Detecting What Matters: A Novel Approach for Out-of-Distribution 3D Object Detection in Autonomous Vehicles

• URL: http://arxiv.org/abs/2506.23426v1
• Date: Sun, 29 Jun 2025 23:21:05 GMT
• Title: Detecting What Matters: A Novel Approach for Out-of-Distribution 3D Object Detection in Autonomous Vehicles
• Authors: Menna Taha, Aya Ahmed, Mohammed Karmoose, Yasser Gadallah,
• Abstract summary: We propose a novel object detection approach that shifts the emphasis from conventional class-based classification to object harmfulness determination.<n>Our method identifies them as either 'harmful' or 'harmless' based on whether they pose a danger to the AV.<n>Our results demonstrate that the proposed model effectively detects OOD objects, evaluates their harmfulness, and classifies them accordingly.
• Score: 1.4524462132789562
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Autonomous vehicles (AVs) use object detection models to recognize their surroundings and make driving decisions accordingly. Conventional object detection approaches classify objects into known classes, which limits the AV's ability to detect and appropriately respond to Out-of-Distribution (OOD) objects. This problem is a significant safety concern since the AV may fail to detect objects or misclassify them, which can potentially lead to hazardous situations such as accidents. Consequently, we propose a novel object detection approach that shifts the emphasis from conventional class-based classification to object harmfulness determination. Instead of object detection by their specific class, our method identifies them as either 'harmful' or 'harmless' based on whether they pose a danger to the AV. This is done based on the object position relative to the AV and its trajectory. With this metric, our model can effectively detect previously unseen objects to enable the AV to make safer real-time decisions. Our results demonstrate that the proposed model effectively detects OOD objects, evaluates their harmfulness, and classifies them accordingly, thus enhancing the AV decision-making effectiveness in dynamic environments.

Related papers

• A Real-Time Defense Against Object Vanishing Adversarial Patch Attacks for Object Detection in Autonomous Vehicles [0.0]
  ADAV (Adversarial Defense for Autonomous Vehicles) is a novel defense methodology against object vanishing patch attacks.<n>ADAV runs in real-time and leverages contextual information from prior frames in an AV's video feed.<n>ADAV is evaluated using real-world driving data from the Berkeley Deep Drive BDD100K dataset.
  arXiv  Detail & Related papers  (2024-12-09T05:21:14Z)
• Uncertainty Estimation for 3D Object Detection via Evidential Learning [63.61283174146648]
  We introduce a framework for quantifying uncertainty in 3D object detection by leveraging an evidential learning loss on Bird's Eye View representations in the 3D detector. We demonstrate both the efficacy and importance of these uncertainty estimates on identifying out-of-distribution scenes, poorly localized objects, and missing (false negative) detections.
  arXiv  Detail & Related papers  (2024-10-31T13:13:32Z)
• Revisiting Out-of-Distribution Detection in LiDAR-based 3D Object Detection [12.633311483061647]
  Out-of-distribution (OOD) objects can lead to misclassifications, posing a significant risk to the safety and reliability of automated vehicles. We propose a new evaluation protocol that allows the use of existing datasets without modifying the point cloud. The effectiveness of our method is validated through experiments on the newly proposed nuScenes OOD benchmark.
  arXiv  Detail & Related papers  (2024-04-24T13:48:38Z)
• Detecting Out-of-Distribution Objects through Class-Conditioned Inpainting [42.67743584239442]
  Real-world deployment often introduces novel and unexpected objects, referred to as out-of-distribution (OOD) objects.<n>Modern object detectors are typically overconfident, making it unreliable to use their predictions alone for OOD detection.<n>We utilize an off-the-shelf text-to-image generative model, such as Stable Diffusion, which is trained with objective functions distinct from those of discriminative object detectors.
  arXiv  Detail & Related papers  (2024-02-05T18:50:27Z)
• Diffusion-Based Particle-DETR for BEV Perception [94.88305708174796]
  Bird-Eye-View (BEV) is one of the most widely-used scene representations for visual perception in Autonomous Vehicles (AVs) Recent diffusion-based methods offer a promising approach to uncertainty modeling for visual perception but fail to effectively detect small objects in the large coverage of the BEV. Here, we address this problem by combining the diffusion paradigm with current state-of-the-art 3D object detectors in BEV.
  arXiv  Detail & Related papers  (2023-12-18T09:52:14Z)
• SalienDet: A Saliency-based Feature Enhancement Algorithm for Object Detection for Autonomous Driving [160.57870373052577]
  We propose a saliency-based OD algorithm (SalienDet) to detect unknown objects. Our SalienDet utilizes a saliency-based algorithm to enhance image features for object proposal generation. We design a dataset relabeling approach to differentiate the unknown objects from all objects in training sample set to achieve Open-World Detection.
  arXiv  Detail & Related papers  (2023-05-11T16:19:44Z)
• PROB: Probabilistic Objectness for Open World Object Detection [15.574535196804042]
  Open World Object Detection (OWOD) is a new computer vision task that bridges the gap between classic object detection (OD) benchmarks and object detection in the real world. We introduce a novel probabilistic framework for objectness estimation, where we alternate between probability distribution estimation and objectness likelihood of known objects. The resulting Probabilistic Objectness transformer-based open-world detector, PROB, integrates our framework into traditional object detection models.
  arXiv  Detail & Related papers  (2022-12-02T20:04:24Z)
• Evaluating Object (mis)Detection from a Safety and Reliability Perspective: Discussion and Measures [1.8492669447784602]
  We propose new object detection measures that reward the correct identification of objects that are most dangerous and most likely to affect driving decisions. We apply our model on the recent autonomous driving dataset nuScenes, and we compare nine object detectors. Results show that, in several settings, object detectors that perform best according to the nuScenes ranking are not the preferable ones when the focus is shifted on safety and reliability.
  arXiv  Detail & Related papers  (2022-03-04T09:31:20Z)
• ObjectSeeker: Certifiably Robust Object Detection against Patch Hiding Attacks via Patch-agnostic Masking [95.6347501381882]
  Object detectors are found to be vulnerable to physical-world patch hiding attacks. We propose ObjectSeeker as a framework for building certifiably robust object detectors.
  arXiv  Detail & Related papers  (2022-02-03T19:34:25Z)
• Robust and Accurate Object Detection via Adversarial Learning [111.36192453882195]
  This work augments the fine-tuning stage for object detectors by exploring adversarial examples. Our approach boosts the performance of state-of-the-art EfficientDets by +1.1 mAP on the object detection benchmark.
  arXiv  Detail & Related papers  (2021-03-23T19:45:26Z)
• Slender Object Detection: Diagnoses and Improvements [74.40792217534]
  In this paper, we are concerned with the detection of a particular type of objects with extreme aspect ratios, namely textbfslender objects. For a classical object detection method, a drastic drop of $18.9%$ mAP on COCO is observed, if solely evaluated on slender objects.
  arXiv  Detail & Related papers  (2020-11-17T09:39:42Z)

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.