$γ$-Quant: Towards Learnable Quantization for Low-bit Pattern Recognition
This addresses energy and bandwidth constraints in low-bit-depth sensor applications, such as wearable devices, by optimizing data processing for automated analysis rather than human perception.
The paper tackles the inefficiency of using high-bit data for automated pattern recognition by proposing a learnable non-linear quantization method, demonstrating that raw data quantized to as few as 4 bits can perform on par with 12-bit data in tasks like object detection and human activity recognition.
Most pattern recognition models are developed on pre-proce\-ssed data. In computer vision, for instance, RGB images processed through image signal processing (ISP) pipelines designed to cater to human perception are the most frequent input to image analysis networks. However, many modern vision tasks operate without a human in the loop, raising the question of whether such pre-processing is optimal for automated analysis. Similarly, human activity recognition (HAR) on body-worn sensor data commonly takes normalized floating-point data arising from a high-bit analog-to-digital converter (ADC) as an input, despite such an approach being highly inefficient in terms of data transmission, significantly affecting the battery life of wearable devices. In this work, we target low-bandwidth and energy-constrained settings where sensors are limited to low-bit-depth capture. We propose $γ$-Quant, i.e.~the task-specific learning of a non-linear quantization for pattern recognition. We exemplify our approach on raw-image object detection as well as HAR of wearable data, and demonstrate that raw data with a learnable quantization using as few as 4-bits can perform on par with the use of raw 12-bit data. All code to reproduce our experiments is publicly available via https://github.com/Mishalfatima/Gamma-Quant