DagnyT/hardnet

Issues

• indoor dataset

  #47 opened 4 years ago by DIVE128
  0

• HPatches result with matlab and python

  #36 opened 5 years ago by aIR1000
  2

• the trainning datasets

  #46 opened 4 years ago by Nestart
  0

• local variable 'out_a' referenced before assignment

  #45 opened 4 years ago by Nestart
  1

• FileNotFoundError: [Errno 2] No such file or directory: 'data/sets/liberty.zip'

  #44 opened 4 years ago by InstantWindy
  1

• Loss function

  #43 opened 4 years ago by zzxp
  0

• where is pytorch_sift ?

  #26 opened 6 years ago by yunyundong
  8

• Key Error occurred when running scripts of Hpathces

  #17 opened 6 years ago by gsygsy96
  10

• Training Set Label Formatting

  #42 opened 5 years ago by ss32
  2

• report a bug maybe

  #41 opened 5 years ago by nbl97
  0

• invalid index of a 0-dim tensor. Use tensor.item() to convert a 0-dim tensor to a Python number

  #40 opened 5 years ago by hiterjoshua
  1

• Is this loss the same with Batch Hard Loss ?

  #39 opened 5 years ago by XuyangBai
  3

• Test on yosemite dataset

  #38 opened 5 years ago by Shu-HowTing
  1

• About the measurement metric

  #29 opened 5 years ago by yunyundong
  1

• what is the version of pytorch

  #34 opened 5 years ago by andeyeluguo
  2

• The link to HardNet++ is broken

  #37 opened 5 years ago by moshes7
  1

• Key Error occurred when running bash run_me.sh

  #35 opened 5 years ago by Danee-wawawa
  1

• Verification task on HPatches

  #33 opened 6 years ago by coding456
  0

• Training epochs

  #32 opened 6 years ago by aIR1000
  2

• Code of Oxford5k experiments

  #31 opened 6 years ago by appleleaves
  2

• HardNet+ with --augmentation or --fliprot

  #30 opened 6 years ago by aIR1000
  11

• hardnet training files

  #28 opened 6 years ago by aIR1000
  2

• Has a worse performance HardNetClassicalHardNegMiningSiftInit, Compared to HardNetMultipleDatasets?

  #27 opened 6 years ago by yunyundong
  6

• RuntimeError: cuda runtime error (30) : unknown error at /pytorch/aten/src/THC/THCTensorRandom.cu:25

  #20 opened 6 years ago by dymymao
  1

• Waht's the meaning of this line?

  #25 opened 6 years ago by yunyundong
  3

• all dataset as training set

  #24 opened 6 years ago by appleleaves
  0

• performance in the Hpatches dataset on retrieval task

  #23 opened 6 years ago by appleleaves
  1

• something confused when training hardnet with hpatches

  #22 opened 6 years ago by Miranda0920
  4

• Insufficient distance margin ?

  #21 opened 6 years ago by IQ17
  2

• Training error : OOM

  #19 opened 6 years ago by Miranda0920
  1

• The Test FPR95 accuracy is always similar regard of different datasets trained ???

  #18 opened 6 years ago by Sunshine352
  2

• Could you tell me how to train on oxford5k?

  #16 opened 6 years ago by gsygsy96
  21

• use hardnet as the local feature extractor

  #14 opened 6 years ago by yhethanchen
  8

• The test code of the caffe model is wrong

  #11 opened 6 years ago by hurricane2018
  4

• The result is different from your paper

  #13 opened 6 years ago by hurricane2018
  2

• Can you explain more about those different HardNet ?

  #12 opened 6 years ago by hurricane2018
  1

• how to evaluate the Liberty using Notredame

  #10 opened 6 years ago by hurricane2018
  5

• Interrupt when running the HardNet.py sample

  #9 opened 6 years ago by empty16
  4

• Convolutional sliding window

  #7 opened 6 years ago
  0

• PS model

  #6 opened 6 years ago by bhack
  11

• Error when the number of keypoints is a multiple of the batch size (128)

  #5 opened 7 years ago by fb82
  1

• Concerning the "class TripletPhotoTour"

  #4 opened 7 years ago by johnnysclai
  1

• Sample code to generate hpatch descriptors not working.

  #1 opened 7 years ago by trueutkarsh
  1