PoseDriver: A Unified Approach to Multi-Category Skeleton Detection for Autonomous Driving

2026-03-24

AI Summary

This work addresses the absence of a unified architecture for simultaneous skeleton detection of diverse object categories—such as lane markings and bicycles—in autonomous driving scenarios. To this end, we propose PoseDriver, a bottom-up, unified framework for multi-category skeleton detection that treats each object class as an independent task and achieves efficient structural perception using only image inputs. Our study introduces the first unified skeletal representation capable of modeling heterogeneous objects like lanes and bicycles, presents a novel skeleton-based approach to lane detection, and releases the first bicycle skeleton dataset. Experiments demonstrate that PoseDriver achieves state-of-the-art performance on OpenLane for lane skeleton detection and exhibits strong cross-category transferability and generalization capabilities.

Technology Category

• Computer Vision: Object Detection & Categorization
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Planning, Routing, and Scheduling: Activity and Plan Recognition

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Object skeletons offer a concise representation of structural information, capturing essential aspects of posture and orientation that are crucial for autonomous driving applications. However, a unified architecture that simultaneously handles multiple instances and categories using only the input image remains elusive. In this paper, we introduce PoseDriver, a unified framework for bottom-up multi-category skeleton detection tailored to common objects in driving scenarios. We model each category as a distinct task to systematically address the challenges of multi-task learning. Specifically, we propose a novel approach for lane detection based on skeleton representations, achieving state-of-the-art performance on the OpenLane dataset. Moreover, we present a new dataset for bicycle skeleton detection and assess the transferability of our framework to novel categories. Experimental results validate the effectiveness of the proposed approach.

Problem

• skeleton detection
• multi-category
• autonomous driving
• unified architecture
• object representation

Innovation

• unified skeleton detection
• multi-category pose estimation
• bottom-up parsing
• lane skeleton representation
• transferable framework