Semantic Scene Segmentation for Robotics Applications

TL;DR

This paper evaluates current semantic scene segmentation models' inference speed and suitability for robotics applications with low-power hardware and high-resolution inputs, providing a comparative analysis for optimal model selection.

Contribution

It offers a comprehensive comparison of top segmentation models' performance under robotics-specific computational constraints, guiding practical deployment choices.

Findings

Identifies models with best inference speed on low-power GPUs

Highlights trade-offs between accuracy and computational efficiency

Provides recommendations for model selection in robotics contexts

Abstract

Semantic scene segmentation plays a critical role in a wide range of robotics applications, e.g., autonomous navigation. These applications are accompanied by specific computational restrictions, e.g., operation on low-power GPUs, at sufficient speed, and also for high-resolution input. Existing state-of-the-art segmentation models provide evaluation results under different setups and mainly considering high-power GPUs. In this paper, we investigate the behavior of the most successful semantic scene segmentation models, in terms of deployment (inference) speed, under various setups (GPUs, input sizes, etc.) in the context of robotics applications. The target of this work is to provide a comparative study of current state-of-the-art segmentation models so as to select the most compliant with the robotics applications requirements.