EfficientRep:An Efficient Repvgg-style ConvNets with Hardware-aware Neural Network Design

TL;DR

This paper introduces EfficientRep, a hardware-aware ConvNet architecture inspired by RepVGG, optimized for hardware efficiency and integrated into YOLOv6 for improved object detection performance.

Contribution

The paper proposes a novel hardware-aware neural network design method and develops EfficientRep series networks that are GPU-friendly and enhance YOLOv6 object detection.

Findings

EfficientRep networks are highly compatible with GPU hardware.

Integration with YOLOv6 improves detection efficiency.

Designed networks outperform traditional models in hardware utilization.

Abstract

We present a hardware-efficient architecture of convolutional neural network, which has a repvgg-like architecture. Flops or parameters are traditional metrics to evaluate the efficiency of networks which are not sensitive to hardware including computing ability and memory bandwidth. Thus, how to design a neural network to efficiently use the computing ability and memory bandwidth of hardware is a critical problem. This paper proposes a method how to design hardware-aware neural network. Based on this method, we designed EfficientRep series convolutional networks, which are high-computation hardware(e.g. GPU) friendly and applied in YOLOv6 object detection framework. YOLOv6 has published YOLOv6N/YOLOv6S/YOLOv6M/YOLOv6L models in v1 and v2 versions.