Pedestrian recognition has achieved the state-of-the-art performance due to the progress of recent convolutional neural network (CNN). However, mainstream CNN models are too complicated to emerging Computing-In-Memory (CIM) architectures for hardware implementation, because enormous parameters and massive intermediate processing results may incur severe “memory bottleneck”. This paper proposed a design methodology of Parameter Substitution with Nodes Compensation (PSNC) to significantly reduce parameters of CNN model without inference accuracy degradation. Based on the PSNC methodology, an ultra-lightweight convolutional neural network (UL-CNN) was designed. The UL-CNN model is a specially optimized convolutional neural network aiming at a flash-based CIM architecture (Conv-Flash) and to apply for recognizing person. The implementation result of running UL-CNN on Conv-Flash shows that the inference accuracy is up to 94.7%. Compared to LeNet-5, on the premise of the similar operations and accuracy, the amounts of UL-CNN’s parameters are less than 37% of LeNet-5 at the same dataset benchmark. Such parameter reduction can dramatically speed up the training process and economize on-chip storage overhead, as well as save the power consumption of the memory access. With the aid of UL-CNN, the Conv-Flash architecture can provide the best energy efficiency compared to other platforms (CPU, GPU, FPGA, etc.), which consumes only 2.2 × 105J to complete pedestrian recognition for one frame.

中文翻译：

---

UL-CNN：一种针对行人识别的基于闪存的内存计算架构的超轻量级卷积神经网络

由于最近卷积神经网络（CNN）的进步，行人识别已经达到了最先进的性能。然而，主流的 CNN 模型对于新兴的内存计算（CIM）架构来说过于复杂，无法进行硬件实现，因为巨大的参数和大量的中间处理结果可能会导致严重的“内存瓶颈”。本文提出了一种带节点补偿的参数替换（PSNC）的设计方法，可以在不降低推理精度的情况下显着减少 CNN 模型的参数。基于PSNC方法，设计了一种超轻量级的卷积神经网络（UL-CNN）。UL-CNN模型是针对基于flash的CIM架构（Conv-Flash）专门优化的卷积神经网络，应用于人脸识别。在 Conv-Flash 上运行 UL-CNN 的实现结果表明，推理准确率高达 94.7%。与 LeNet-5 相比，在相同的操作和准确性的前提下，UL-CNN 的参数量在相同数据集基准上小于 LeNet-5 的 37%。这种参数减少可以显着加快训练过程并节省片上存储开销，并节省内存访问的功耗。借助 UL-CNN，Conv-Flash 架构相比其他平台（CPU、GPU、FPGA 等）可以提供最佳的能效，完成一帧行人识别仅消耗 2.2×105J。在同一数据集基准上，UL-CNN 的参数量不到 LeNet-5 的 37%。这种参数减少可以显着加快训练过程并节省片上存储开销，并节省内存访问的功耗。借助 UL-CNN，Conv-Flash 架构相比其他平台（CPU、GPU、FPGA 等）可以提供最佳的能效，完成一帧行人识别仅消耗 2.2×105J。在同一数据集基准上，UL-CNN 的参数量不到 LeNet-5 的 37%。这种参数减少可以显着加快训练过程并节省片上存储开销，并节省内存访问的功耗。借助 UL-CNN，Conv-Flash 架构相比其他平台（CPU、GPU、FPGA 等）可以提供最佳的能效，完成一帧行人识别仅消耗 2.2×105J。