Mask-ToF: Learning Microlens Masks for Flying Pixel Correction in Time-of-Flight Imaging
This addresses a pervasive artifact in ToF depth sensing that can adversely affect downstream 3D vision tasks, representing a domain-specific improvement.
The paper tackled the problem of flying pixels (FP) in time-of-flight (ToF) depth imaging, which cause erroneous depth estimates around edges, by introducing Mask-ToF, a method that learns microlens masks to modulate light mixtures and reduce FPs, resulting in cutting FP counts in half on real data.
We introduce Mask-ToF, a method to reduce flying pixels (FP) in time-of-flight (ToF) depth captures. FPs are pervasive artifacts which occur around depth edges, where light paths from both an object and its background are integrated over the aperture. This light mixes at a sensor pixel to produce erroneous depth estimates, which can adversely affect downstream 3D vision tasks. Mask-ToF starts at the source of these FPs, learning a microlens-level occlusion mask which effectively creates a custom-shaped sub-aperture for each sensor pixel. This modulates the selection of foreground and background light mixtures on a per-pixel basis and thereby encodes scene geometric information directly into the ToF measurements. We develop a differentiable ToF simulator to jointly train a convolutional neural network to decode this information and produce high-fidelity, low-FP depth reconstructions. We test the effectiveness of Mask-ToF on a simulated light field dataset and validate the method with an experimental prototype. To this end, we manufacture the learned amplitude mask and design an optical relay system to virtually place it on a high-resolution ToF sensor. We find that Mask-ToF generalizes well to real data without retraining, cutting FP counts in half.