Severity: Warning
Message: file_get_contents(https://...@gmail.com&api_key=61f08fa0b96a73de8c900d749fcb997acc09&a=1): Failed to open stream: HTTP request failed! HTTP/1.1 429 Too Many Requests
Filename: helpers/my_audit_helper.php
Line Number: 197
Backtrace:
File: /var/www/html/application/helpers/my_audit_helper.php
Line: 197
Function: file_get_contents
File: /var/www/html/application/helpers/my_audit_helper.php
Line: 271
Function: simplexml_load_file_from_url
File: /var/www/html/application/helpers/my_audit_helper.php
Line: 1075
Function: getPubMedXML
File: /var/www/html/application/helpers/my_audit_helper.php
Line: 3195
Function: GetPubMedArticleOutput_2016
File: /var/www/html/application/controllers/Detail.php
Line: 597
Function: pubMedSearch_Global
File: /var/www/html/application/controllers/Detail.php
Line: 511
Function: pubMedGetRelatedKeyword
File: /var/www/html/index.php
Line: 317
Function: require_once
98%
921
2 minutes
20
Perception plays a crucial role in autonomous driving and computer vision, particularly in interpreting traffic scenes from monocular cameras. In this article, we present a comprehensive collection of traffic scene datasets organized into four distinct groups: (1) Traffic Scene Datasets, (2) Top-View Datasets - both introduced in the authors' earlier research, (3) MultiHeightView Datasets and (4) Depth Datasets. The Traffic Scene Datasets include RealStreet, which captures authentic traffic scenarios, and SynthStreet, its synthetic counterpart. To enhance the diversity of available data, we also introduce the SupporterReal subset, which augments real datasets that lack a corresponding synthetic twin. In addition, SupporterVirtual offers a virtual environment without a direct real-world equivalent. For top-view perspectives, the datasets RealTop and SynthTop provide real and synthetic traffic scene data, respectively. The MultiHeightView dataset provides synchronized RGB image sequences of urban traffic scenes captured from three fixed camera heights: 0.5 m, 1.0 m, and 1.5 m designed to support zero-shot monocular depth model. Depth dataset provides synchronized RGB, and depth images of real-world urban traffic scenes collected using RealSense stereo camera. A perceptual similarity analysis using the LPIPS metric confirms that synthetic images closely resemble real-world scenes, supporting their use in synthetic-to-real domain transfer learning. Data collection and generation were conducted between 2023 and 2025, utilizing standardized formats to ensure compatibility with deep learning frameworks and to promote broader reuse in computer vision applications.
Download full-text PDF |
Source |
---|---|
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC12398884 | PMC |
http://dx.doi.org/10.1016/j.dib.2025.111976 | DOI Listing |