Yao Yao
Hi @gxd1994, some of the codes are depended on altizure internal library so I cannot share with you. But it should be easy to implement using either ANN search or...
Hi @demonleach we are using a modified incremental SfM pipeline. I doubt that if you are using the raw OpenMVG software you cannot get high-quality results similar to our provided...
@kwea123 Thank you for your assistance!
In R-MVSNet we use **inverse depth** for large depth range reconstructions. The definition of the expectation (soft-argmin) of inverse depth is not intuitive so we did not apply this operation...
This repo separates different scenes: https://github.com/kwea123/BlendedMVS_scenes
How did you get the per image point cloud output? The Fusibile should produce only one merged point cloud.
@UannaFF the score is the view selection score, which is used to indicate the match quality between the two views. You can refer to the MVSNet paper for how to...
Hi @UannaFF I just upload a script to convert colmap sfm result to mvsnet input. Could you try first convert the openmvg sfm to colmap and then to mvsnet?
Which dataset are you testing? For DTU dataset with input resolutions of 1600 x 1184, the depth map res is 400 x 296 and you don't need to upsample the...
There might be some problems using python3. Can python2 successfully pass this part? You can try using cv2.imread() to replace scipy.misc.imread()