Mixed Precision l1 Solver for Compressive Depth Reconstruction: An ADMM Case Study

Yun Wu, Andrew Michael Wallace, Andreas Aßmann, Brian Stewart

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Rapid reconstruction of depth images from sparsely sampled data is important for many machine learning applications, including robot or vehicle assistance or autonomy, which require low power LiDAR sensing for eye safety, and resource reduction for FPGA or solid state implementation, especially with constrained energy budgets. A new compressive depth reconstruction design approach is proposed using a compact ADMM solver for the lasso problem, which varies the precision scaling in an iterative optimization process. Implementations on an FPGA architecture show over 55% savings in hardware resources and 78% in power with only minor reduction in reconstructed depth image quality compared to single float precision.
Original languageEnglish
Title of host publication2021 International Workshop on Signal Processing Systems
Publication statusAccepted/In press - 5 Jul 2021
Event2021 International Workshop on Signal Processing Systems - Coimbra, Portugal
Duration: 19 Oct 202121 Oct 2021
https://sips2021.org/

Workshop

Workshop2021 International Workshop on Signal Processing Systems
Abbreviated titleSiPS 2021
Country/TerritoryPortugal
CityCoimbra
Period19/10/2121/10/21
Internet address

Fingerprint

Dive into the research topics of 'Mixed Precision l1 Solver for Compressive Depth Reconstruction: An ADMM Case Study'. Together they form a unique fingerprint.

Cite this