Skip to content

特征点 Do you need robust and fast local feature extraction? You are in the right place!

Jupyter Notebook · active 2024-04-302026-08-11 (UTC)

Complete coverage26,401 / 26,401 hourly files (100%) · 2 absent upstream2023-08-152026-08-19 (UTC)
Events
2K
Pushes
28
Pull requests
17
Issues
135
Stars
1.4K
Forks
165

Activity over time

Daily event counts in the loaded window

Line chart, 834 days from 2024-04-30 to 2026-08-11. Pushes: 28 total, peak 4 in a day. Pull requests: 17 total, peak 3 in a day. Issues: 135 total, peak 5 in a day. Comments: 278 total, peak 12 in a day. Stars: 1,378 total, peak 94 in a day.

  • Pushes
  • Pull requests
  • Issues
  • Comments
  • Stars

Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.

Top contributors

Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity

ContributorContributionsPushesPRsComments
guipotje12326379
noahzn270015
zw-92190016
Asherchi140010
acai66110110
the-cat-crying10008
noahzhy9008
IamShubhamGupto8017
longzeyilang7005
Parskatt7014
felipecadar6105
Xiaole-An6004
AI-Passionner6003
m15kh6002
Odin-byte5004
ZhouShunlong5002
hbl8434497915001
schwert265003
MingYangGeng5000
cw3145003

Recent activity

Latest issues, pull requests and releases

  • Issue comment#118wangjunmars-bit2026-02-09 14:31
    Overlap between the MegaDepth data used for training and the MegaDepth1500 data used for evaluation
  • Issue comment#94hhl1342026-01-22 12:14
    Problem when training xfeat+lightglue
  • Issue#118kengomitsui2026-01-19 02:09
    Overlap between the MegaDepth data used for training and the MegaDepth1500 data used for evaluation
  • Issue comment#118kengomitsui2026-01-13 23:59
    Overlap between the MegaDepth data used for training and the MegaDepth1500 data used for evaluation
  • Issue comment#76OUCmeenoi2025-12-11 07:16
    根据论文写配置
  • Issue comment#74jvbshsai2025-11-18 09:53
    Question about the MegaDepth dataset for training.
  • Issue comment#113jvbshsai2025-11-18 09:51
    Clarification on MegaDepth Dataset Preparation & Broken Link
  • Issue comment#4813034638222025-10-16 03:06
    training
  • Issue comment#56ZHjiuang2025-10-16 02:27
    XFeat + GIM
  • Issue#117octaAIteam2025-10-01 14:00
    Batch inference not faster than single inference
  • Issue comment#9413034638222025-09-29 09:44
    Problem when training xfeat+lightglue
  • Issue comment#9413034638222025-09-28 08:07
    Problem when training xfeat+lightglue
  • Issue#114IkariKlein2025-08-19 08:53
    About InterpolateSparse2d
  • Issue#113skbhat2025-08-14 05:32
    Clarification on MegaDepth Dataset Preparation & Broken Link
  • Issue comment#103dinglei07192025-08-09 06:14
    Megadepth dataset and training convergence
  • Issue#112jirongli2025-08-06 05:12
    Question about the smaller model
  • Pull request#109Raghav-B2025-07-10 07:27
  • Issue comment#34July09282025-07-09 02:26
    XFeat + LightGlue
  • Issue comment#94July09282025-07-07 07:55
    Problem when training xfeat+lightglue
  • Issue comment#102sr-feng2025-07-03 08:17
    how to convert lightglue to RKNN format, thank you
  • Issue#106HumanDr2025-06-23 15:25
    an error in xfeat+lg_torch_hub.ipynb
  • Issue comment#61cuixing1582025-05-26 08:34
    Can xfeat be reproduced by MATLAB?
  • Issue#104noahzhy2025-05-12 15:10
    Question about input first two layers, mean + InstanceNorm2d
  • Issue comment#57Odin-byte2025-05-12 14:02
    Eval-MegaDepth-1500
  • Issue comment#94ZhouShunlong2025-05-12 07:38
    Problem when training xfeat+lightglue

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 1,378 stars here means stars gained during the window, not the repo's star count.