search

QVPR / VPRSNN

Code for our IEEE RAL + ICRA 2022 paper "Spiking Neural Networks for Visual Place Recognition via Weighted Neuronal Assignments", and our ICRA2023 paper "Ensembles of Compact, Region-specific & Regularized Spiking Neural Networks for Scalable Place Recognition"
MIT License
52 stars 6 forks source link

Could you share the learned weights for the Oxford RobotCar data? #24

Closed berthaSZ closed 3 months ago

berthaSZ commented 6 months ago

Hello @Somayeh-h,

Could you please share the learned weights for the Oxford RobotCar dataset? I checked #4, #10, and #16. However, I could not find either the data you used or the corresponding learned weights. Therefore, I thought this would require posting a new issue.

I processed my own subset of Oxford data by following the preprocessing instructions in your paper. However, I am getting very low precision at 100% recall values.

Best, Bertha

Somayeh-h commented 3 months ago

Hi @berthaSZ,

Thanks for reaching out about this! Again, apologies for the delayed reply.

As you raised a similar issue in the VPRTempo repo [https://github.com/QVPR/VPRTempo/issues/15], Adam has computed the corresponding image names in the original Oxford RobotCar dataset. I have also added the .csv files for the images I used in this repo in the ./dataset_imagenames folder. Please try again training the model, and reach out if any issues.

Best, Somayeh