Lane detection requires recovering thin, elongated, and frequently occluded lane structures under challenging driving conditions. While anchor-based detectors provide efficient candidate generation, their performance is limited by two coupled issues: backbone features often lose structural continuity along partially visible lanes, and classification confidence may decouple from line-level localization quality, allowing inaccurate anchors to persist before non-maximum suppression (NMS).
arXiv:2608. 09610v1 Announce Type: cross Abstract: Lane detection requires recovering thin, elongated, and frequently occluded lane structures under challenging driving conditions.
arXiv:2610.09118v1 Announce Type: new Abstract: Topology reasoning jointly detects 3D lanes and traffic elements from multi-view images and infers their structural connectivity. Current methods model...
Autonomous driving requires understanding the road as a graph of drivable lanes and their connectivity, beyond the ego lane alone, to follow routes through intersections and reason about cross- and merging-traffic. Recent perception models infer such lane topology, i.
arXiv:2608. 16338v1 Announce Type: cross Abstract: Video lane detection requires predictions that remain stable across frames, yet severe vehicle occlusions can break temporal cues.
The paper surveys one‑stage object detectors for autonomous driving, covering the evolution from early models like YOLOv1 and SSD to recent real‑time architectures such as YOLOv10 and anchor‑free detectors like FCOS and CenterNet. It compares these methods on design choices, feature‑fusion strategies, loss functions, deployment trade‑offs, and benchmark performance, while also summarizing datasets, evaluation metrics, open challenges, and future research directions. The survey emphasizes how one‑stage detectors balance speed, accuracy, efficiency, and robustness, noting the gap between benchmark results and dependable real‑world performance.
arXiv:2609. 28366v1 Announce Type: cross Abstract: Vision-language models (VLMs) offer a promising approach to long-tail autonomous driving, but existing driving datasets provide limited supervision for connecting decision-critical visual evidence with reasoning and planning.
arXiv:2609.05516v1 Announce Type: cross Abstract: Unified perception enables autonomous driving systems to perform object detection, drivable-area segmentation, and lane segmentation within a single...
arXiv:2602. 23499v4 Announce Type: replace-cross Abstract: Collecting a high-quality dataset is a critical task that demands meticulous attention to detail, as overlooking certain aspects can render the entire dataset unusable.
The paper introduces Adaptive Dual-Constrained Line Aggregation (ADLA), a framework that extracts line segments across multiple detection paradigms by aggregating pixels from an edge strength map under orientation coherence and bounded orthogonal distance constraints. ADLA dynamically updates line centroids and orientations, incorporates edge strength into various stages, and reduces parameter tuning. Experiments on generic, wireframe, and Manhattan line segment datasets show strong performance, achieving F^H scores of 0.8665, 0.8720, and 0.7297 respectively.
arXiv:2608. 16480v1 Announce Type: cross Abstract: We present RISE (Roadside Infrastructure Sequence Understanding and Evaluation), a framework spanning metric 3D tracking and structured vision-language reasoning in roadside sequences.
The paper introduces FlexDepth, a family of self‑supervised monocular depth estimation models designed for robust driving perception. FlexDepth uses a two‑stage static‑dynamic decoupled training strategy and a Scale‑Driven Decoder that selects components based on scale size, enabling efficient feature fusion and high‑precision depth maps. Experiments on driving benchmarks show state‑of‑the‑art performance across arbitrary scales with minimal computational cost, with the smallest model (Flex‑Nano) achieving 37.6 FPS on mobile devices.
The paper introduces TopoEnhance, a framework that improves the reliability of scene topology graphs for autonomous driving by treating topology enhancement as a denoising reconstruction task. It learns to recover structural consistency from corrupted ground‑truth graphs, thereby correcting logical inconsistencies and unreliable connections. Experiments show consistent gains in both continuous topology metrics (TOP score) and discrete connectivity (Topology Jaccard Similarity) across multiple baselines without retraining.