Efficient Accelerator for Dilated and Transposed Convolution with Decomposition
Not on the map. Why?
Abstract (arXiv)
Hardware acceleration for dilated and transposed convolution enables real time execution of related tasks like segmentation, but current designs are specific for these convolutional types or suffer from complex control for reconfigurable designs. This paper presents a design that decomposes input or weight for dilated and transposed convolutions respectively to skip redundant computations and thus executes efficiently on existing dense CNN hardware as well. The proposed architecture can cut down 87.8% of the cycle counts to achieve 8.2X speedup over a naive execution for the ENet case.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 10 of its 19 references that are in the map. First posted on arXiv 2 May 2022.
Builds on
10 of 19 references.
References · 10
- 2015Deep Residual Learning for Image Recognitioncited by 52,096
- 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
- 2016The Cityscapes Dataset for Semantic Urban Scene Understandingcited by 4,444
- 2014Going Deeper with Convolutionscited by 4,241
- 2017MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applicationscited by 3,845
- 2016DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFscited by 3,241
- 2013Rich feature hierarchies for accurate object detection and semantic segmentationcited by 2,162
- 2015Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networkscited by 2,054
- 2016Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learningcited by 1,512
- 2016ENet: A Deep Neural Network Architecture for Real-Time Semantic Segmentationcited by 291
Built on by
Citing papers per year: 2023: 1.
1 paper.