Dilated convolution
A dilated convolution applies a filter whose sampling positions are spaced apart, expanding the receptive field without proportionally increasing kernel size or parameters.
Sutskever's List
A dilated convolution applies a convolutional filter with spaced sampling positions to enlarge the receptive field without proportionally increasing filter size.
- How it is used
- Preserves full-resolution feature maps while aggregating broad context for dense pixel prediction.
- How deep it goes
- Explains spaced sampling and stacked dilation rates, comparing receptive-field growth with parameter growth.
- Examples
- Uses dilation rates 1, 2, 4, and 8 and a modified VGG-16 backbone.
- What you can do after
- Evaluate ways to expand receptive fields while retaining resolution and avoiding larger kernels or pooling.
Fight Fraud with Machine Learning
Atrous convolution is an image convolution operation that spaces kernel elements apart to expand the input area covered without increasing the number of kernel parameters.
- How it is used
- Supports image segmentation by expanding contextual coverage while retaining pixel-level predictions for forged and genuine regions.
- How deep it goes
- Explains kernel spacing, dilation rates, field-of-view expansion, and parameter efficiency without detailing kernel arithmetic or implementation code.
- Examples
- Uses forged-receipt pixel masks and the DeepLabV3 architecture as concrete applications.
- What you can do after
- Apply it via DeepLabV3 to capture wider image context for pixel-wise document-forgery segmentation.
Read how Fight Fraud with Machine Learning explains dilated convolution →
What to know first
- convolution kernelA convolution kernel is an array or tensor of weights applied to local regions of structured input to compute convolution responses.
Related concepts
- stridecontrasts with · Stride is the step or spacing between successive positions in an array access or successive placements of an operation over input data.
- DeepLabV3used by · DeepLabV3 is a convolutional-neural-network-based image segmentation architecture that uses atrous convolution and spatial pyramid pooling to capture information at multiple scales.
- receptive fieldproduces · A receptive field is the set or region of input elements whose information can influence a particular neural-network unit, representation, or output.
- convolution kerneluses · A convolution kernel is an array or tensor of weights applied to local regions of structured input to compute convolution responses.
- convolutional layerkind of · A convolutional layer is a neural-network layer that applies shared learned filters to local regions of structured input to produce feature representations.
Questions about dilated convolution
What is dilated convolution?
A dilated convolution applies a filter whose sampling positions are spaced apart, expanding the receptive field without proportionally increasing kernel size or parameters.
Which Manning books explain dilated convolution?
Sutskever's List and Fight Fraud with Machine Learning.
What should I know before learning dilated convolution?
Manning books that teach dilated convolution usually introduce convolution kernel first.
What is the difference between dilated convolution and stride?
A dilated convolution applies a filter whose sampling positions are spaced apart, expanding the receptive field without proportionally increasing kernel size or parameters. Stride is the step or spacing between successive positions in an array access or successive placements of an operation over input data.
What is dilated convolution a kind of?
Dilated convolution is a kind of convolutional layer. A convolutional layer is a neural-network layer that applies shared learned filters to local regions of structured input to produce feature representations.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.

