“…High integration density, low power, low fabrication cost, CMOS compatibility [1] and full back-end integration, make OxRAM a suitable candidate for many applications. Recent studies have also proposed different OxRAM devices for image-sensing and imageprocessing related application use-cases such as event-detection [4], feature extraction [5], sparse encoding [6], recognition [7], compression [8], overexposure/underexposure control [9,10], dynamic range improvement [11] and real-time background tracking [12]. At the same time, unlike conventional vision systems, biological vision systems are known to perform more efficiently [13][14][15] on real-world vision tasks.…”