Synthetic Image Generation for Robot Simulation: Quantifying the Impact of Model Modifications on Perception

摘要

Modeling cameras for the simulation of autonomous robotics is critical for generating synthetic images with appropriate realism to effectively evaluate a perception algorithm in simulation. In many cases though, simulated images are produced by traditional rendering techniques that exclude or superficially handle processing steps and aspects encountered in the actual camera pipeline. The purpose of this contribution is to quantify the effect that modifying the camera model has on the perception algorithm evaluated in simulation. We investigate what happens if one ignores aspects tied to processes from within the physical camera, e.g., lens distortion, noise, and signal processing; scene effects, e.g., lighting and reflection; and rendering quality. The results quantifiably indicate that, for the evaluated task, modeling modifications that result in large-scale changes to color, scene, and location had far greater impact on perception than model aspects concerned with local, feature-level artifacts.