Learning a Generative Model of Images by Factoring Appearance and Shape

By Nicolas Le Roux, Nicolas Manfred Otto Heess, Jamie Shotton, John M. Winn. 2011

106 citations recorded by the source corpus.

Subfields: Cognitive Neuroscience, Computer Vision and Pattern Recognition, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cognitive Neuroscience

In Computer Vision and Pattern Recognition

In Signal Processing

Citations in the corpus

On the Binding Problem in Artificial Neural Networks (2020, methodology): “Instance slots have been used in several approaches to learning object representations, including Masked Restricted Boltzman Machines (M-RBMs; Le Roux et al., 2011), Neural ExpectationMaximization (N-EM; Greff et al., 2017), and IODINE (Greff et al., 2019).”
Towards causal generative scene models via competition of experts (2020, methodology): “We therefore consider it a more truthful approach to modelling visual scenes than, e.g., spatial mixture models, in line with Le Roux et al. (2011).”
Attend, Infer, Repeat: Fast Scene Understanding with Generative Models (2016, background): “These models range from highly speciﬁed (Milch et al., 2005; Mansinghka et al., 2013), to partially speciﬁed (Zhu & Mumford, 2006; Roux et al., 2011; Heess et al., 2011; Eslami & Williams, 2014; Tang et al., 2013; 2014), to largely unspeciﬁed (Hinton, 2002; Salakhutdinov & Hin-ton, 2009; Eslami et…”
Tagger: Deep Unsupervised Perceptual Grouping (2016, background): “Many of the works in that area [14, 29, 27, 11] build upon Restricted Boltzmann Machines.”
Neural Expectation Maximization (2017, methodology): “Masked RBMs [20] for example extend Restricted Boltzmann Machines with a latent mask inferred through Block-Gibbs sampling.”
Exploring Compositional High Order Pattern Potentials for Structured Output Learning (2013, methodology): “For example, they have been used to model object shape [2], images under occlusion [18], and noisy images [30].”
Robust Boltzmann Machines for recognition and denoising (2012, methodology): “Recently, models involving Restricted Boltzmann Machines have also been applied to image segmentation [17] and foreground-background modeling [3].”
Training restricted Boltzmann machines: An introduction (2014, background): “, to generate image textures [34, 28]), in particular from the marginal distributions of interest, see right plot of Fig.”
Learning and Selecting Features Jointly with Point-wise Gated Boltzmann Machines (2013, background): “Algorithm mnist-back-rand mnist-back-image mnist-rot-back-image mnist-rot-back-rand .39 - CAE-2 (Rifai et al., 2011) can be viewed as having a similar formulation to the masked RBM (Le Roux et al., 2011; Heess et al., 2011).”
Annealing between distributions by averaging moments (2013, background): “AIS with moment averaging performs well empirically at estimating partition functions of restricted Boltzmann machines (RBMs), which form the building blocks of many deep learning models.”
A Case for Object Compositionality in Deep Generative Models of Images (2018, methodology): “Different implementations consider RBMs (Le Roux et al., 2011), VAEs (Nash et al., 2017), or (recurrent) auto-encoders inspired by EM-like inference procedures (Greff et al., 2016; 2017) to generate these patches.”
A Deep Sum-Product Architecture for Robust Facial Attributes Analysis (2013, methodology): “The subsequent methods [21, 20, 18, 16] overcame this problem by learning noise models from training data.”
"Natural Images, Gaussian Mixtures and Dead Leaves" (2012, background): “There have been several works in this direction already [19,20,21], and we feel this may hold promise for creating links to higher level visual tasks such as segmentation, recognition and more.”
Learning generative models of mid-level structure in natural images (2012, methodology): “This work is an extension of the “Masked Restricted Boltzmann Machine” proposed by Le Roux et al. (2011) and it allows explicit reasoning about the independent shapes and relative depths of occluding objects.”
Approximation properties of DBNs with binary hidden units and real-valued visible units (2013, unknown): “However, arguably the most prominent applications in recent times involving RBMs consider models in which the visible variables are real-valued (e.g., Salakhutdinov & Hinton, 2007; Lee et al., 2009; Taylor et al., 2010; Le Roux et al., 2011).”