SPARK: Low Latency Single-Camera 3D Pose Estimation for Autonomous Racing using Keypoints

2026-06-16

AI Summary

This work addresses the challenge of real-time 3D motion perception of non-cooperative opponents in high-speed autonomous racing, where existing LiDAR-based approaches suffer from high latency and limited deployability on edge devices. To overcome these limitations, the authors propose a lightweight monocular vision-based 3D pose estimation algorithm that uniquely integrates keypoint detection with fixed geometric priors derived from track structure. By optimizing a YOLO-based architecture for efficient keypoint localization and leveraging赛道 geometry to constrain pose inference—particularly for distant targets—the method significantly enhances estimation accuracy. Experimental results on real-world racing datasets demonstrate that the proposed approach outperforms state-of-the-art monocular and LiDAR-based methods, achieving higher precision with lower computational latency.

Abstract

In autonomous racing, fast detection of other participants' movements is required to plan safe, collision-free trajectories with non-cooperative opponents. LiDAR detection is inherently slower and harder to deploy on edge devices than vision methods, causing delayed detections that limit object tracking performance during high-dynamic maneuvering. Utilizing monocular 3D detection enables an easy-to-deploy, low-latency detection of other participants on the racetrack. We present SPARK, a single-camera pose-estimation algorithm for autonomous racing using keypoint detection. It achieves long-range detection with high accuracy, exceeding the performance of state-of-the-art monocular camera detection algorithms while maintaining lower latency. By employing well-optimized YOLO models and leveraging the fixed geometry in the autonomous racing domain, the algorithm also exhibits low latency and resource usage. We evaluate the performance of our approach on real-world autonomous racing data and compare it to state-of-the-art LiDAR and camera detection algorithms. The source code is available at: https://github.com/TUMFTM/SPARK-camera-det

Problem

• autonomous racing
• 3D pose estimation
• monocular camera
• low latency
• object detection

Innovation

• monocular 3D pose estimation
• low latency
• autonomous racing
• keypoint detection
• edge deployment