Data for: Motion adaptive deblurring with single photon cameras

Single-photon avalanche diodes (SPADs) are a rapidly developing
image sensing technology with extreme lowlight sensitivity and picosecond
timing resolution. These unique capabilities have enabled SPADs
to be used in applications like LiDAR, non-line-of-sight imaging and
fluorescence microscopy that require imaging in
photon-starved scenarios. In this work we harness these
capabilities for dealing with motion blur in a passive imaging
setting in low illumination conditions. Our key insight is that
the data captured by a SPAD array camera can be represented as
a 3D spatio-temporal tensor of photon detection events
which can be integrated along arbitrary spatio-temporal
trajectories with dynamically varying integration windows, depending on
scene motion. We propose an algorithm that estimates pixel motion from
photon timestamp data and dynamically adapts the integration windows to
minimize motion blur. Our simulation results show the
applicability of this algorithm to a variety of motion profiles
including translation, rotation and local object motion. We also
demonstrate the real-world feasibility of our method on data
captured using a 32 × 32 SPAD camera.