Hardware-aware mobile building block evaluation for computer vision
This work addresses the need for efficient neural network design on embedded hardware, though it is incremental as it builds on existing comparison paradigms.
The authors tackled the problem of evaluating neural network building blocks for computer vision in a hardware-aware way, showing that their methodology provides insights into hardware cost-accuracy trade-offs and can speed up inference by up to 2x on specific hardware accelerators.
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.