arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2501.15045cs.CVcs.AI
收藏

迈向自动驾驶中鲁棒的无监督注意力预测

Towards Robust Unsupervised Attention Prediction in Autonomous Driving

Beijing University of Posts and Telecommunications(北京邮电大学)

Mengshi Qi, Xiaoyang Bi, Pengfei Zhu, Huadong Ma

AI总结:

提出一种鲁棒的无监督注意力预测方法,通过不确定性挖掘分支、知识嵌入模块和新型数据增强方法 RoboMixup 弥合领域差异并提升鲁棒性,同时构建了 DriverAttention-C 基准,性能媲美全监督方法。

摘要:

鲁棒地预测自动驾驶系统的注意力感兴趣区域对驾驶安全至关重要,但由于获取大规模注意力标签的劳动密集性,以及自动驾驶场景与自然场景之间的领域差异,这带来了重大挑战。复杂的交通环境进一步加剧了这些挑战,包括恶劣天气下的摄像头损坏、噪声干扰以及长尾分布带来的中心偏差。为了解决这些问题,我们提出了一种鲁棒的无监督注意力预测方法。Uncertainty Mining Branch(不确定性挖掘分支)通过分析自然场景上多个预训练模型的共性和差异来细化预测,而 Knowledge Embedding Block(知识嵌入模块)通过引入驾驶知识来自适应地增强伪标签,从而弥合领域差异。此外,我们引入了 RoboMixup,这是一种新颖的数据增强方法,通过软注意力和动态增强提高对损坏的鲁棒性,并通过将随机裁剪集成到 Mixup 中作为正则化器来缓解中心偏差。为了系统地评估自动驾驶注意力预测的鲁棒性,我们引入了 DriverAttention-C 基准,包含超过 100k 帧,分为三个子集:BDD-A-C、DR(eye)VE-C 和 DADA-2000-C。我们的方法在三个公共数据集和提出的鲁棒性基准上实现了等同于或超越全监督最先进方法的性能,在 KLD 和 CC 指标上分别将相对损坏退化降低了 58.8% 和 52.8%,并将中心偏差鲁棒性分别提高了 12.4% 和 11.4%。代码和数据可在 https://github.com/zaplm/DriverAttention 获取。

英文摘要:

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.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑