University Building Recognition Dataset in Thailand for the mission-oriented IoT sensor system
Takara Taniguchi, Yudai Ueda, Atsuya Muramatsu, Kohki Hashimoto, Ryo Yagi, Hideya Ochiai, Chaodit Aswakul
TL;DR
The paper addresses the need for edge-centric, privacy-preserving learning for mission-oriented IoT sensing, where centralized federated learning poses single-point failures and lacks task-specific datasets. It introduces WAFL-ViT, a fully decentralized learning approach, and a campus-specific building recognition dataset (CUBR) tailored for Chulalongkorn University to evaluate performance in a real-world IoT setting. Through experiments comparing WAFL against self-training across ViT, VGG, ResNet, and MobileNet backbones, the work demonstrates that WAFL achieves higher accuracy on CUBR. The authors also release CUBR on GitHub, highlighting its potential to enable robust, device-to-device collaboration for mission fulfillment in campus environments.
Abstract
Many industrial sectors have been using of machine learning at inference mode on edge devices. Future directions show that training on edge devices is promising due to improvements in semiconductor performance. Wireless Ad Hoc Federated Learning (WAFL) has been proposed as a promising approach for collaborative learning with device-to-device communication among edges. In particular, WAFL with Vision Transformer (WAFL-ViT) has been tested on image recognition tasks with the UTokyo Building Recognition Dataset (UTBR). Since WAFL-ViT is a mission-oriented sensor system, it is essential to construct specific datasets by each mission. In our work, we have developed the Chulalongkorn University Building Recognition Dataset (CUBR), which is specialized for Chulalongkorn University as a case study in Thailand. Additionally, our results also demonstrate that training on WAFL scenarios achieves better accuracy than self-training scenarios. Dataset is available in https://github.com/jo2lxq/wafl/.