Hardware-aware mobile building block evaluation for computer vision
Not on the map. Why?
Abstract (arXiv)
In this work we propose a methodology to accurately evaluate and compare the performance of efficient neural network building blocks for computer vision in a hardware-aware manner. Our comparison uses pareto fronts based on randomly sampled networks from a design space to capture the underlying accuracy/complexity trade-offs. We show that our approach allows to match the information obtained by previous comparison paradigms, but provides more insights in the relationship between hardware cost and accuracy. We use our methodology to analyze different building blocks and evaluate their performance on a range of embedded hardware platforms. This highlights the importance of benchmarking building blocks as a preselection step in the design process of a neural network. We show that choosing the right building block can speed up inference by up to a factor of 2x on specific hardware ML accelerators.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 29 of its 31 references that are in the map. First posted on arXiv 26 Aug 2022.
Builds on
29 of 31 references.
References · 29
- 2015Deep Residual Learning for Image Recognitioncited by 51,809
- 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
- 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,415
- 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,773
- 2017Squeeze-and-Excitation Networkscited by 4,763
- 2017MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applicationscited by 3,827
- 2016Aggregated Residual Transformations for Deep Neural Networkscited by 2,540
- 2019Searching for MobileNetV3cited by 1,984
- 2017ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devicescited by 1,194
- 2018ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Designcited by 1,132
- 2021EfficientNetV2: Smaller Models and Faster Trainingcited by 1,127
- 2017Learning Transferable Architectures for Scalable Image Recognitioncited by 1,097
17 more
- 2020Designing Network Design Spacescited by 954
- 2018MnasNet: Platform-Aware Neural Architecture Search for Mobilecited by 937
- 2018Regularized Evolution for Image Classifier Architecture Searchcited by 902
- 2018Efficient Neural Architecture Search via Parameter Sharingcited by 831
- 2019Once-for-All: Train One Network and Specialize it for Efficient Deploymentcited by 778
- 2017Progressive Neural Architecture Searchcited by 483
- 2018FBNet: Hardware-Aware Efficient ConvNet Design via Differentiable Neural Architecture Searchcited by 480
- 2019Single Path One-Shot Neural Architecture Search with Uniform Samplingcited by 411
- 2016Designing Neural Network Architectures using Reinforcement Learningcited by 277
- 2019Single-Path NAS: Designing Hardware-Efficient ConvNets in less than 4 Hourscited by 87
- 2019On Network Design Spaces for Visual Recognitioncited by 68
- 2020MobileDets: Searching for Object Detection Architectures for Mobile Acceleratorscited by 55
- 2019Visual Wake Words Datasetcited by 46
- 2020Accelerator-aware Neural Network Design using AutoMLcited by 30
- 2021Google Neural Network Models for Edge Devices: Analyzing and Mitigating Machine Learning Inference Bottleneckscited by 22
- 2020Optimizing Grouped Convolutions on Edge Devicescited by 12
- 2022Searching for Efficient Neural Architectures for On-Device ML on Edge TPUscited by 4
Built on by
Citing papers per year: 2023: 1.
1 paper.