P2M: A Processing-in-Pixel-in-Memory Paradigm for Resource-Constrained TinyML Applications

TL;DR

This paper introduces P2M, a novel in-memory processing paradigm integrated into pixel arrays to enable energy-efficient, bandwidth-reducing CNN processing directly within CMOS image sensors for TinyML applications.

Contribution

It presents a new pixel array design supporting analog multi-channel, multi-bit convolution, batch normalization, and ReLU, enabling in-pixel CNN layer processing as a drop-in replacement.

Findings

Reduces data transfer bandwidth by ~21x.

Lowers energy-delay product by ~11x for TinyML CNNs.

Maintains test accuracy comparable to standard methods.

Abstract

The demand to process vast amounts of data generated from state-of-the-art high resolution cameras has motivated novel energy-efficient on-device AI solutions. Visual data in such cameras are usually captured in the form of analog voltages by a sensor pixel array, and then converted to the digital domain for subsequent AI processing using analog-to-digital converters (ADC). Recent research has tried to take advantage of massively parallel low-power analog/digital computing in the form of near- and in-sensor processing, in which the AI computation is performed partly in the periphery of the pixel array and partly in a separate on-board CPU/accelerator. Unfortunately, high-resolution input images still need to be streamed between the camera and the AI processing unit, frame by frame, causing energy, bandwidth, and security bottlenecks. To mitigate this problem, we propose a novel Processing-in-Pixel-in-memory (P2M) paradigm, that customizes the pixel array by adding support for analog multi-channel, multi-bit convolution, batch normalization, and ReLU (Rectified Linear Units). Our solution includes a holistic algorithm-circuit co-design approach and the resulting P2M paradigm can be used as a drop-in replacement for embedding memory-intensive first few layers of convolutional neural network (CNN) models within foundry-manufacturable CMOS image sensor platforms. Our experimental results indicate that P2M reduces data transfer bandwidth from sensors and analog to digital conversions by ~21x, and the energy-delay product (EDP) incurred in processing a MobileNetV2 model on a TinyML use case for visual wake words dataset (VWW) by up to ~11x compared to standard near-processing or in-sensor implementations, without any significant drop in test accuracy.