Published on Mon Sep 28 2015

Learning FRAME Models Using CNN Filters

Yang Lu, Song-Chun Zhu, Ying Nian Wu

The convolutional neural network (ConvNet or CNN) has proven to be very successful in many tasks such as those in computer vision. In this conceptual paper, we study the generative perspective of the discriminative CNN.

0
0
0
Abstract

The convolutional neural network (ConvNet or CNN) has proven to be very successful in many tasks such as those in computer vision. In this conceptual paper, we study the generative perspective of the discriminative CNN. In particular, we propose to learn the generative FRAME (Filters, Random field, And Maximum Entropy) model using the highly expressive filters pre-learned by the CNN at the convolutional layers. We show that the learning algorithm can generate realistic and rich object and texture patterns in natural scenes. We explain that each learned model corresponds to a new CNN unit at a layer above the layer of filters employed by the model. We further show that it is possible to learn a new layer of CNN units using a generative CNN model, which is a product of experts model, and the learning algorithm admits an EM interpretation with binary latent variables.

Thu Aug 31 2017
Computer Vision
Learning Inference Models for Computer Vision
Computer vision can be understood as the ability to perform inference on image data. We propose techniques for improving inference in MCMC sampling and message-passing inference. We generalize the existing use of bilateral filters and then propose new neural network architectures.
0
0
0
Thu Nov 01 2018
Neural Networks
A Bayesian Perspective of Convolutional Neural Networks through a Deconvolutional Generative Model
Deconvolutional Generative Model (DGM) is a new probabilistic generative model. The DGM uses a CNN to design the prior distribution in the model. In addition, likelihood estimation in the DGM yields training losses for CNNs.
0
0
0
Tue May 31 2016
Machine Learning
Dynamic Filter Networks
In a traditional convolutional layer, the learned filters stay fixed after training. In contrast, we introduce a new framework, the Dynamic Filter Network, where filters are generated dynamically conditioned on an input. We demonstrate the effectiveness of the dynamic filter network on the tasks of video and stereo prediction.
0
0
0
Sat Jan 04 2020
Computer Vision
FrequentNet : A New Interpretable Deep Learning Baseline for Image Classification
This paper has proposed a new baseline deep learning model of more benefits for image classification. Different from the convolutional neural network(CNN)practice where filters are trained by back propagation to represent different patterns.
6
1
2
Fri Sep 14 2018
Computer Vision
Deep CNN Frame Interpolation with Lessons Learned from Natural Language Processing
A major area of growth within deep learning has been the study and implementation of convolutional neural networks. Recent developments in fields such as Natural Language Processing are demonstrating that this paradigm may be incorrect. We present a hypothesis that provides a novel explanation for the robustness of CNN models.
0
0
0
Tue Oct 02 2018
Machine Learning
Deep Decoder: Concise Image Representations from Untrained Non-convolutional Networks
Deep neural networks have become highly effective tools for compressing images. This success can be attributed in part to their ability to represent and generate natural images well. In this paper, we propose an untrained simple image model, called the deep decoder, which is a deep neural network.
0
0
0