Search papers, labs, and topics across Lattice.
RPTU, DFKI
3
1
6
LITEWAY slashes model size and energy consumption for wearable human activity recognition without sacrificing performance, achieving up to 9.52x size reduction.
Structured video-grounded semantics can boost synthetic IMU generation, leading to a 19.86% improvement in tail-class recognition over traditional methods.
By explicitly modeling how people differ, adversarial learning can be steered to produce activity recognition models that generalize better to new users.
