mlsanityData to 7 Oct 2026

Real-time Accident Anticipation for Autonomous Driving Through Monocular Depth-Enhanced 3D Modeling

arXiv 2409.01256 ↗First posted 2 Sep 2024cs.CV cs.AIcited by 12

Not on the map. Why?

Abstract (arXiv)

The primary goal of traffic accident anticipation is to foresee potential accidents in real time using dashcam videos, a task that is pivotal for enhancing the safety and reliability of autonomous driving technologies. In this study, we introduce an innovative framework, AccNet, which significantly advances the prediction capabilities beyond the current state-of-the-art (SOTA) 2D-based methods by incorporating monocular depth cues for sophisticated 3D scene modeling. Addressing the prevalent challenge of skewed data distribution in traffic accident datasets, we propose the Binary Adaptive Loss for Early Anticipation (BA-LEA). This novel loss function, together with a multi-task learning strategy, shifts the focus of the predictive model towards the critical moments preceding an accident. {We rigorously evaluate the performance of our framework on three benchmark datasets--Dashcam Accident Dataset (DAD), Car Crash Dataset (CCD), and AnAn Accident Detection (A3D), and DADA-2000 Dataset--demonstrating its superior predictive accuracy through key metrics such as Average Precision (AP) and mean Time-To-Accident (mTTA).

Key facts. Cited by 12 papers in the map (data through 7 Oct 2026). Builds on 27 of its 56 references that are in the map. First posted on arXiv 2 Sep 2024.

Builds on

27 of 56 references.

References · 27

  1. 2017Attention Is All You Needcited by 59,932
  2. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  3. 2019MMDetection: Open MMLab Detection Toolbox and Benchmarkcited by 1,379
  4. 2017Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semanticscited by 1,345
  5. 2021Video Swin Transformercited by 1,090
  6. 2021Multiscale Vision Transformerscited by 896
  7. 2023ZoeDepth: Zero-shot Transfer by Combining Relative and Metric Depthcited by 646
  8. 2020Hypersim: A Photorealistic Synthetic Dataset for Holistic Indoor Scene Understandingcited by 593
  9. 2018T-GCN: A Temporal Graph ConvolutionalNetwork for Traffic Predictioncited by 399
  10. 2023Metric3D: Towards Zero-shot Metric 3D Prediction from A Single Imagecited by 357
  11. 2022UniFormerV2: Spatiotemporal Learning by Arming Image ViTs with Video UniFormercited by 116
  12. 2020Uncertainty-based Traffic Accident Anticipation with Spatio-Temporal Relational Learningcited by 91
15 more
  1. 2024World Models for Autonomous Driving: An Initial Surveycited by 79
  2. 2019Unsupervised Traffic Accident Detection in First-Person Videoscited by 69
  3. 2019DADA: Driver Attention Prediction in Driving Accident Scenarioscited by 69
  4. 2018Anticipating Traffic Accidents with Adaptive Loss and Large-scale Incident DBcited by 51
  5. 2021DRIVE: Deep Reinforced Accident Anticipation with Visual Explanationcited by 46
  6. 2021A Dynamic Spatial-temporal Attention Network for Early Anticipation of Traffic Accidentscited by 42
  7. 2021Single Image Depth Estimation: An Overviewcited by 35
  8. 2017Agent-Centric Risk Assessment: Accident Anticipation and Risky Region Localizationcited by 33
  9. 2023Vision-Based Traffic Accident Detection and Anticipation: A Surveycited by 31
  10. 2023GPT-4 Enhanced Multimodal Grounding for Autonomous Driving: Leveraging Cross-Modal Attention with Large Language Modelscited by 31
  11. 2024CRASH: Crash Recognition and Anticipation System Harnessing with Context-Aware and Temporal Focus Attentionscited by 14
  12. 2024MFTraj: Map-Free, Behavior-Driven Trajectory Prediction for Autonomous Drivingcited by 12
  13. 2022No driver, No Regulation? --Online Legal Driving Behavior Monitoring for Self-driving Vehiclescited by 10
  14. 2021Towards explainable artificial intelligence (XAI) for early anticipation of traffic accidentscited by 6
  15. 2024When, Where, and What? A Novel Benchmark for Accident Anticipation and Localization with Large Language Modelscited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 6; 2026: 6.

12 papers.

  1. 2025Nexar Dashcam Collision Prediction Dataset and Challengecited by 21
  2. 2025Minds on the Move: Decoding Trajectory Prediction in Autonomous Driving with Cognitive Insightscited by 10
  3. 2025World Model-Based End-to-End Scene Generation for Accident Anticipation in Autonomous Drivingcited by 7
  4. 2025Domain-Enhanced Dual-Branch Model for Efficient and Interpretable Accident Anticipationcited by 5
  5. 2025Accident Anticipation via Temporal Occurrence Predictioncited by 4
  6. 2026RiskProp: Collision-Anchored Self-Supervised Risk Propagation for Early Accident Anticipationcited by 2
  7. 2025ROAR: Robust Accident Recognition and Anticipation for Autonomous Drivingcited by 0
  8. 2026Learning from the Unseen: Generative Data Augmentation for Geometric-Semantic Accident Anticipationcited by 0
  9. 2026FLaRA: Predicting Future Latent Representations for Accident Anticipationcited by 0
  10. 2026CARA: Concept-Aware Risk Attention for Interpretable Collision Anticipationcited by 0
  11. 2026Observe Before You Alert: Adaptive Driver Alerting with Vision-Language Modelscited by 0
  12. 2026Progressive Risk Estimation for Accident Anticipationcited by 0