LimitNet: Progressive, Content-Aware Image Offloading for Extremely Weak Devices & Networks
This addresses the challenge of time-sensitive inference for IoT devices in remote areas with limited hardware and network capabilities, offering a novel solution for efficient offloading.
The paper tackles the problem of offloading image inference tasks from weak IoT devices to the cloud over low-bandwidth networks by introducing LimitNet, a progressive, content-aware compression model that achieves higher accuracy (e.g., 14.01 p.p. on ImageNet1000) and saves significant bandwidth (e.g., 61.24% on ImageNet1000) compared to state-of-the-art methods.
IoT devices have limited hardware capabilities and are often deployed in remote areas. Consequently, advanced vision models surpass such devices' processing and storage capabilities, requiring offloading of such tasks to the cloud. However, remote areas often rely on LPWANs technology with limited bandwidth, high packet loss rates, and extremely low duty cycles, which makes fast offloading for time-sensitive inference challenging. Today's approaches, which are deployable on weak devices, generate a non-progressive bit stream, and therefore, their decoding quality suffers strongly when data is only partially available on the cloud at a deadline due to limited bandwidth or packet losses. In this paper, we introduce LimitNet, a progressive, content-aware image compression model designed for extremely weak devices and networks. LimitNet's lightweight progressive encoder prioritizes critical data during transmission based on the content of the image, which gives the cloud the opportunity to run inference even with partial data availability. Experimental results demonstrate that LimitNet, on average, compared to SOTA, achieves 14.01 p.p. (percentage point) higher accuracy on ImageNet1000, 18.01 pp on CIFAR100, and 0.1 higher mAP@0.5 on COCO. Also, on average, LimitNet saves 61.24% bandwidth on ImageNet1000, 83.68% on CIFAR100, and 42.25% on the COCO dataset compared to SOTA, while it only has 4% more encoding time compared to JPEG (with a fixed quality) on STM32F7 (Cortex-M7).