GithubHelp home page GithubHelp logo

Comments (2)

kevinlin311tw avatar kevinlin311tw commented on July 17, 2024

Thank you for your interests in our work!

We were mainly using the first strategy (similar to GraphCMR) for both human mesh and hand mesh estimation tasks.

We later explored the second strategy (w/ conv1d layer) and see if we can improve further. It seems like it can slightly improves human mesh estimation about ~0.5 PA-MPJPE, but no improvement for hand mesh estimation.

We think these probably suggest that fine-grained local visual details are important to our tasks. This motivates us to explore better strategy with grid features (i.e., add grid features as input tokens and allow interactions between all the joints, vertices and grid features). We presented more details in our recent paper Mesh Graphormer.

from meshtransformer.

siyuzou avatar siyuzou commented on July 17, 2024

Thanks for the quick reply!

Interesting! I'll read the new paper later :)

from meshtransformer.

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.