this context, we propose two strategies: (i) directly processing event-camera data with traditional video-processing