A Survey on 3D Model Generation from Images
Keywords:
3D Model, GAN, DIB-R, Image, Neural NetAbstract
3D models are used for a variety of domains including video games, movies, architecture, illustration, engineering, and commercial advertising. We have seen significant progress in 3D model generation and reconstruction in recent years. There are multiple approaches, or method to do it. We discuss about different approaches in this paper, such as 3D Generative Adversarial Network (GAN), Differentiable Interpolation-based Renderer (DIB-R), Hierarchical Surface Prediction. We also discuss advantages, and limitations of these approaches. In the end it shows the results produced by them.
References
- Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modelling, Jiajun Wu*, Chengkai Zhang*, Tianfan Xue*, William T. Freeman*, Joshua B. Tenenbaum* (2016)
- 3D Reconstruction from Single 2D Image, Department of CSE, Maharaja Institute of Technology Mysore, India, January 2016
- Image Generation with Disentangled 3D representation Jun Yan Zhu* (6 Dec 2018)
- Unsupervised learning of 3D representations from natural images, Thu Nguyen-Phuoc*(1 Oct 2019)
- Learning to predict 3D objects with an interpolation- based differentiable renderer. (2019)
- https://en.wikipedia.org/wiki/3D_rendering
- Hierarchical Surface Prediction for 3D Object Reconstruction Christian Häne, Shubham Tulsiani, Jitendra Malik (3 Apr 2017)
Downloads
Published
Issue
Section
License
Copyright (c) IJSRST
![Creative Commons License](http://i.creativecommons.org/l/by/4.0/88x31.png)
This work is licensed under a Creative Commons Attribution 4.0 International License.