LoRFT: Benchmarking Long-Range Vehicle Trajectory Reconstruction from Fixed Highway Cameras

2026-07-22 · Source: Computer Vision and Pattern Recognition · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Robotics & Autonomous Systems, Computer Vision · Depth: Expert, quick

Summary

LoRFT is introduced as the first open benchmark for long-range vehicle trajectory reconstruction from fixed highway cameras, addressing challenges like perspective compression and scale decay that fragment tracklets. This benchmark comprises 22 expressway surveillance scenes, 366,109 video frames, 6,601 manually verified trajectories, 2,694,889 bounding boxes, road-geometry annotations, and evaluation scripts. Alongside LoRFT, the authors propose Map-RSTNet, a map-aware residual sequence-to-sequence model. Map-RSTNet reconstructs distant trajectories using a road-geometry-aligned state space and dynamically refreshes local road geometry during decoding. On the LoRFT benchmark, Map-RSTNet significantly improves performance, reducing ADE by 11.0%, FDE by 15.4%, and 5-second RMSE by 10.5% relative to the strongest baseline, demonstrating the value of road-geometry-aware reconstruction.

Key takeaway

For Computer Vision Engineers developing traffic monitoring or autonomous driving systems, LoRFT provides a crucial benchmark to validate long-range trajectory reconstruction models. You should consider integrating road-geometry-aware techniques, like those in Map-RSTNet, into your algorithms. This approach can significantly extend the usable range of vehicle trajectory data from existing fixed highway cameras, improving the accuracy of spatio-temporal evidence for safety analysis and traffic management.

Key insights

The LoRFT benchmark and Map-RSTNet model advance long-range vehicle trajectory reconstruction using road-geometry awareness.

Principles

• Road geometry improves distant trajectory reconstruction.
• Perspective effects hinder long-range tracking.
• Benchmarks drive progress in complex vision tasks.

Method

Map-RSTNet reconstructs distant trajectories by aligning them to road geometry in a state space and dynamically refreshing local road geometry during decoding.

In practice

• Utilize LoRFT for evaluating trajectory models.
• Integrate road geometry into tracking algorithms.
• Extend existing camera infrastructure utility.

Topics

• Vehicle Trajectory Reconstruction
• Fixed Highway Cameras
• Computer Vision Benchmarks
• Map-RSTNet
• Road Geometry
• Traffic Safety Analysis

Best for: Research Scientist, AI Scientist, Computer Vision Engineer, Robotics Engineer

Related on AIssential

• LoRFT: Benchmarking Long-Range Vehicle Trajectory Reconstruction from Fixed Highway Cameras — Fixed highway cameras can extend usable trajectory records by reconstructing fragmented far-range tracklets using road-geometry-aware modeling.
• Trajectory-aware Cross-view Geo-localization with Sequential Observations — Fusing sequential video and abstract route descriptions significantly enhances cross-view geo-localization performance.
• RECO: Region-Aware Compensation for Extrinsic Perturbations in Roadside 3D Detection — RECO improves roadside 3D detection by compensating for camera extrinsic perturbations using region-aware pose offsets and a novel reprojection loss.
• TRIG: Trajectory-Rig Decoupled Metric Geometry Learning — Decoupling ego-trajectory and camera-rig components in multi-camera systems improves metric geometry and pose estimation for autonomous driving.
• Geometry-Aware Infrastructure-Anchored Denoiser for UWB Sensing and Work-Zone Reconstruction — Geometry-aware range denoising significantly improves UWB-based work-zone reconstruction accuracy by coupling temporal modeling with spatial consistency.
• Toward Semantic Communication for Real-time Mobile 3D Reconstruction — Semantic communication for real-time 3D reconstruction benefits from pixel-wise confidence maps to guide geometric estimation.

See Counsel's argued verdicts on the open AI decisions leaders are weighing →

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Computer Vision and Pattern Recognition.