Towards Robust Unsupervised Attention Prediction in Autonomous Driving

• URL: http://arxiv.org/abs/2501.15045v2
• Date: Wed, 29 Jan 2025 03:43:00 GMT
• Title: Towards Robust Unsupervised Attention Prediction in Autonomous Driving
• Authors: Mengshi Qi, Xiaoyang Bi, Pengfei Zhu, Huadong Ma,
• Abstract summary: We propose a robust unsupervised attention prediction method for self-driving systems.<n>An Uncertainty Mining Branch refines predictions by analyzing commonalities and differences across multiple pre-trained models on natural scenes.<n>A Knowledge Embedding Block bridges the domain gap by incorporating driving knowledge to adaptively enhance pseudo-labels.<n>A novel data augmentation method improves robustness against corruption through soft attention and dynamic augmentation.
• Score: 40.84001015982244
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Robustly predicting attention regions of interest for self-driving systems is crucial for driving safety but presents significant challenges due to the labor-intensive nature of obtaining large-scale attention labels and the domain gap between self-driving scenarios and natural scenes. These challenges are further exacerbated by complex traffic environments, including camera corruption under adverse weather, noise interferences, and central bias from long-tail distributions. To address these issues, we propose a robust unsupervised attention prediction method. An Uncertainty Mining Branch refines predictions by analyzing commonalities and differences across multiple pre-trained models on natural scenes, while a Knowledge Embedding Block bridges the domain gap by incorporating driving knowledge to adaptively enhance pseudo-labels. Additionally, we introduce RoboMixup, a novel data augmentation method that improves robustness against corruption through soft attention and dynamic augmentation, and mitigates central bias by integrating random cropping into Mixup as a regularizer. To systematically evaluate robustness in self-driving attention prediction, we introduce the DriverAttention-C benchmark, comprising over 100k frames across three subsets: BDD-A-C, DR(eye)VE-C, and DADA-2000-C. Our method achieves performance equivalent to or surpassing fully supervised state-of-the-art approaches on three public datasets and the proposed robustness benchmark, reducing relative corruption degradation by 58.8% and 52.8%, and improving central bias robustness by 12.4% and 11.4% in KLD and CC metrics, respectively. Code and data are available at https://github.com/zaplm/DriverAttention.

Related papers

• OCCUQ: Exploring Efficient Uncertainty Quantification for 3D Occupancy Prediction [9.742801351723482]
  Current methods often overlook uncertainties arising from adversarial conditions or distributional shifts. We propose an efficient adaptation of an uncertainty estimation technique for 3D occupancy prediction. Our approach consistently demonstrates reliable uncertainty measures, indicating its potential for enhancing the robustness of autonomous driving systems.
  arXiv  Detail & Related papers  (2025-03-13T17:50:07Z)
• Anomaly Detection in Cooperative Vehicle Perception Systems under Imperfect Communication [4.575903181579272]
  We propose a cooperative-perception-based anomaly detection framework (CPAD) CPAD is a robust architecture that remains effective under communication interruptions. Empirical results demonstrate that our approach outperforms standard anomaly classification methods in F1-score, AUC.
  arXiv  Detail & Related papers  (2025-01-28T22:41:06Z)
• Traffic and Safety Rule Compliance of Humans in Diverse Driving Situations [48.924085579865334]
  Analyzing human data is crucial for developing autonomous systems that replicate safe driving practices. This paper presents a comparative evaluation of human compliance with traffic and safety rules across multiple trajectory prediction datasets.
  arXiv  Detail & Related papers  (2024-11-04T09:21:00Z)
• DSDFormer: An Innovative Transformer-Mamba Framework for Robust High-Precision Driver Distraction Identification [23.05821759499963]
  Driver distraction remains a leading cause of traffic accidents, posing a critical threat to road safety globally. We propose DSDFormer, a framework that integrates the strengths of Transformer and Mamba architectures. We also introduce Temporal Reasoning Confident Learning (TRCL), an unsupervised approach that refines noisy labels by leveragingtemporal correlations in video.
  arXiv  Detail & Related papers  (2024-09-09T13:16:15Z)
• The RoboDrive Challenge: Drive Anytime Anywhere in Any Condition [136.32656319458158]
  The 2024 RoboDrive Challenge was crafted to propel the development of driving perception technologies. This year's challenge consisted of five distinct tracks and attracted 140 registered teams from 93 institutes across 11 countries. The competition culminated in 15 top-performing solutions.
  arXiv  Detail & Related papers  (2024-05-14T17:59:57Z)
• Edge-Assisted ML-Aided Uncertainty-Aware Vehicle Collision Avoidance at Urban Intersections [12.812518632907771]
  We present a novel framework that detects preemptively collisions at urban crossroads. We exploit the Multi-access Edge Computing platform of 5G networks.
  arXiv  Detail & Related papers  (2024-04-22T18:45:40Z)
• Leveraging Driver Field-of-View for Multimodal Ego-Trajectory Prediction [69.29802752614677]
  RouteFormer is a novel ego-trajectory prediction network combining GPS data, environmental context, and the driver's field-of-view. To tackle data scarcity and enhance diversity, we introduce GEM, a dataset of urban driving scenarios enriched with synchronized driver field-of-view and gaze data.
  arXiv  Detail & Related papers  (2023-12-13T23:06:30Z)
• Unsupervised Self-Driving Attention Prediction via Uncertainty Mining and Knowledge Embedding [51.8579160500354]
  We propose an unsupervised way to predict self-driving attention by uncertainty modeling and driving knowledge integration. Results show equivalent or even more impressive performance compared to fully-supervised state-of-the-art approaches.
  arXiv  Detail & Related papers  (2023-03-17T00:28:33Z)
• Cognitive Accident Prediction in Driving Scenes: A Multimodality Benchmark [77.54411007883962]
  We propose a Cognitive Accident Prediction (CAP) method that explicitly leverages human-inspired cognition of text description on the visual observation and the driver attention to facilitate model training. CAP is formulated by an attentive text-to-vision shift fusion module, an attentive scene context transfer module, and the driver attention guided accident prediction module. We construct a new large-scale benchmark consisting of 11,727 in-the-wild accident videos with over 2.19 million frames.
  arXiv  Detail & Related papers  (2022-12-19T11:43:02Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.