Box-Level Tube Tracking and Refinement for Vehicles Anomaly Detection

Traffic Anomaly detection is an essential computer vision task and plays a critical role in video structure analysis and urban traffic analysis. In this paper, we propose a box-level tracking and refinement algorithm to identify anomaly detection in road scenes. We first link the detection results to construct candidate spatio-temporal tubes via greedy search. Then the box-level refinement scheme is introduced to employ auxiliary detection cues to promote the abnormal predictions, which consists of spatial fusion, still-thing filter, temporal fusion, and feedforward optimization. Still-thing filter and feedforward optimization employ complementary detection concepts to promote the abnormal predictions, which helps determine an accurate abnormal period. The experimental results show that our approach is superior in the Traffic Anomaly Detection Track test set of the NVIDIA AI CITY 2021 CHALLENGE, which ranked second in this competition, with a 93.18% F1-score and 3.1623 root mean square error It reveals that the proposed approach contributes to fine-grained anomaly detection in actual traffic accident scenarios and promoting the development of intelligent transportation.

1. Introduction

Traffic Anomaly detection is a fundamental computer vision task and plays a critical role in video structure analysis and potential downstream applications, e.g., accident forecasting, urban traffic analysis and evidence investigation. Traffic anomaly detection has been extensively studied in the computer vision field for a long time [6, 11, 7, 5, 10, 3, 4, 19, 16, 22, 20, 23]. These researches have been conducted to leverage a series of statistic patterns to model abnormal concepts, e.g., Hidden Markov Model [7], [5], Markov Random Field [6], [11], sparse reconstruction [10], [3], [21], [9] and autoencoders [4], [19].