A Column Streaming-Based Convolution Engine and Mapping Algorithm for CNN-based Edge AI accelerators
This work addresses performance and resource constraints for edge AI devices like UAVs and wearables, but it appears incremental as it builds on existing accelerator designs.
The paper tackled the challenge of designing efficient CNN accelerators for edge AI applications with strict area and power constraints by proposing a column streaming-based convolution engine, achieving similar execution cycles to a commercial accelerator for a 227 x 227 feature map while avoiding zero-padding penalties.
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.