Thermodynamics of natural images
arXiv:0806.2694 · doi:10.1103/PhysRevLett.110.018701
Abstract
The scale invariance of natural images suggests an analogy to the statistical mechanics of physical systems at a critical point. Here we examine the distribution of pixels in small image patches and show how to construct the corresponding thermodynamics. We find evidence for criticality in a diverging specific heat, which corresponds to large fluctuations in how "surprising" we find individual images, and in the quantitative form of the entropy vs. energy. The energy landscape derived from our thermodynamic framework identifies special image configurations that have intrinsic error correcting properties, and neurons which could detect these features have a strong resemblance to the cells found in primary visual cortex.
References in corpus (2)
Cited by in corpus (32)
- Are biological systems poised at criticality?
- Diversity in Machine Learning
- Searching for collective behavior in a network of real neurons
- Inverse statistical problems: from the inverse Ising problem to data science
- Natural images from the birthplace of the human eye
- Dynamical criticality in the collective activity of a population of retinal neurons
- Multi-scale structural complexity of natural patterns
- Perspectives on theory at the interface of physics and biology
- Signatures of criticality arise in simple neural population models with correlations
- Beauty and structural complexity
- Input nonlinearities can shape beyond-pairwise correlations and improve information transmission by neural populations
- Rényi entropy, abundance distribution and the equivalence of ensembles
- Hamiltonian Annealed Importance Sampling for partition function estimation
- A statistical physics perspective on criticality in financial markets
- Pairwise maximum-entropy models and their Glauber dynamics: bimodality, bistability, non-ergodicity problems, and their elimination via inhibition
- Biochemical feedback and its application to immune cells II: dynamics and critical slowing down
- Systems poised to criticality through Pareto selective forces
- Dead leaves and the dirty ground: low-level image statistics in transmissive and occlusive imaging environments
- An Information Theoretic Measure for Robot Expressivity
- Improving predictability of time series using maximum entropy methods
- Optimal encoding in stochastic latent-variable Models
- Boltzmann-Machine Learning of Prior Distributions of Binarized Natural Images
- The statistical mechanics of Twitter
- Extrinsic vs Intrinsic Criticality in Systems with Many Components
- Semiparametric energy-based probabilistic models
- On the typical properties of inverse problems in statistical mechanics
- Kauffman cellular automata on quasicrystal topology
- Active image restoration
- The Wilson Machine for Image Modeling
- Multiscale Relevance of Natural Images
- Universal Scale Laws for Colors and Patterns in Imagery
- Thermodynamics of Information Retrieval