QVPR/Patch-NetVLAD issues and pull requests

#91 - Add links to the added model on HuggingFace that is trained on Google Landmarks

Pull Request - State: closed - Opened by Somayeh-h 6 months ago

#90 - Unable to download pretrained models

Issue - State: closed - Opened by lijincheng2020 11 months ago - 1 comment

#89 - Large time taken for feature matching between query and database images for pitts30k

Issue - State: open - Opened by ur10 11 months ago - 5 comments

#88 - Tokyo24/7

Issue - State: closed - Opened by LKELN 12 months ago - 1 comment

#87 - pittsburgh_WPCA4096

Issue - State: closed - Opened by whuzs about 1 year ago - 1 comment

#86 - patchnetvlad recall

Issue - State: closed - Opened by whuzs about 1 year ago - 1 comment

#85 - Ambiguity in the reported results

Issue - State: open - Opened by Anuradha-Uggi about 1 year ago

#84 - How recalls on MSLS, Robotcar seasons, and CMU seasons are produced?

Issue - State: closed - Opened by Anuradha-Uggi over 1 year ago - 6 comments

#83 - questions about grouth.npz

Issue - State: closed - Opened by WentangChen over 1 year ago - 2 comments

#82 - Cross-match between patches

Issue - State: closed - Opened by euncheolChoi over 1 year ago - 2 comments

#81 - Qustion about the function get_integral_feature(feat_in)

Issue - State: closed - Opened by euncheolChoi over 1 year ago - 2 comments

#80 - how to make a custom dataset?

Issue - State: closed - Opened by WentangChen over 1 year ago - 1 comment

#79 - How about the training time-consuming of hard samples mining module?

Issue - State: closed - Opened by whu-lyh over 1 year ago - 3 comments

#78 - Add image crop transform

Pull Request - State: open - Opened by f-fl0 over 1 year ago

#77 - Support MPS

Issue - State: open - Opened by Tobias-Fischer over 1 year ago
Labels: enhancement

#76 - Remove the WPCA layer before train

Issue - State: closed - Opened by longxiangjingyun over 1 year ago - 4 comments

#75 - problem on training

Issue - State: closed - Opened by bestwo01 over 1 year ago - 6 comments

#74 - Loss function

Issue - State: closed - Opened by bestwo01 over 1 year ago - 1 comment

#73 - Fix bounds on faiss search to never go out of bounds

Pull Request - State: closed - Opened by doraemon96 over 1 year ago - 1 comment

#72 - Small issue on feature_match

Issue - State: closed - Opened by doraemon96 over 1 year ago - 1 comment

#71 - How to get the value of Extended CMU Seasons?

Issue - State: closed - Opened by MAX-OTW over 1 year ago - 5 comments

#70 - How to arrange the mapillary dataset when training the code

Issue - State: closed - Opened by bestwo01 over 1 year ago - 6 comments

#69 - cannot find the datasets...

Issue - State: closed - Opened by Raywoo1 over 1 year ago - 3 comments

#68 - Groundtruths and details of the datasets used

Issue - State: closed - Opened by Anuradha-Uggi over 1 year ago - 1 comment

#67 - Couldn't reproduce results in the paper

Issue - State: closed - Opened by silaopi almost 2 years ago

#66 - Why choose homography?

Issue - State: closed - Opened by setchin almost 2 years ago - 5 comments

#65 - What is MSLS Challenge dataset?

Issue - State: closed - Opened by kaiyi98 almost 2 years ago - 1 comment

#64 - Question about “posDistThr” of the Nordland dataset

Issue - State: closed - Opened by Lu-Feng about 2 years ago - 4 comments

#63 - Why do feat_regions divide by (patch_size ** 2)？

Issue - State: closed - Opened by libenchong about 2 years ago - 2 comments

#62 - Testing pretrained models on Pitts30k

Issue - State: closed - Opened by UsmanMaqbool about 2 years ago - 2 comments

#61 - download datasets

Issue - State: closed - Opened by libenchong about 2 years ago - 1 comment

#60 - Reproduce results of NetVLAD on RobotCar Seasons v2

Issue - State: closed - Opened by MAX-OTW about 2 years ago - 5 comments

#59 - Msls processing

Issue - State: closed - Opened by cazhang over 2 years ago - 2 comments

#58 - The training process

Issue - State: closed - Opened by Orbis36 over 2 years ago - 1 comment

#57 - Downloading Nordland dataset

Issue - State: closed - Opened by gmberton over 2 years ago - 7 comments

#56 - Shape of Local features

Issue - State: closed - Opened by wjun0830 over 2 years ago - 1 comment

#55 - About “out of memory” question

Issue - State: closed - Opened by buulzj over 2 years ago - 2 comments

#54 - question about training loss

Issue - State: closed - Opened by LyyyyyyyN over 2 years ago - 5 comments

#53 - How to down NordLand test dataset

Issue - State: closed - Opened by HeartbreakSurvivor over 2 years ago - 2 comments

#52 - About the result on Mapillary

Issue - State: closed - Opened by Jeff-Zilence almost 3 years ago - 9 comments

#51 - Why only train conv5 of VGG?

Issue - State: closed - Opened by jmorlana almost 3 years ago - 1 comment

#50 - Reproduce the results of DELG in the paper

Issue - State: closed - Opened by hellocasper almost 3 years ago - 4 comments

#49 - Why filter msls for im2im use "idx[len(idx) // 2]"

Issue - State: closed - Opened by ChengFengGu almost 3 years ago - 1 comment

#48 - Why is the performance improved so much after a pca module？

Issue - State: closed - Opened by HawkYJH almost 3 years ago - 1 comment

#47 - Some questions about loss and GPU memory

Issue - State: closed - Opened by JinRanYAO almost 3 years ago - 1 comment

#46 - Incorrect Aachen Day-Night result

Issue - State: closed - Opened by HeartbreakSurvivor almost 3 years ago - 4 comments

#45 - Can't reproduce Robotcar results

Issue - State: open - Opened by RuotongWANG almost 3 years ago - 11 comments

#44 - Fixing val.py incorrect recalls issue #43

Pull Request - State: closed - Opened by StephenHausler about 3 years ago - 10 comments

#43 - validation / evaluation wrong indices

Issue - State: closed - Opened by Tarekbouamer about 3 years ago - 1 comment

#42 - For academic communication

Issue - State: closed - Opened by DonMuv about 3 years ago - 3 comments

#41 - About calculating keypoints

Issue - State: closed - Opened by ShijieXue749 about 3 years ago - 2 comments

#40 - About Tokoy dataset

Issue - State: closed - Opened by DonMuv about 3 years ago - 3 comments

#39 - Added pitts30k as an option for PCA training

Pull Request - State: closed - Opened by StephenHausler about 3 years ago - 2 comments

#38 - About RobotCar v2 results

Issue - State: closed - Opened by RuotongWANG about 3 years ago - 4 comments

#37 - Can add_pca.py be used on the pitts dataset

Issue - State: closed - Opened by DonMuv about 3 years ago - 7 comments

#36 - Store global features image-by-image

Issue - State: closed - Opened by Tobias-Fischer about 3 years ago - 2 comments

#35 - can't assign a tuple to a torch.cuda.FloatTensor

Issue - State: closed - Opened by DonMuv about 3 years ago - 1 comment

#34 - 'can only test a child process'

Issue - State: closed - Opened by DonMuv about 3 years ago - 7 comments

#33 - sorry，how i can get the dataset

Issue - State: closed - Opened by XiaozhuLove about 3 years ago - 1 comment

#32 - Indoor experiment

Issue - State: closed - Opened by oym1994 about 3 years ago - 2 comments

#31 - keywords: 'optimizer' and 'epoch'

Issue - State: closed - Opened by cxy1996 about 3 years ago - 2 comments

#30 - How to convert pose to utm

Issue - State: closed - Opened by DonMuv about 3 years ago - 4 comments

#29 - permission for loading the model [certificate verify failed]

Issue - State: closed - Opened by jaaaaaang about 3 years ago - 6 comments

#28 - PackageNotFoundError

Issue - State: closed - Opened by jaaaaaang about 3 years ago - 4 comments

#27 - how to use pool=patchnetvlad to train the network

Issue - State: closed - Opened by zhaiqx about 3 years ago - 5 comments

#26 - how to reproduce Mapillary val recall

Issue - State: closed - Opened by ZZQzzq about 3 years ago - 2 comments

#25 - reproduce Mapillary val recall

Issue - State: closed - Opened by bixiaopeng0 about 3 years ago - 3 comments

#24 - Does this training code support Pittsburgh dataset?

Issue - State: closed - Opened by heng94 over 3 years ago - 1 comment

#23 - Training Time

Issue - State: closed - Opened by michaelschleiss over 3 years ago - 1 comment

#22 - Move checkpoint loading before DataParallel

Pull Request - State: closed - Opened by michaelschleiss over 3 years ago - 2 comments

#21 - Question about Nordland dataset.

Issue - State: closed - Opened by jinyummiao over 3 years ago - 3 comments

#20 - Train add

Pull Request - State: closed - Opened by StephenHausler over 3 years ago - 3 comments

#19 - About the hard disk memory

Issue - State: closed - Opened by heng94 over 3 years ago - 3 comments

#18 - Does utmQ correspond to the query file or the index file?

Issue - State: closed - Opened by DonMuv over 3 years ago - 2 comments

#17 - training code

Issue - State: closed - Opened by yu1206 over 3 years ago - 1 comment

#16 - How to convert h5 files to npz

Issue - State: closed - Opened by DonMuv over 3 years ago - 3 comments

#15 - About reproduce the results in the paper

Issue - State: closed - Opened by CaiYingFeng over 3 years ago - 2 comments

#14 - Feature matching (two files)

Issue - State: closed - Opened by decadezhang over 3 years ago - 2 comments

#13 - Pytorch GPU Installation

Issue - State: closed - Opened by WHBSmith over 3 years ago - 1 comment

#12 - Match realtime add

Pull Request - State: closed - Opened by StephenHausler over 3 years ago - 4 comments

#11 - Changes to detect number of clusters from model file

Pull Request - State: closed - Opened by Tobias-Fischer over 3 years ago - 1 comment

GitHub / QVPR/Patch-NetVLAD issues and pull requests