RT-ST-GCN: Enabling Realtime Continual Inference at the Edge
Published in 2024 IEEE 34th International Workshop on Machine Learning for Signal Processing (MLSP), 2024
We propose a hardware-driven optimization, termed RT-ST-GCN, which scales down the computational bottleneck of ST-GCN to achieve realtime predictions up to 50 fps.
Recommended citation: Yudayev, M., Filtjens, B., & Balasch, J. (2024, September). "RT-ST-GCN: Enabling Realtime Continual Inference at the Edge." In 2024 IEEE 34th International Workshop on Machine Learning for Signal Processing (MLSP) (pp. 1-6). IEEE. doi: 10.1109/MLSP58920.2024.10734754.
Download Paper