search

cdcseacave / openMVS

open Multi-View Stereo reconstruction library
http://cdcseacave.github.io
GNU Affero General Public License v3.0
3.18k stars 891 forks source link

Papers the Point Cloud Reconstruction Algorithm is based on #218

Open MaXL130 opened 7 years ago

MaXL130 commented 7 years ago

Hi openMVS-team,

I read through your documentation, and found that you are pointing towards the paper "PatchMatch: A Randomized Correspondence Algorithm for Structural Image Editing".

In another issue (this one) the answer references another paper, called "Accurate Multiple View 3D Reconstruction Using Patch-Based Stereo for Large-Scale Scenes".

So I just wanted to make sure those are the right sources to dig into if I want to learn more about the point cloud reconstruction algorithm? Are there more publications that are relevant?

Thanks for your help, Max

kejianwei commented 7 years ago

I have the same question. I am also wondering what reference(s) the main topics(dense reconstruction, mesh reconstruction, mesh refinement, mesh texturing) are based on.

Thank you all in advance.

Jianwei

MaXL130 commented 7 years ago

@kejianwei Have you already found sth, beside this thread?

cdcseacave commented 7 years ago

The papers listed in the wiki are the closest to the algorithms implemented in OpenMVS. However none of the implementations follow the original paper 100%, but rather they are a mixture of multiple ideas, either found in many other papers, either original.

MaXL130 commented 7 years ago

Hey, thanks for your answer! So is the paper I referenced above one of those "findings", used for point reconstruction? I just noticed, that I referenced the wrong thread in my original post, I meant this one. There is a link to a paper (see above), that the dense point reconstruction is based on. Is this the main framework for this explicit module?