Pixel Deconvolutional Networks

Abstract

Deconvolutional layers have been widely used in a variety of deep models for up-sampling, including encoder-decoder networks for semantic segmentation and deep generative models for unsupervised learning. One of the key limitations of deconvolutional operations is that they result in the so-called checkerboard problem. This is caused by the fact that no direct relationship exists among adjacent pixels on the output feature map. To address this problem, we propose the pixel deconvolutional layer (PixelDCL) to establish direct relationships among adjacent pixels on the up-sampled feature map. Our method is based on a fresh interpretation of the regular deconvolution operation. The resulting PixelDCL can be used to replace any deconvolutional layer in a plug-and-play manner without compromising the fully trainable capabilities of original models. The proposed PixelDCL may result in slight decrease in efficiency, but this can be overcome by an implementation trick. Experimental results on semantic segmentation demonstrate that PixelDCL can consider spatial features such as edges and shapes and yields more accurate segmentation outputs than deconvolutional layers. When used in image generation tasks, our PixelDCL can largely overcome the checkerboard problem suffered by regular deconvolution operations.

11 pages

References in corpus (11)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• Stochastic Backpropagation and Approximate Inference in Deep Generative Models
• Generative Adversarial Text to Image Synthesis
• Conditional Image Generation with PixelCNN Decoders
• DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs
• Learning Deconvolution Network for Semantic Segmentation
• MADE: Masked Autoencoder for Distribution Estimation
• Composing graphical models with neural networks for structured representations and fast inference
• Winner-Take-All Autoencoders
• Parallel Multiscale Autoregressive Density Estimation

Cited by in corpus (5)

• On the Reconstruction of Face Images from Deep Face Templates
• Super-Resolution using Convolutional Neural Networks without Any Checkerboard Artifacts
• FlatteNet: A Simple Versatile Framework for Dense Pixelwise Prediction
• Fixed smooth convolutional layer for avoiding checkerboard artifacts in CNNs
• Versailles-FP dataset: Wall Detection in Ancient