A Bayesian Based Deep Unrolling Algorithm for Single-Photon Lidar Systems

  title={A Bayesian Based Deep Unrolling Algorithm for Single-Photon Lidar Systems},
  author={Jakeoung Koo and Abderrahim Halimi and Stephen Mclaughlin},
  journal={IEEE Journal of Selected Topics in Signal Processing},
Deploying 3D single-photon Lidar imaging in real world applications presents multiple challenges including imaging in high noise environments. Several algorithms have been proposed to address these issues based on statistical or learning-based frameworks. Statistical methods provide rich information about the inferred parameters but are limited by the assumed model correlation structures, while deep learning methods show state-of-the-art performance but limited inference guarantees, preventing… 

A Bayesian Based Unrolling Approach to Single-Photon Lidar Imaging through Obscurants

A deep learning model for 3D single-photon Lidar imaging through obscurants, i.e., in the presence of a high and non-uniform background, which enables an estimation of parameters uncertainties, suitable for decision making.

3D Target Detection and Spectral Classification for Single-photon LiDAR Data

3D single-photon LiDAR imaging has an important role in many applications. However, full deployment of this modality will require the analysis of low signal to noise ratio target returns and a very

Real-Time Reconstruction of 3D Videos From Single-Photon LiDaR Data in the Presence of Obscurants

Comparisons with state-of-the-art algorithms on simulated and real underwater data demonstrate the benefit of the proposed framework for target detection, and for fast and robust depth estimation at multiple frames per second.



Robust and Guided Bayesian Reconstruction of Single-Photon 3D Lidar Data: Application to Multispectral and Underwater Imaging

The proposed Bayesian model and its estimation algorithm are validated on both synthetic and real images showing competitive results regarding the quality of the inferences and the computational complexity when compared to the state-of-the-art algorithms.

Expectation-Maximization Based Approach to 3D Reconstruction From Single-Waveform Multispectral Lidar Data

A novel Bayesian approach for estimating spectral and range profiles from single-photon Lidar waveforms associated with single surfaces in the photon-limited regime, which demonstrates a significant speed-up without significant degradation of the reconstruction performance when compared to existing methods in the photons-starved regime.

Robust Restoration of Sparse Multidimensional Single-Photon LiDAR Images

A new reconstruction algorithm for single-photon 3-D Lidar images is presented that can deal with multiple tasks, and is validated on synthetic and real data and in challenging realistic scenarios including sparse photon regimes for fast imaging, the presence of high background due to obscurants, and the joint processing of multispectral and/or multitemporal data.

Fast Task-Based Adaptive Sampling for 3D Single-Photon Multispectral Lidar Data

3D single-photon LiDAR imaging plays an important role in numerous applications. However, long acquisition times and significant data volumes present a challenge for LiDAR imaging. This paper

Bayesian 3D Reconstruction of Complex Scenes from Single-Photon Lidar Data

A new approach to 3D reconstruction using single-photon, single-wavelength Lidar data, which is capable of identifying multiple surfaces in each pixel is presented, and a multiresolution approach is adopted, processing the data from a coarse to the finest scale.

Single-photon 3D imaging with deep sensor fusion

This paper proposes a data-driven method for photon-efficient 3D imaging which leverages sensor fusion and computational reconstruction to rapidly and robustly estimate a dense depth map from low photon counts.

Photon-Efficient 3D Imaging with A Non-local Neural Network

A non-local neural network trained on simulated data that achieves decent reconstruction fidelity even under photon counts (and signal-to-background ratio, SBR) as low as 1 photon/pixel (and 0.01 SBR), which significantly surpasses the state-of-the-art.

Robust real-time 3D imaging of moving scenes through atmospheric obscurant using single-photon LiDAR

This paper presents the reconstruction of depth profiles of moving objects through high levels of obscurant equivalent to five attenuation lengths between transceiver and target at stand-off distances up to 150 m, and demonstrates a way forward for high-resolution imaging from mobile platforms in degraded visual environments.

A Few Photons Among Many: Unmixing Signal and Noise for Photon-Efficient Active Imaging

A new approach to depth and reflectivity estimation that emphasizes the unmixing of contributions from signal and noise sources is introduced, and improved performance of both reflectivity and depth estimates over state-of-the-art methods, especially at low SBR.

Real-time 3D reconstruction from single-photon lidar data using plug-and-play point cloud denoisers

A new computational framework for real-time three-dimensional (3D) scene reconstruction from single-photon data is shown by combining statistical models with highly scalable computational tools from the computer graphics community and enables robust, real- time target reconstruction of complex moving scenes.