Deep disentangled representations for volumetric reconstruction
By Edward Grant, Pushmeet Kohli, Marcel A. J. van Gerven. 2016
31 citations recorded by the source corpus.
We introduce a convolutional neural network for inferring a compact disentangled graphical description of objects from 2D images that can be used for volumetric reconstruction. The network comprises an encoder and a twin-tailed decoder. The encoder generates a disentangled graphics code. The first decoder generates a volume, and the second decoder reconstructs the input image using a novel training regime that allows the graphics code to learn a separate representation of the 3D object and a description of its lighting and pose conditions. We demonstrate this method by generating volumes and disentangled graphical descriptions from images and videos of faces and chairs.
Subfields: Computational Mechanics, Computer Vision and Pattern Recognition
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computational Mechanics
- Maxim Tatarchenko
- Andrew Markham
- Bo Yang
- Agathoniki Trigoni
- Hongkai Wen
- Sen Wang
- Thomas Brox
- Alexey Dosovitskiy
In Computer Vision and Pattern Recognition
- Andrea Vedaldi
- Andrew Markham
- Luc Van Gool
- Jitendra Malik
- Ming-Hsuan Yang
- Ruslan Salakhutdinov
- Dacheng Tao
- Thomas Brox
Citations in the corpus
- Octree Generating Networks: Efficient Convolutional Architectures for High-resolution 3D Outputs (2017, methodology): “Applications include single- and multi-view 3D object reconstruction trained in supervised [14, 5, 7, 17] and unsupervised [37, 12, 19] ways, probabilistic generative modeling of 3D shapes [36, 35, 29], semantic segmentation [2, 4] and shape deformation [39].”
- 3D Object Reconstruction from a Single Depth View with Adversarial Learning (2017, background): “In the past two years, a number of works [13] [60] [15] [21] apply such generative models to learn latent space to represent 3D object shapes, and then to solve simple discriminative tasks such as new image generation, object classification, recognition and shape retrieval.”
- 3D Object Dense Reconstruction from a Single Depth View (2018, background): “In the past few years, a number of works [18] [19] [20] [21] applied such generative models to learn latent space to represent 3D object shapes, in order to solve simple discriminative tasks such as new image generation, object classification, recognition and shape retrieval.”