Search papers, labs, and topics across Lattice.
This paper introduces Zero-Fi, a novel framework for zero-shot human activity recognition using Wi-Fi signals, which overcomes the limitations of existing methods that rely on labeled samples for known activities. By employing a contrastive signal-language alignment approach, Zero-Fi learns to create unified representations from Wi-Fi signal features and aligns them with natural language descriptions of activities in a shared embedding space. Experimental results on large-scale datasets show that Zero-Fi effectively recognizes previously unseen activities, showcasing the power of cross-modal alignment in expanding the capabilities of Wi-Fi sensing technologies.
Zero-Fi can recognize new human activities using Wi-Fi signals without any labeled samples, revolutionizing the potential for real-time activity monitoring.
Wi-Fi-based human activity recognition has advanced substantially, but most existing methods assume a closed set of activities and require labeled Wi-Fi samples for every target class, limiting their ability to recognize unseen activities. We present Zero-Fi, a contrastive signal-language alignment framework for zero-shot Wi-Fi-based human activity recognition. Zero-Fi learns unified representations from complementary Wi-Fi signal features and aligns them with the semantic representations of natural-language activity descriptions in a shared embedding space. This cross-modal alignment enables Zero-Fi to recognize new activity classes without requiring labeled Wi-Fi samples or model adaptation for those classes. Experiments on large-scale public benchmark datasets demonstrate effective zero-shot recognition of held-out activity classes, highlighting the potential of signal-language alignment to extend Wi-Fi sensing beyond predefined activity classes.
