1 citations · 2 across the 10 of their papers we have counts for
7 papers · 1 filter
Exploring Image Generation via Mutually Exclusive Probability Spaces and Local Correlation Hypothesis
Chenqiu Zhao, Anup Basu
A common assumption in probabilistic generative models for image generation is that learning the global data distribution suffices to generate novel images via sampling. We investi…
Accelerating Inference of Networks in the Frequency Domain
Chenqiu Zhao, Guanfang Dong, Anup Basu
It has been demonstrated that networks' parameters can be significantly reduced in the frequency domain with a very small decrease in accuracy. However, given the cost of frequency…
Affine-Transformation-Invariant Image Classification by Differentiable Arithmetic Distribution Module
Zijie Tan, Guanfang Dong, Chenqiu Zhao +1
Although Convolutional Neural Networks (CNNs) have achieved promising results in image classification, they still are vulnerable to affine transformations including rotation, trans…
Is Deep Learning Network Necessary for Image Generation?
Chenqiu Zhao, Guanfang Dong, Anup Basu
Recently, images are considered samples from a high-dimensional distribution, and deep learning has become almost synonymous with image generation. However, is a deep learning netw…
Frequency Regularization: Restricting Information Redundancy of Convolutional Neural Networks
Chenqiu Zhao, Guanfang Dong, Shupei Zhang +2
Convolutional neural networks have demonstrated impressive results in many computer vision tasks. However, the increasing size of these networks raises concerns about the informati…
Learning Temporal Distribution and Spatial Correlation Towards Universal Moving Object Segmentation
Guanfang Dong, Chenqiu Zhao, Xichen Pan +1
The goal of moving object segmentation is separating moving objects from stationary backgrounds in videos. One major challenge in this problem is how to develop a universal model f…
Universal Background Subtraction based on Arithmetic Distribution Neural Network
Chenqiu Zhao, Kangkang Hu, Anup Basu
We propose a universal background subtraction framework based on the Arithmetic Distribution Neural Network (ADNN) for learning the distributions of temporal pixels. In our ADNN mo…