colmap

colmap copied to clipboard

Add ALIKED/LightGlue feature based on libtorch

9

* Adds support for extracting ALIKED features. * Adds support for matching ALIKED features using simple dot-product based nearest neighbor search. * Adds support for matching ALIKED and SIFT using...

ahojnnes

what does "POINTS2D[] as (X, Y, POINT3D_ID)" mean?

3

Are the POINTS2D in images.txt the xy coordinates extracted during feature extraction, or are they the reprojected points after reconstruction?

bob-tinner

Mixture of known poses of varying quality

1

**Is your feature request related to a problem? Please describe.** I'm always frustrated when I want to use COLMAP in the case where I have perfect poses for some frames...

KFDK

CUDA Illegal Memory Access in patch_match_stereo with Large Images (8000x6000)

4

### **Description** When running `colmap patch_match_stereo` on high-resolution images (8000x6000), I encounter a CUDA illegal memory access error during execution. However, when I limit the `--PatchMatchStereo.max_image_size` to **7700**, the process...

beautifulv0id

Patch Match Stereo on subset of images

1

Hi, I was wondering if it was possible to only compute depth maps for only a subset of images? (While still using all images as potential source frames) I haven't...

reynoldscem-oculo

model_converter on linux fails after inserting known camera poses

4

Hello, I am currently trying to reconstruct a model from known camera poses. I achieved this on my windows machine however on ubuntu it fails with the final model conversion...

M45t3rL4

Reconstruct with very large images

1

Hello, I am wondering what the best practice is to generate accurate camera poses and sparse point cloud with large images >4K? Currently using 4K images and around 950 images,...

abrahamezzeddine

ERROR:root:Feature matching failed with code 34304. Exiting.

5

I would like to train myself to take images with 3DGS and process my images with colmap. when i run :python convert.py -s data/mydata，It throws the following error： python convert.py...

Justinitaa

Sparse reconstruction from point triangulator results using existing camera poses in an irrational result compared to mapper

11

(This issue was first posted in [glomap](https://github.com/colmap/glomap/issues/161) repo) **Describe the bug** I am using habitat simulator with the [coordinates system y-up and right-handed](https://aihabitat.org/docs/habitat-sim/coordinate-frame-tutorial.html#coordinate-frame-conventions-y-up-and-right-handed) (I tried `-QX, QW, -QZ, QY, TX,...

lemonci

Camera model does not exist for OPENCV

7

**Describe the bug** When I run `pycolmap.import_images()` with the OPENCV camera model, I get this error (`what(): Camera model does not exist`) ``` ============================================================================== Feature matching ============================================================================== I0124 10:48:02.596794 712...

michael-hoss