40 opened by Shuhul24 - 1
question about the relative_pose_error
#42 opened by Master-cai - 1
train => disk +lg
#72 opened by vampire1204 - 4
Training on my own dataset
#34 opened by CalleRC - 4
- 5
- 1
stuck at the "Creating dataset ImagePairs" when evaluate lightglue on megadepth
#70 opened by Master-cai - 0
- 2
- 0
- 3
- 2
Mismatched model shape upon MegaDepth finetuning
#64 opened by mattiasmar - 1
Questions about SIFT+LightGlue
#63 opened by Uchan1996 - 1
Struture of the MegaDepth dataset
#60 opened by QuangVinh1280 - 3
About train
#28 opened by liutao23 - 0
KeyError: 'loss/total'
#62 opened by QuangVinh1280 - 0
Pre-training Error
#61 opened by QuangVinh1280 - 1
Possible typo?
#44 opened by AlbertoJaenal - 0
error when installing
#57 opened by 2659494286 - 3
KeyError: 'loss/total'
#47 opened by flashcolab - 0
Homography pretraining not converging
#53 opened by ikaftan - 0
Trained Weights
#54 opened by flashcolab - 4
Possibly faster training
#39 opened by ducha-aiki - 3
Eval on WxBS
#35 opened by ducha-aiki - 1
Questions about training data
#33 opened by ACuOoOoO - 2
- 0
Discrepancy in training config w.r.t. paper
#43 opened by ducha-aiki - 0
I have a problem after training is complete
#41 opened by qiao89757 - 1
About training LightGlue in two stages
#36 opened by EATMustard - 1
Issues with lrscheduler
#31 opened by AnasAlsharo - 5
torch.multiprocessing.spawn.ProcessExitedException: process 3 terminated with signal SIGKILL
#19 opened by GONGXI1994 - 5
Performance differences between 2 versions of SuperPoint with pretrained weights
#23 opened by nobobobo - 3
- 2
Homography disk requirement
#14 opened by sbucaille - 4
About installation
#16 opened by PeeKaBo0L