CPC H04N 7/188 (2013.01) [G06F 18/2193 (2023.01); G06T 9/002 (2013.01); G06V 10/10 (2022.01); G06V 10/147 (2022.01); G06V 10/764 (2022.01); G06V 20/44 (2022.01); G06V 20/52 (2022.01); G06V 40/25 (2022.01); H04L 43/028 (2013.01); H04L 43/04 (2013.01); H04N 7/181 (2013.01); H04N 19/00 (2013.01); H04N 19/14 (2014.11); H04N 19/188 (2014.11); H04N 25/47 (2023.01)] | 9 Claims |
3. A method comprising:
receiving a plurality of input packets from a dynamic vision sensor (DVS) by a sensor input, wherein the DVS is configured to output a plurality of pixel values;
generating, by a coincidence detector, a plurality of thickened edges based on the plurality of input packets, wherein the plurality of thickened edges comprises a first thickened edge, and the first thickened edge indicates a coincidence between a first pixel value and a second pixel value of the plurality of pixel values, and the coincidence is detected by identifying that the first pixel value and the second pixel value have a same detected polarity;
obtaining a packet payload based on event counts of the plurality of thickened edges by an aggregator; and
transmitting, to a detector, a compressed data packet, wherein the compressed data packet is based on the packet payload by a transmitter.
|