Over the past few years there has been a growing interest in monolithic arrays of single photon avalanche diodes (SPAD) for spatially resolved detection of faint ultrafast optical signals. SPADs implemented in CMOS-compatible planar technologies offer the typical advantages of microelectronic devices (small size, ruggedness, low voltage, low power, etc.). Furthermore, they have inherently higher photon detection efficiency than PMTs and are able to provide, beside sensitivities down to single-photons, very high acquisition speeds (i.e. either high frame-rates or very short integration time-slots). SPADs offer several advantages over other commercially available imagers. For example, CCDs and similar imagers lack in speed because their readout process is based on a slow charge-transfer mechanisms. CMOS APS, on the other hand, are unable to detect very faint optical signals, due to poor sensitivity and noisy electronics. In order to make SPAD array more and more competitive it is necessary to face several issues: dark counts, quantum efficiency, crosstalk, timing performance. These issues will be discussed in the context of two possible approaches to such a challenge: employing a standard industrial CMOS technology or developing a dedicated technology. Advances recently attained will be outlined with reference to both photon counting and Time correlated single photon counting detector arrays.© (2009) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.