Generative Latent Alignment for Interpretable Radar Based Occupancy Detection in Ambient Assisted Living
This addresses privacy concerns in Ambient Assisted Living by providing interpretable radar-based occupancy detection, but it is incremental as it builds on existing methods like VAEs and CLIP for a specific domain.
The paper tackled the problem of making mmWave radar presence detection more interpretable for Ambient Assisted Living by proposing a Generative Latent Alignment framework, which learns a latent representation aligned with semantic anchors and uses Grad-CAM to visualize spatial evidence for decisions, with qualitative observations showing compact blobs for person detection and diffuse evidence for empty rooms.
In this work, we study how to make mmWave radar presence detection more interpretable for Ambient Assisted Living (AAL) settings, where camera-based sensing raises privacy concerns. We propose a Generative Latent Alignment (GLA) framework that combines a lightweight convolutional variational autoencoder with a frozen CLIP text encoder to learn a low-dimensional latent representation of radar Range-Angle (RA) heatmaps. The latent space is softly aligned with two semantic anchors corresponding to "empty room" and "person present", and Grad-CAM is applied in this aligned latent space to visualize which spatial regions support each presence decision. On our mmWave radar dataset, we qualitatively observe that the "person present" class produces compact Grad-CAM blobs that coincide with strong RA returns, whereas "empty room" samples yield diffuse or no evidence. We also conduct an ablation study using unrelated text prompts, which degrades both reconstruction and localization, suggesting that radar-specific anchors are important for meaningful explanations in this setting.