Physics-based Differentiable Depth Sensor Simulation

TL;DR

This paper presents a physics-based, differentiable simulation pipeline for generating realistic 2.5D depth scans, enabling improved training of deep learning models for depth-based recognition tasks.

Contribution

It introduces the first end-to-end differentiable depth sensor simulation built on physics-based rendering and custom algorithms, allowing parameter tuning and integration into larger vision systems.

Findings

Enhanced deep learning model performance on real scans.

Improved simulation fidelity over previous static methods.

Effective domain adaptation for depth-based tasks.

Abstract

Gradient-based algorithms are crucial to modern computer-vision and graphics applications, enabling learning-based optimization and inverse problems. For example, photorealistic differentiable rendering pipelines for color images have been proven highly valuable to applications aiming to map 2D and 3D domains. However, to the best of our knowledge, no effort has been made so far towards extending these gradient-based methods to the generation of depth (2.5D) images, as simulating structured-light depth sensors implies solving complex light transport and stereo-matching problems. In this paper, we introduce a novel end-to-end differentiable simulation pipeline for the generation of realistic 2.5D scans, built on physics-based 3D rendering and custom block-matching algorithms. Each module can be differentiated w.r.t sensor and scene parameters; e.g., to automatically tune the simulation for new devices over some provided scans or to leverage the pipeline as a 3D-to-2.5D transformer within larger computer-vision applications. Applied to the training of deep-learning methods for various depth-based recognition tasks (classification, pose estimation, semantic segmentation), our simulation greatly improves the performance of the resulting models on real scans, thereby demonstrating the fidelity and value of its synthetic depth data compared to previous static simulations and learning-based domain adaptation schemes.