Smoothed dilated convolutions for improved dense prediction

By Zhengyang Wang, Shuiwang Ji. 2018

188 citations recorded by the source corpus.

Dilated convolutions, also known as atrous convolutions, have been widely explored in deep convolutional neural networks (DCNNs) for various dense prediction tasks. However, dilated convolutions suffer from the gridding artifacts, which hampers the performance. In this work, we propose two simple yet effective degridding methods by studying a decomposition of dilated convolutions. Unlike existing models, which explore solutions by focusing on a block of cascaded dilated convolutional layers, our methods address the gridding artifacts by smoothing the dilated convolution itself. In addition, we point out that the two degridding approaches are intrinsically related and define separable and shared (SS) operations, which generalize the proposed methods. We further explore SS operations in view of operations on graphs and propose the SS output layer, which is able to smooth the entire DCNNs by only replacing the output layer. We evaluate our degridding methods and the SS output layer thoroughly, and visualize the smoothing effect through effective receptive field analysis. Results show that our methods degridding yield consistent improvements on the performance of dense prediction tasks, while adding negligible amounts of extra training parameters. And the SS output layer improves the performance by 3.3% and contains only 9% training parameters of the original output layer.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Pay Less Attention with Lightweight and Dynamic Convolutions (2019, background): “Wang & Ji (2018) ties the weights of all channels (H = 1).”
Gated Context Aggregation Network for Image Dehazing and Deraining (2018, methodology): “To understand this issue more clearly, a very recent work [37] analyzes the dilated convolution in a compositional way.”
Aggregated Contextual Transformations for High-Resolution Image Inpainting (2021, background): “However, as widely acknowledged, serialized dilated convolutions tend to encode features of predefined gridding patterns rather than patterns of interest for context reasoning [13], [23], [47].”