Partial Convolution based Padding

Guilin Liu, Kevin J. Shih, Ting-Chun Wang, Fitsum A. Reda, Karan Sapra +3 more
11/28/2018
cs.CV

Official code: NVIDIA/partialconv (1,281 stars, last commit 2023-05-16).

Abstract

In this paper, we present a simple yet effective padding scheme that can be used as a drop-in module for existing convolutional neural networks. We call it partial convolution based padding, with the intuition that the padded region can be treated as holes and the original input as non-holes. Specifically, during the convolution operation, the convolution results are re-weighted near image borders based on the ratios between the padded area and the convolution sliding window area. Extensive experiments with various deep network models on ImageNet classification and semantic segmentation demonstrate that the proposed padding scheme consistently outperforms standard zero padding with better accuracy.

View on arXivView PDF

Code Implementations(1)

NVIDIA/partialconvOfficial100%

A New Padding Scheme: Partial Convolution based Padding

1,281212PythonNov 5, 20183 years agoNOASSERTION

Similar Papers

PadChannel: Improving CNN Performance through Explicit Padding Encoding

88% match2023

Padding Module: Learning the Padding in Deep Neural Networks

86% match2023

Improving Translation Invariance in Convolutional Neural Networks with Peripheral Prediction Padding

85% match2023

Padding-free Convolution based on Preservation of Differential Characteristics of Kernels

81% match2023

Deep Convolutional Neural Networks with Zero-Padding: Feature Extraction and Learning

81% match1 repo32023

On the Interplay of Convolutional Padding and Adversarial Robustness

79% match1 repo22023

Cite this paper

@article{liu2018partial,
  title  = {Partial Convolution based Padding},
  author = {Guilin Liu and Kevin J. Shih and Ting-Chun Wang and Fitsum A. Reda and Karan Sapra and Zhiding Yu and Andrew Tao and Bryan Catanzaro},
  year   = {2018},
  eprint = {1811.11718},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1811.11718}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1811.11718.svg)](https://codeofpaper.com/paper/1811.11718)

Discussion