High-speed compressive range imaging based on active illumination

We report a compressive imaging method based on active illumination, which reconstructs a 3D scene at a frame rate beyond the acquisition speed limit of the camera. We have built an imaging prototype that projects temporally varying illumination pattern and demonstrated a joint reconstruction algorithm that iteratively retrieves both the range and high-temporal-frequency information from the 2D low-frame rate measurement. The reflectance and depth-map videos have been reconstructed at 1000 frames per second (fps) from the measurement captured at 200 fps. The range resolution is in agreement with the resolution calculated from the triangulation methods based on the same system geometry. We expect such an imaging method could become a simple solution to a wide range of applications, including industrial metrology, 3D printing, and vehicle navigations.