Using detector arrays can speed up lidar systems by parallelizing acquisition.
However, current SPAD arrays have time bins longer than
typical laser pulse durations, resulting in measurement errors dominated
by quantization. We propose an optical time-of-flight system
that uses subtractive dither to improve image depth resolution.
Modeling the measurement noise with a generalized Gaussian distribution
further improves estimation error in simulations, although
model mismatch prevents the same advantage for our experimental
data. Experimental results with the ratio of laser pulse standard deviation
to quantization bin duration equal to 0.15 and using an average
of 267 photons per pixel show a reduction in RMS error as large as
9-fold over estimates from coarsely quantized data.