Statistical Modeling and Conceptualization of Visual Patterns

By Song-Chun Zhu. 2003

136 citations recorded by the source corpus.

Subfields: 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 Computer Vision and Pattern Recognition

Citations in the corpus

On the Anatomy of MCMC-based Maximum Likelihood Learning of Energy-Based Models (2019, methodology): “[42] refers to energy-based models as descriptive models.”
Learning Descriptor Networks for 3D Shape Synthesis and Analysis (2018, methodology): “We call this type of models the descriptive models or descriptor networks following [34], because the models describe the data based on bottom-up descriptive features learned from the data.”
Divergence Triangle for Joint Training of Generator Model, Energy-Based Model, and Inferential Model (2018, methodology): “[9] called the energy-based models as descriptive models.”
Cooperative Training of Descriptor and Generator Networks (2016, methodology): “To illustrate the descriptive models and the generative models, (Zhu, 2003) used the FRAME (Filters, Random ﬁeld, And Maximum Entropy) model of (Zhu et al., 1997) as a prototype of the descriptive models, and used the sparse coding model of (Olshausen & Field, 1997) as a prototype of the generative…”
Alternating Back-Propagation for Generator Network (2016, methodology): “We model a textured motion (Wang and Zhu, 2003) or a dynamic texture (Doretto et al., 2003) by a non-linear dynamic system Yt = f(Zt;W ) + t, and Zt+1 = AZt + ηt, where we assume the latent factors follow a vector auto-regressive model, whereA is a d×dmatrix, and ηt ∼ N(0, Q) is the innovation.”
Pros and Cons of GAN Evaluation Measures (2018, background): “Another important property of natural image statistics is the non-Gaussianity [76, 106, 85].”
Unsupervised segmentation of natural images via lossy data compression (2008, background): “For a more detailed survey of these methods, the reader is referred to [42,15,8,27].”
Learning Latent Space Energy-Based Prior Model (2020, result): “Earlier papers on this strategy are [64, 19, 4, 8, 31].”
Efficient Multilevel Brain Tumor Segmentation With Integrated Bayesian Model Classification (2008, unknown): “The methods also put prior probabilities on the model variables defining what is termed a generative model [34].”