place-recognition-db

place-recognition-db copied to clipboard

https://github.com/prime-slam/place-recognition-db/blob/43e4aa21801b55d9a016db79b108c4c68308c879/vprdb/vpr_systems/cos_place/cos_place.py#L49 ```RuntimeError: Attempting to deserialize object on a CUDA device but torch.cuda.is_available() is False. If you are running on a CPU-only machine, please use torch.load with map_location=torch.device('cpu') to map your...

ksenmel

https://github.com/prime-slam/place-recognition-db/blob/43e4aa21801b55d9a016db79b108c4c68308c879/vprdb/global_localization/global_localization.py#L77-L80 Currently, local features are computed all at once, which can cause problems if the dataset is large enough

vnmsklnk

Match two databases inefficiency

1

https://github.com/prime-slam/place-recognition-db/blob/43e4aa21801b55d9a016db79b108c4c68308c879/vprdb/core/utils.py#L66 At the moment, the function checks the coverage of each frame with each

vnmsklnk