Searcharxiv⌕ Search
Home/Archive/arXiv

arXiv · 2611.00066

Comparing Monocular and Multi-View 3D Pedestrian Reconstruction Strategies for Urban Intersection Sensing

Kauan Mariani·Matheus Carvalho·Joel Perca·João Rulff·Jorge Poco
View original source ↗Read source document ↗Export metadata ↓

Abstract

Understanding how pedestrians move and interact at urban intersections is central to traffic-safety assessment, mobility planning, and smart-city sensing. Such analyses rely on physically grounded descriptions of motion: 3D body pose, metric trajectories, and pose-derived descriptors such as stature and walking speed. Recovering these from fixed roadside cameras is difficult: viewpoints are oblique and long-range, occlusions are frequent, and neither ground-truth 3D pose nor cross-camera identity labels are typically available. We compare two strategies for deriving pedestrian motion descriptors from a multi-camera intersection dataset: a monocular learned pipeline that combines text-prompted segmentation, tracking, and learned 3D body estimation, and a calibrated multi-view geometric pipeline that combines 2D pose estimation, tracking, cross-camera association, and triangulation. Because 3D ground truth is unavailable, we evaluate both with reference-free criteria relevant to intersection analysis: pedestrian coverage, anthropometric plausibility, walking-speed distributions, computational cost, and failure modes. The two strategies exhibit complementary trade-offs. The learned pipeline is simpler to deploy and covers more pedestrians, but suffers from scale ambiguity and systematic height compression; the geometric pipeline yields metric reconstructions more efficiently, but requires calibration, discards pedestrians seen by only one camera, and can produce severe outliers in dense crowds. We distill these findings into practical guidance for selecting a reconstruction strategy and configuring camera infrastructure according to the target analysis task.

Explore related subjects

cs.CV

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Kauan Mariani, Matheus Carvalho, Joel Perca, João Rulff, Jorge Poco. 2026-10-14. Comparing Monocular and Multi-View 3D Pedestrian Reconstruction Strategies for Urban Intersection Sensing. https://arxiv.org/abs/2611.00066

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Graph-Aggregated Prototype Learning for Remote Sensing Semantic Change Detection

Semantic change detection (SCD) extends the binary change detection task to provide not only the change locations but also the detailed "from-to" categories in multi-temporal remote sensing data. Such detailed semantic insights into changes offer considerable advantages for a wide array of applications. However, since SCD involves the simultaneous optimization of multiple tasks, the model is prone to negative transfer due to task-specific learning difficulties and conflicting gradient flows. To address this issue, we propose Graph-Aggregated Prototype Learning for Remote Sensing Semantic Change Detection(GAPL-SCD). In this framework, a multi-task joint optimization method is designed to optimize the primary task of semantic segmentation and change detection, along with the auxiliary task of graph aggregation prototype learning. Adaptive weight allocation and gradient rotation methods are used to alleviate the conflict between training tasks and improve multi-task learning capabilities. Specifically, the graph aggregation prototype learning module constructs an interaction graph using high-level features. Prototypes serve as class proxies, enabling category-level domain alignment across time points and reducing interference from irrelevant changes. Additionally, the proposed self-query multi-level feature interaction and bi-temporal feature fusion modules further enhance multi-scale feature representation, improving performance in complex scenes. Experimental results on the SECOND and Landsat-SCD datasets demonstrate that our method achieves state-of-the-art performance, with significant improvements in accuracy and robustness for SCD task.

cs.CV↗
arXiv2026

Compact Attention: Exploiting Structured Spatio-Temporal Sparsity for Fast Video Generation

The quadratic computational complexity of self-attention mechanisms imposes a critical constraint on transformer-based video generation, particularly for synthesizing ultra-long and high-definition sequences. While most sparse attention techniques employ fixed or heuristic patterns, they often struggle to accommodate the heterogeneous and layer-specific redundancies inherent in video diffusion transformers (VDiTs). We find that different attention heads consistently concentrate most of their attention mass in different coarse query-key regions, a behavior we call functional specialization. Although individual attention values vary with the generated content, these regions remain similar across prompts and random seeds. Leveraging these insights, we introduce Compact Attention, a hardware-aware sparse attention framework for efficient long-video synthesis. We quantify this stability and use it to construct searchable, reusable, head-specific masks. Based on the observation, we propose a deformable dual-window sparse pattern mechanism that flexibly approximates local and axial attention regions within one tile-aligned mask family. To avoid manually choosing a mask for each head, we develop a recall-driven offline auto-search algorithm that identifies high-quality sparse masks while leveraging temporal stability for mask reuse. Extensive evaluations on Wan2.1 (14B) and Hunyuan-Video demonstrate that Compact Attention achieves 2.25x/3.03x attention speedups over FlashAttention-2 and 1.20x/1.74x over FlashAttention-3, respectively, while maintaining visual quality and temporal consistency comparable to full-attention baselines. Project Page: https://yo-ava.github.io/Compact-Attention.github.io/

cs.CV↗
arXiv2026

Backpropagation-Free Test-Time Adaptation via Probabilistic Gaussian Alignment

Test-time adaptation (TTA) enhances the zero-shot robustness under distribution shifts by leveraging unlabeled test data during inference. Despite notable advances, several challenges still limit its broader applicability. First, most methods rely on backpropagation or iterative optimization, which limits scalability and hinders real-time deployment. Second, they lack explicit modeling of class-conditional feature distributions. This modeling is crucial for producing reliable decision boundaries and calibrated predictions, but it remains underexplored due to the lack of both source data and supervision at test time. In this paper, we propose ADAPT, an Advanced Distribution-Aware and backPropagation-free Test-time adaptation method. We reframe TTA as a Gaussian probabilistic inference task by modeling class-conditional likelihoods using gradually updated class means and a shared covariance matrix. This enables closed-form, training-free inference. To correct potential likelihood bias, we introduce lightweight regularization guided by CLIP priors and a historical knowledge bank. ADAPT requires no source data, no gradient updates, and no full access to target data, supporting both online and transductive settings. Extensive experiments across diverse benchmarks demonstrate that our method achieves state-of-the-art performance under a wide range of distribution shifts with superior scalability and robustness.

cs.CV↗