TSLA: A Task-Specific Learning Adaptation for Semantic Segmentation on Autonomous Vehicles Platform

TL;DR

This paper introduces TSLA, a method for customizing semantic segmentation networks for autonomous vehicles by dynamically adapting model complexity based on hardware constraints and driving scenarios, using Bayesian Optimization for hyperparameter tuning.

Contribution

It proposes a novel task-specific learning adaptation framework with a three-tier control mechanism and Bayesian Optimization to optimize model performance under resource limitations.

Findings

Enhanced resource efficiency on embedded devices

Improved segmentation accuracy tailored to scenarios

Effective hyperparameter search under computational constraints

Abstract

Autonomous driving platforms encounter diverse driving scenarios, each with varying hardware resources and precision requirements. Given the computational limitations of embedded devices, it is crucial to consider computing costs when deploying on target platforms like the NVIDIA\textsuperscript{\textregistered} DRIVE PX 2. Our objective is to customize the semantic segmentation network according to the computing power and specific scenarios of autonomous driving hardware. We implement dynamic adaptability through a three-tier control mechanism -- width multiplier, classifier depth, and classifier kernel -- allowing fine-grained control over model components based on hardware constraints and task requirements. This adaptability facilitates broad model scaling, targeted refinement of the final layers, and scenario-specific optimization of kernel sizes, leading to improved resource allocation and performance. Additionally, we leverage Bayesian Optimization with surrogate modeling to efficiently explore hyperparameter spaces under tight computational budgets. Our approach addresses scenario-specific and task-specific requirements through automatic parameter search, accommodating the unique computational complexity and accuracy needs of autonomous driving. It scales its Multiply-Accumulate Operations (MACs) for Task-Specific Learning Adaptation (TSLA), resulting in alternative configurations tailored to diverse self-driving tasks. These TSLA customizations maximize computational capacity and model accuracy, optimizing hardware utilization.