Partial Convolution based Padding
arXiv:1811.11718
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.
11 pages; code is available at https://github.com/NVIDIA/partialconv
Cited by in corpus (18)
- Review: Deep Learning in Electron Microscopy
- CMID: A Unified Self-Supervised Learning Framework for Remote Sensing Image Understanding
- Position, Padding and Predictions: A Deeper Look at Position Information in CNNs
- Mind the Pad -- CNNs can Develop Blind Spots
- Continual 3D Convolutional Neural Networks for Real-time Processing of Videos
- A Tailored Convolutional Neural Network for Nonlinear Manifold Learning of Computational Physics Data using Unstructured Spatial Discretizations
- Who Make Drivers Stop? Towards Driver-centric Risk Assessment: Risk Object Identification via Causal Inference
- Transposer: Universal Texture Synthesis Using Feature Maps as Transposed Convolution Filter
- Context-aware Padding for Semantic Segmentation
- SRWarp: Generalized Image Super-Resolution under Arbitrary Transformation
- Uncertainties of Satellite-based Essential Climate Variables from Deep Learning
- Painting Outside as Inside: Edge Guided Image Outpainting via Bidirectional Rearrangement with Progressive Step Learning
- FOSNet: An End-to-End Trainable Deep Neural Network for Scene Recognition
- Normalized Label Distribution: Towards Learning Calibrated, Adaptable and Efficient Activation Maps
- MarioNette: Self-Supervised Sprite Learning
- Repopulating Street Scenes
- One Shot 3D Photography
- The Impact of Hole Geometry on Relative Robustness of In-Painting Networks: An Empirical Study