Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

v1.1 #78

Merged
merged 43 commits into from
Jul 17, 2021
Merged

v1.1 #78

merged 43 commits into from
Jul 17, 2021

Conversation

sarlinpe
Copy link
Member

  • [BREAKING] improved structure of the SfM folders (triangulation and reconstruction), see #76
  • Support for image retrieval (NetVLAD, DIR) and more local features (SIFT, R2D2)
  • Support for more datasets: Aachen v1.1, Extended CMU Seasons, RobotCar Seasons, 4Seasons, Cambridge Landmarks, 7-Scenes
  • Simplified pipeline and API
  • Spatial matcher
  • Support for arbitrary paths of features and matches
  • Support for matching multiple feature files together

sarlinpe and others added 30 commits August 1, 2020 23:46
- fix crash when the images are a list of `str` instead of `Path`
- skip query images for which no images were retrieved
* Added SIFT features using PyCOLMAP.

* Added asserts for input image.

* Fix style

Co-authored-by: Mihai Dusmanu <mihai.dusmanu@gmail.com>
See #69
* Added NetVLAD implementation.

* No whitening layer if whiten=False.

* Host the weights on the CVG storage

* Fix style

Co-authored-by: Mihai Dusmanu <mihai.dusmanu@gmail.com>
See #68
sarlinpe and others added 13 commits June 26, 2021 19:01
Co-authored-by: mhumenbe <martin.humenberger@naverlabs.com>
* add r2d2 feature extractor

* rename required_data_keys to required_inputs

* pass mean and std as list literal to the tvf.Normalize function and store the tvf.Normalize function norm_rgb as a member function in the class. The device will be take care in tvf.functional.Normalize

* rename conf keys and resolve error&warnings based on flake8 (E402 remained)

* fix some typos. Validate with the original repo on brooklyn.png (>95% keypoints/descriptors are the same). Successfully run on CPU but the results seem not to match very well

* remove a print
* Add pipelines for Aachen, CMU, RobotCar, Cambridge, 7Scenes
* Add prepend_camera_name argument
* Minor fixes
* Add SfM correction with dense depth
* Batch conversion
* Remove notebook Aachen v1.1
…76)

* Move the COLMAP .bin files to the top level SfM directory
* Update the notebooks
* Propagate the changes to the new pipelines
* Remove the Aachen v1.1 notebook again
@sarlinpe sarlinpe merged commit 91f40bf into master Jul 17, 2021
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

6 participants