Search papers, labs, and topics across Lattice.
This paper introduces a privacy-preserving clothing classification scheme using Vision Transformers (ViT) to enable secure occupant-centric control (OCC) systems. They address the challenge of accuracy degradation when applying conventional privacy-preserving methods to image classification for thermal comfort estimation. Their scheme, applied to the DeepFashion dataset, maintains high accuracy on encrypted images, showing no degradation compared to plain images, unlike pixel-based methods.
Encrypting images no longer has to mean sacrificing accuracy in clothing classification tasks, thanks to a clever Vision Transformer approach.
A privacy-preserving clothing classification scheme is presented to enable secure occupant-centric control (OCC) systems. Although the utilization of camera images for HVAC control has been widely studied to optimize thermal comfort, privacy protection of occupant images has not been considered in prior works. While various privacy-preserving methods have been proposed for image classification, applying conventional schemes results in severe accuracy degradation. In this paper, we introduce a privacy-preserving classification method using Vision Transformer (ViT) applied to clothing insulation estimation. In an experiment using the DeepFashion dataset categorized by clothing insulation, while the conventional pixel-based method suffers a severe accuracy drop, our scheme maintains a high accuracy on encrypted images, showing no degradation from plain images across all categories.
