Git Product home page Git Product logo

gaussian_heart's Introduction

Introduction of Gaussian Function into Rendering

1. Gaussian Splatting

image

1.1. Real time training and rendering

https://drive.google.com/file/d/177gq0HyPMX5qBCGiX3VMURhziZ0GAJFP/view?usp=sharing

1.2. Approximating Image with 2D Gaussian

https://drive.google.com/file/d/1J28Y-mxXk8i5E9HhaplPZKQkSwkhN4qu/view?usp=drive_link

1.3. Comparison of Splat Numbering

1.3.1. 100 Splats

https://drive.google.com/file/d/1ltsq_9Ri469USjnBw5opx6TbHEOmlDeD/view?usp=sharing

1.3.2. 500 Splats

https://drive.google.com/file/d/1h33vSN7GYtXxzBZ1CLDf0-YuNkqcSjWk/view?usp=sharing

1.4. Approximating Image with 3D Gaussian

https://drive.google.com/file/d/1DqhDCL6DswitASSbG1KIPAgEGIU2LqUW/view?usp=sharing

1.5. Approximating 3D Point Cloud with 3D Gaussians

1.5.1. Epochs: 1000, Points: 50000 , Splats: 200

https://drive.google.com/file/d/1vp0NTCzieo5Lzrpdz5LEwTN2Awl84QVi/view?usp=sharing

1.5.2. Epochs: 1000, Points: 50000, Splats: 500

https://drive.google.com/file/d/1YO2t6D0MmVpA04h4do4FzLXZQKWmOPki/view?usp=sharing

1.6. Approximating 3D Point Cloud with 3D Gaussians and then Projecting Points and Gaussians to 2D plane

1.6.1. Reference Rendering

https://drive.google.com/file/d/1EoyGrwSLRJECsmUyEd3rhsog5ANs1uGK/view?usp=drive_link

1.6.2. Gaussian based Rendering

https://drive.google.com/file/d/1UbxR3GOjFkGBz4lmioB5B7xeC_8qiweI/view?usp=drive_link

1.7. Effect of Hyperparameters

https://drive.google.com/file/d/1vtAheuQ_iOmOi7uX0aR76Ge3lYApfCbQ/view?usp=sharing

1.8. Comparison of Output

1.8.1. Lego Bulldozer: Train vs Test

Training performance: https://drive.google.com/file/d/15bbiPrhZRFZPjEjRU7S9Ikkxx2UBTzyO/view?usp=sharing

Test Performance: https://drive.google.com/file/d/1QtYhsZzvg10HwDZLKvKwvJr6k1fUXXVn/view?usp=drive_link

1.8.2. Train: Train vs Test

Training performance: https://drive.google.com/file/d/1TisU1tV9IhhbO3ME9ovoQTc4ZBAWrHND/view?usp=drive_link

Test performance: https://drive.google.com/file/d/1V1n6BteJ2oFTiEqFfM23wz_luNuqjBGh/view?usp=drive_link

2. DreamGaussian

image

2.1. Mapping Camera Pose to RGB Image as Backbone of Gaussian Splatting

https://drive.google.com/file/d/1LKPbh0ZxWrzt2fgGq5xURzwVWgrVdVeS/view?usp=drive_link

https://drive.google.com/file/d/1cG7L0fI8KEgUHvixH7AfiFguqSopG9lG/view?usp=drive_link

2.2. Comparison of Stage-1 and Stage-2 Output

https://drive.google.com/file/d/1ytsvpKLIwV1tlmiaARKFr7ilmx5aDHYR/view?usp=sharing

https://drive.google.com/file/d/1WL7-rR662ktqyGrrSFzaEqPIHjwTrb6b/view?usp=sharing

3. Future Works

-- Nerf without parameters

desk_img

bulldozer_img

-- Neuralangelo (its individual componets have been worked upon)

-- Improving point cloud representing with Zero: 1-2-3

4. Reference Code

2D Experiments 3D Experiments (Gaussian Splatting) 3D Experiments (DreamGaussian)

gaussian_heart's People

Contributors

superdianuj avatar

Watchers

 avatar

Recommend Projects

  • React photo React

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

  • Vue.js photo Vue.js

    ๐Ÿ–– 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. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

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 โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.