A Column Streaming-Based Convolution Engine and Mapping Algorithm for CNN-based Edge AI accelerators

TL;DR

This paper introduces a column streaming-based convolution engine and mapping algorithm tailored for CNN-based edge AI accelerators, optimizing performance under strict area and power constraints for portable applications.

Contribution

It proposes a flexible column streaming-based convolution engine with a novel mapping algorithm, improving efficiency for diverse CNN algorithms in edge AI devices.

Findings

Requires similar execution cycles as commercial accelerators for 227x227 feature maps

Avoids zero-padding penalties, enhancing processing efficiency

Supports various CNN algorithms through flexible design

Abstract

Edge AI accelerators have been emerging as a solution for near customers' applications in areas such as unmanned aerial vehicles (UAVs), image recognition sensors, wearable devices, robotics, and remote sensing satellites. These applications not only require meeting performance targets but also meeting strict area and power constraints due to their portable mobility feature and limited power sources. As a result, a column streaming-based convolution engine has been proposed in this paper that includes column sets of processing elements design for flexibility in terms of the applicability for different CNN algorithms in edge AI accelerators. Comparing to a commercialized CNN accelerator, the key results reveal that the column streaming-based convolution engine requires similar execution cycles for processing a 227 x 227 feature map with avoiding zero-padding penalties.