Git Product home page Git Product logo

Comments (6)

kwea123 avatar kwea123 commented on July 18, 2024

I feel sorry, my evaluation is done in resolution 1152x864 because of insufficient memory. Is your metrics done in resolution 1600x1184 then?

from mvsnet_pytorch.

whubaichuan avatar whubaichuan commented on July 18, 2024

@kwea123 The code here uses 1600x1184. Maybe there are some differences in the fusion code. The original MVSNet uses fusibile to fuse all of the depth. But the fusion code here needs to be improved.

from mvsnet_pytorch.

whubaichuan avatar whubaichuan commented on July 18, 2024

@kwea123
您好,最近遇到一个问题:关于mvsnet中D=192和D=256时,overall的值到底是多少。

第一,基于上所有published paper中,当D=256时,overall=0.462,这个问题不大。

第二,但是当D=192时,却大不一样:
a. 在这里提到overall=0.482, 但是它的这个仓库里说是0.414。
b. 在cascade mvsnet point-mvs 和若干其余论文中,overall=0.551。
c. 在mvs2unsup_mvs 中,overall=0.592.

请问当D=192时,overall到底应该如何选择呢?

from mvsnet_pytorch.

kwea123 avatar kwea123 commented on July 18, 2024

跟resolution跟D有關吧,我測的是1152x864所以比這裡寫的0.414要差,它的結果應該是1600x1184的。
據我看,cascade是抄point mvs的結果,其他論文應該也是互抄,而point mvs裡提到這個數字是1152x864x192的結果,這個應該可信。
在unsupervised的兩篇論文中,training details中有提到resolution是640x512,所以他們的結果應該是640x512x192的。

結論:看你要比較哪個resolution就用哪個數字
有些論文抄數字會抄錯,所以最好多方比較... 另外確認一下overall是不是等於(acc+comp)/2,如果不是就是抄錯了

from mvsnet_pytorch.

whubaichuan avatar whubaichuan commented on July 18, 2024

@kwea123 嗯嗯,你跑出来的0.487和point mvs的0.551应该是相同的resolution, 怎么结果会有比较大的差别呢?

from mvsnet_pytorch.

kwea123 avatar kwea123 commented on July 18, 2024

可能因为这个训练得epoch比较长吧,这里是16原文是6

from mvsnet_pytorch.

Related Issues (20)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.