Unsupervised Detection of Entry and Exit Regions from Vehicle Trajectories for Camera-Agnostic Turning Movement Counts

2026-07-12

AI Summary

This work addresses the high cost and limited scalability of traditional turning movement count methods, which rely on manual annotation and camera calibration. The authors propose an unsupervised pipeline that automatically generates persistent polygonal entry and exit zones solely from vehicle trajectory start and end points, leveraging spatial clustering without requiring human intervention, camera calibration, or prior intersection knowledge. This approach introduces reusable spatial regions for the first time, enabling cross-camera deployment and substantially reducing computational overhead. Integrated with object detection, multi-object tracking, and systematic parameter optimization, the method achieves efficient trajectory classification. Evaluated on 25 real-world cameras—including 16 unseen scenarios—it attains a median classification error of approximately 3% and meets engineering standards under the GEH metric, demonstrating superior stability and efficiency compared to existing baselines.

Abstract

Turning movement counts are essential for intersection-level traffic management, yet their collection remains predominantly manual due to the cost of per-camera region annotation. This paper presents an unsupervised pipeline that identifies entry and exit regions directly from raw vehicle trajectories extracted via object detection and multi-object tracking, requiring no manual annotation, camera calibration, or prior knowledge of intersection geometry. Unlike trajectory clustering methods that classify individual trajectories using pairwise similarity and must be re-executed on every new batch, the proposed pipeline clusters initial and terminal point locations to produce persistent spatial region polygons that classify future trajectories by point-in-polygon containment at linear cost. The pipeline comprises six sequential steps, each with configurable parameters evaluated through a systematic statistical analysis spanning 19,152 pipeline executions across 25 surveillance cameras capturing dense heterogeneous traffic in Bengaluru, India, and 10 sequences from the UA-DETRAC benchmark dataset. Both parametric and nonparametric testing frameworks identify three consistently significant parameters and yield an empirically grounded recommended configuration. Under this configuration, the pipeline achieves a median classification error of approximately 3% across all 25 cameras, including 16 held-out locations, with GEH values within accepted engineering thresholds. Compared with two trajectory clustering baselines, the proposed pipeline exhibits greater stability across camera views and lower computational cost, at the expense of higher median error. Extended evaluation demonstrates that calibration clips of at least 60 minutes and peak-traffic selection further improve region estimation quality.

Problem

• turning movement counts
• entry and exit regions
• unsupervised detection
• vehicle trajectories
• camera-agnostic

Innovation

• unsupervised region detection
• trajectory-based turning movement counts
• camera-agnostic traffic analysis
• spatial polygon clustering
• persistent entry-exit regions