Continuous Conditional Random Fields for Efficient Regression in Large Fully Connected Graphs

When used for structured regression, powerful Conditional Random Fields (CRFs) are typically restricted to modeling effects of interactions among examples in local neighborhoods. Using more expressive representation would result in dense graphs, making these methods impractical for large-scale applications. To address this issue, we propose an effective CRF model with linear scale-up properties regarding approximate learning and inference for structured regression on large, fully connected graphs. The proposed method is validated on real-world large-scale problems of image de-noising and remote sensing. In conducted experiments, we demonstrated that dense connectivity provides an improvement in prediction accuracy. Inference time of less than ten seconds on graphs with millions of nodes and trillions of edges makes the proposed model an attractive tool for large-scale, structured regression problems.

• Publication year

  2013

• Venue

  AAAI Conference on Artificial Intelligence

• Publication date

  2013-06-30

• Fields of study

  Computer Science, Environmental Science

• Identifiers
  DOI 10.1609/aaai.v27i1.8685
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Pattern Recognition And Machine Learning

  2016cited by this paper
• Long-Term Occupancy Analysis Using Graph-Based Optimisation in Thermal Imagery

  2013cited by this paper
• Efficient inference for fully-connected CRFs with stationarity

  2012cited by this paper
• Filter-Based Mean-Field Inference for Random Fields with Higher-Order Terms and Product Label-Spaces

  2012cited by this paper
• Patch reprojections for Non-Local methods

  2012cited by this paper
• Mean Field for Continuous High-Order MRFs

  2012cited by this paper
• Non-Local Means Denoising

  2011cited by this paper
• Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials

  2011influential reference
• Nonlocal Means With Dimensionality Reduction and SURE-Based Parameter Selection

  2011influential reference
• Correcting evaluation bias of relational classifiers with network cross validation

  2010cited by this paper
• Fast High‐Dimensional Filtering Using the Permutohedral Lattice

  2010cited by this paper
• (RF)^2 - Random Forest Random Field

  2010cited by this paper
• Continuous Conditional Random Fields for Regression in Remote Sensing

  2010cited by this paper
• Probabilistic Graphical Models: Principles and Techniques - Adaptive Computation and Machine Learning

  2009cited by this paper
• A social recommendation framework based on multi-scale continuous conditional random fields

  2009cited by this paper
• Gaussian KD-trees for fast high-dimensional filtering

  2009cited by this paper
• Principal Neighborhood Dictionaries for Nonlocal Means Image Denoising

  2009cited by this paper
• Fields of Experts

  2009cited by this paper
• Discriminative Learning for Dynamic State Prediction

  2009cited by this paper
• Random Field Model for Integration of Local Information and Global Information

  2008cited by this paper
• Global Ranking Using Continuous Conditional Random Fields

  2008influential reference
• Detecting Carried Objects in Short Video Sequences

  2008cited by this paper
• Learning Gaussian Conditional Random Fields for Low-Level Vision

  2007cited by this paper
• Convergent Tree-Reweighted Message Passing for Energy Minimization

  2006cited by this paper
• A Fast Approximation of the Bilateral Filter Using a Signal Processing Approach

  2006cited by this paper
• Pattern Recognition and Machine Learning

  2006cited by this paper
• Accelerated training of conditional random fields with stochastic gradient methods

  2006cited by this paper
• The MODIS Aerosol Algorithm, Products, and Validation

  2005cited by this paper
• Fields of Experts: a framework for learning image priors

  2005cited by this paper
• Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data

  2001cited by this paper
• Radiative forcing and climate response

  1997cited by this paper
• A Markov random field model for classification of multisource satellite imagery

  1996cited by this paper

• Survey of Deep Learning Techniques for High-Fidelity Material Representation and Generation

  2025cites this paper
• Enhancing Visual Re-Ranking Through Denoising Nearest Neighbor Graph via Continuous CRF

  2024cites this paper
• Ice Thickness From Deep Learning and Conditional Random Fields: Application to Ice-Penetrating Radar Data With Radiometric Validation

  2022cites this paper
• Deep Autoencoder for Hyperspectral Unmixing via Global-Local Smoothing

  2022influential citation
• RMS-FlowNet: Efficient and Robust Multi-Scale Scene Flow Estimation for Large-Scale Point Clouds

  2022cites this paper
• A Survey on Societal Event Forecasting with Deep Learning

  2021influential citation
• Conditional random fields as message passing mechanism in anchor-free network for multi-scale pedestrian detection

  2021cites this paper
• HCRF-Flow: Scene Flow from Point Clouds with Continuous High-order CRFs and Position-aware Flow Embedding

  2021cites this paper
• A survey of human judgement and quantitative forecasting methods

  2021cites this paper
• Deep Structured Prediction for Facial Landmark Detection

  2020cites this paper
• MetroEye: A Weather-Aware System for Real-Time Metro Passenger Flow Prediction

  2020cites this paper
• Computational modeling for visual attention analysis

  2020cites this paper
• Neural Network based Continuous Conditional Random Field for Fine-grained Crime Prediction

  2019cites this paper
• Crowd Counting With Deep Structured Scale Integration Network

  2019cites this paper
• Further results on structured regression for multi-scale networks

  2019cites this paper
• Attentional Neural Fields for Crowd Counting

  2019cites this paper
• Structured Modeling of Joint Deep Feature and Prediction Refinement for Salient Object Detection

  2019cites this paper
• Distance-To-Mean Continuous Conditional Random Fields: Case Study in Traffic Congestion

  2019cites this paper
• MVSCRF: Learning Multi-View Stereo With Conditional Random Fields

  2019cites this paper
• Structured Coupled Generative Adversarial Networks for Unsupervised Monocular Depth Estimation

  2019cites this paper
• Action Recognition in Multi-view Videos

  2018cites this paper
• Dividing and Aggregating Network for Multi-view Action Recognition

  2018cites this paper
• Learning to reconstruct shape and spatially-varying reflectance from a single image

  2018cites this paper
• Materials for Masses: SVBRDF Acquisition with a Single Mobile Phone Image

  2018cites this paper
• InversionNet: An Efficient and Accurate Data-Driven Full Waveform Inversion

  2018cites this paper
• Improving confidence while predicting trends in temporal disease networks

  2018cites this paper
• Monocular Depth Estimation Using Multi-Scale Continuous CRFs as Sequential Deep Networks

  2018cites this paper
• Efficient Message Passing Methods With Fully Connected Models for Early Vision

  2017cites this paper
• Multi-scale Continuous CRFs as Sequential Deep Networks for Monocular Depth Estimation

  2017cites this paper
• Deep Randomly-Connected Conditional Random Fields For Image Segmentation

  2017cites this paper
• Forecasting the rise and fall of volatile point-of-interests

  2017cites this paper
• Deep Gaussian Process for Crop Yield Prediction Based on Remote Sensing Data

  2017cites this paper
• Structured Regression on Multiscale Networks

  2017cites this paper
• Predicting Spatiotemporal Impacts of Weather on Power Systems Using Big Data Science

  2017cites this paper
• Depth estimation from single monocular images using deep hybrid network

  2016cites this paper
• Risk Assessment of a Transmission Line Insulation Breakdown Due to Lightning and Severe Weather

  2016cites this paper
• Image Restoration via Deep-Structured Stochastically Fully-Connected Conditional Random Fields (DSFCRFs) for Very Low-Light Conditions

  2016cites this paper
• Extending the Modelling Capacity of Gaussian Conditional Random Fields while Learning Faster

  2016cites this paper
• Structured Regression on Multilayer Networks

  2016cites this paper
• A future for learning semantic models of man-made environments

  2016cites this paper
• A fast structured regression for large networks

  2016cites this paper
• A Unified Probabilistic Computational Framework for Cross-Domain Compensated Medical Imaging

  2016cites this paper
• Uncertainty Propagation in Long-Term Structured Regression on Evolving Networks

  2016cites this paper
• Learning Depth from Single Monocular Images Using Deep Convolutional Neural Fields

  2015cites this paper
• 회귀나무 분석을 이용한 C-CRF의 특징함수 구성 방법 / Method to Construct Feature Functions of C-CRF Using Regression Tree Analysis

  2015cites this paper
• Forming A Random Field via Stochastic Cliques: From Random Graphs to Fully Connected Random Fields

  2015cites this paper
• Improved fine structure modeling via guided stochastic clique formation in fully connected conditional random fields

  2015cites this paper
• Prediction of New Customer’s Degree of Loyalty of Internet Shopping Mall Using Continuous Conditional Random Field

  2015cites this paper
• Panning for gold: using variograms to select useful connections in a temporal multigraph setting

  2014cites this paper
• Hidden Conditional Random Fields with Deep User Embeddings for Ad Targeting

  2014cites this paper
• A Deep-Structured Fully Connected Random Field Model for Structured Inference

  2014cites this paper
• Efficient Bayesian inference using fully connected conditional random fields with stochastic cliques

  2014cites this paper
• Neural Gaussian Conditional Random Fields

  2014cites this paper
• Deep convolutional neural fields for depth estimation from a single image

  2014cites this paper
• An international conference on mathematical modelling in social sciences and life sciences , statistics and complexity

  2013cites this paper