Single Photon Avalanche Diode sensor arrays operating in direct time of flight mode can perform 3D imaging using pulsed lasers. Operating at high frame rates, SPAD imagers typically generate large volumes of noisy and largely redundant spatio-temporal data. This results in communication bottlenecks and unnecessary data processing. In this work, we propose a set of neuromorphic processing solutions to this problem. By processing the SPAD generated spatio-temporal patterns locally and in an event-based manner, the proposed methods reduce the size of output data transmitted from the sensor by orders of magnitude while increasing the utility of the output data in the context of challenging recognition tasks. To demonstrate these results, the first large scale complex SPAD imaging dataset is presented involving high-speed view-invariant recognition of airplanes with background clutter. The framebased SPAD imaging dataset is converted via several alternative methods into event-based data streams and processed using a range of feature extractor networks and pooling methods. The results of the event-based processing methods are compared to processing the original frame-based dataset via frame-based but otherwise identical architectures. The results show the eventbased methods are superior to the frame-based approach both in terms of classification accuracy and output data-rate.