{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,21]],"date-time":"2026-05-21T05:08:08Z","timestamp":1779340088420,"version":"3.51.4"},"reference-count":36,"publisher":"MDPI AG","issue":"10","license":[{"start":{"date-parts":[[2026,5,8]],"date-time":"2026-05-08T00:00:00Z","timestamp":1778198400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Vision foundation models (VFMs) enable high-quality 2D instance masks, yet their lifted pseudo-point clouds suffer from scale ambiguity, structural noise, and temporal inconsistency, limiting their utility in 3D annotation. Existing automatic labeling methods either rely on expensive light detection and ranging (LiDAR) sensors or fail to enforce physical plausibility in dynamic roadside scenes. This study proposes a LiDAR-free radar\u2013visual auto-labeling framework that leverages cross-modal spatio-temporal consistency between millimeter-wave radar trajectories and visual pseudo-point clouds to self-correct 3D geometry. The method first associates radar points, 2D masks, and pseudo-point clouds into object-centric sequences. Then, an uncertainty-aware pose fusion module combines motion-derived and structure-derived orientations using automatically solved road priors. Finally, the pseudo-point cloud is refined in canonical space by optimizing stable semantic landmarks from temporally consistent masks and propagating their corrections globally. Evaluated on a real-world roadside dataset, the method achieves 49.1% bird\u2019s-eye-view (BEV) intersection over union (IoU) and 43.0% 3D IoU, outperforming a radar\u2013camera fusion baseline by 5.5\/5.9 points. Downstream experiments further show that the generated pseudo-labels and semantic enhancement are useful under the evaluated detector configurations, while broader validation remains future work.<\/jats:p>","DOI":"10.3390\/s26102956","type":"journal-article","created":{"date-parts":[[2026,5,8]],"date-time":"2026-05-08T19:10:52Z","timestamp":1778267452000},"page":"2956","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["LiDAR-Free 3D Auto-Labeling via Radar\u2013Visual Spatio-Temporal Consistency"],"prefix":"10.3390","volume":"26","author":[{"ORCID":"https:\/\/orcid.org\/0009-0000-1989-8998","authenticated-orcid":false,"given":"Boning","family":"Zhu","sequence":"first","affiliation":[{"name":"Beijing Laboratory of Advanced Information Networks, Beijing University of Posts and Telecommunications, Beijing 100876, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5666-1401","authenticated-orcid":false,"given":"Zhiqun","family":"Hu","sequence":"additional","affiliation":[{"name":"Beijing Laboratory of Advanced Information Networks, Beijing University of Posts and Telecommunications, Beijing 100876, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhaoming","family":"Lu","sequence":"additional","affiliation":[{"name":"Beijing Laboratory of Advanced Information Networks, Beijing University of Posts and Telecommunications, Beijing 100876, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2026,5,8]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Dao, M.Q., Caesar, H., Berrio, J.S., Shan, M., Worrall, S., Fr\u00e9mont, V., and Malis, E. (2024, January 2\u20135). Label-efficient 3d object detection for road-side units. Proceedings of the 2024 IEEE Intelligent Vehicles Symposium (IV), Jeju Island, Republic of Korea.","DOI":"10.1109\/IV55156.2024.10588666"},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Engelmann, F., Stuckler, J., and Leibe, B. (2017, January 24\u201331). SAMP: Shape and Motion Priors for 4D Vehicle Reconstruction. Proceedings of the 2017 IEEE Winter Conference on Applications of Computer Vision (WACV), Santa Rosa, CA, USA.","DOI":"10.1109\/WACV.2017.51"},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"Liu, C., Qian, X., Huang, B., Qi, X., Lam, E., Tan, S.C., and Wong, N. (2022). Multimodal transformer for automatic 3d annotation and object detection. Computer Vision\u2014ECCV 2022, Proceedings of the 17th European Conference, Tel Aviv, Israel, 23\u201327 October 2022, Springer.","DOI":"10.1007\/978-3-031-19839-7_38"},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"531","DOI":"10.1007\/s11263-022-01710-9","article-title":"PV-RCNN++: Point-voxel feature set abstraction with local vector representation for 3D object detection","volume":"131","author":"Shi","year":"2023","journal-title":"Int. J. Comput. Vis."},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Ma, T., Yang, X., Zhou, H., Li, X., Shi, B., Liu, J., Yang, Y., Liu, Z., He, L., and Qiao, Y. (2023, January 1\u20136). DetZero: Rethinking Offboard 3D Object Detection with Long-term Sequential Point Clouds. Proceedings of the 2023 IEEE\/CVF International Conference on Computer Vision (ICCV), Paris, France.","DOI":"10.1109\/ICCV51070.2023.00620"},{"key":"ref_6","unstructured":"Yang, B., Bai, M., Liang, M., Zeng, W., and Urtasun, R. (2021). Auto4d: Learning to label 4d objects from sequential point clouds. arXiv."},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Zakharov, S., Kehl, W., Bhargava, A., and Gaidon, A. (2020, January 13\u201319). Autolabeling 3D Objects With Differentiable Rendering of SDF Shape Priors. Proceedings of the 2020 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA.","DOI":"10.1109\/CVPR42600.2020.01224"},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Wang, R., Xu, S., Dai, C., Xiang, J., Deng, Y., Tong, X., and Yang, J. (2025, January 11\u201315). Moge: Unlocking accurate monocular geometry estimation for open-domain images with optimal training supervision. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Nashville, TN, USA.","DOI":"10.1109\/CVPR52734.2025.00496"},{"key":"ref_9","unstructured":"Wang, R., Xu, S., Dong, Y., Deng, Y., Xiang, J., Lv, Z., Sun, G., Tong, X., and Yang, J. (2025). MoGe-2: Accurate Monocular Geometry with Metric Scale and Sharp Details. arXiv."},{"key":"ref_10","first-page":"21875","article-title":"Depth anything v2","volume":"Volume 37","author":"Yang","year":"2024","journal-title":"Advances in Neural Information Processing Systems, Proceedings of the 38th Conference on Neural Information Processing Systems (NeurIPS 2024), Vancouver, BC, Canada, 10\u201315 December 2024"},{"key":"ref_11","unstructured":"Lin, H., Chen, S., Liew, J., Chen, D.Y., Li, Z., Shi, G., Feng, J., and Kang, B. (2025). Depth Anything 3: Recovering the Visual Space from Any Views. arXiv."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"11782","DOI":"10.1109\/TITS.2025.3569399","article-title":"SGV3D: Toward Scenario Generalization for Vision-Based Roadside 3D Object Detection","volume":"26","author":"Yang","year":"2025","journal-title":"IEEE Trans. Intell. Transp. Syst."},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Lin, H., Peng, S., Chen, J., Peng, S., Sun, J., Liu, M., Bao, H., Feng, J., Zhou, X., and Kang, B. (2025, January 11\u201315). Prompting depth anything for 4k resolution accurate metric depth estimation. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Nashville, TN, USA.","DOI":"10.1109\/CVPR52734.2025.01591"},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Zhang, F., Yu, Z., Li, C., Zhang, R., Bai, X., Zhou, Z., Cao, S.Y., Wang, F., and Shen, H.L. (2025, January 19\u201323). Structure-Aware Radar-Camera Depth Estimation. Proceedings of the 2025 IEEE International Conference on Robotics and Automation (ICRA), Atlanta, GA, USA.","DOI":"10.1109\/ICRA55743.2025.11128760"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"1623","DOI":"10.1109\/TPAMI.2020.3019967","article-title":"Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-Shot Cross-Dataset Transfer","volume":"44","author":"Ranftl","year":"2022","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Yang, L., Kang, B., Huang, Z., Xu, X., Feng, J., and Zhao, H. (2024, January 16\u201322). Depth anything: Unleashing the power of large-scale unlabeled data. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.","DOI":"10.1109\/CVPR52733.2024.00987"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"10579","DOI":"10.1109\/TPAMI.2024.3444912","article-title":"Metric3D v2: A Versatile Monocular Geometric Foundation Model for Zero-Shot Metric Depth and Surface Normal Estimation","volume":"46","author":"Hu","year":"2024","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Piccinelli, L., Yang, Y.H., Sakaridis, C., Segu, M., Li, S., Gool, L.V., and Yu, F. (2024, January 16\u201322). UniDepth: Universal Monocular Metric Depth Estimation. Proceedings of the 2024 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA.","DOI":"10.1109\/CVPR52733.2024.00963"},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Nabati, R., and Qi, H. (2021, January 3\u20138). CenterFusion: Center-based Radar and Camera Fusion for 3D Object Detection. Proceedings of the 2021 IEEE Winter Conference on Applications of Computer Vision (WACV), Waikoloa, HI, USA.","DOI":"10.1109\/WACV48630.2021.00157"},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Wu, Z., Chen, G., Gan, Y., Wang, L., and Pu, J. (June, January 29). MVFusion: Multi-View 3D Object Detection with Semantic-Aligned Radar and Camera Fusion. Proceedings of the 2023 IEEE International Conference on Robotics and Automation (ICRA), London, UK.","DOI":"10.1109\/ICRA48891.2023.10161329"},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Kim, Y., Shin, J., Kim, S., Lee, I.J., Choi, J.W., and Kum, D. (2023, January 2\u20136). Crn: Camera radar net for accurate, robust, efficient 3d perception. Proceedings of the IEEE\/CVF International Conference on Computer Vision, Paris, France.","DOI":"10.1109\/ICCV51070.2023.01615"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Lin, Z., Liu, Z., Xia, Z., Wang, X., Wang, Y., Qi, S., Dong, Y., Dong, N., Zhang, L., and Zhu, C. (2024, January 16\u201322). Rcbevdet: Radar-camera fusion in bird\u2019s eye view for 3d object detection. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.","DOI":"10.1109\/CVPR52733.2024.01414"},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Musiat, A., Reichardt, L., Schulze, M., and Wasenm\u00fcller, O. (2024, January 24\u201327). RadarPillars: Efficient Object Detection from 4D Radar Point Clouds. Proceedings of the 2024 IEEE 27th International Conference on Intelligent Transportation Systems (ITSC), Edmonton, AB, Canada.","DOI":"10.1109\/ITSC58415.2024.10919920"},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Long, Y., Morris, D., Liu, X., Castro, M., Chakravarty, P., and Narayanan, P. (2021, January 19\u201325). Radar-Camera Pixel Depth Association for Depth Completion. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Nashville, TN, USA.","DOI":"10.1109\/CVPR46437.2021.01232"},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Li, H., Ma, Y., Gu, Y., Hu, K., Liu, Y., and Zuo, X. (2024, January 13\u201317). RadarCam-Depth: Radar-Camera Fusion for Depth Estimation with Learned Metric Scale. Proceedings of the 2024 IEEE International Conference on Robotics and Automation (ICRA), Yokohama, Japan.","DOI":"10.1109\/ICRA57147.2024.10610929"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"9500418","DOI":"10.1109\/TIM.2025.3647989","article-title":"CalibRefine: Deep Learning-Based Online Automatic Targetless LiDAR-Camera Calibration With Iterative and Attention-Driven Post-Refinement","volume":"75","author":"Cheng","year":"2026","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Cheng, L., and Cao, S. (2023, January 28\u201331). Online Targetless Radar-Camera Extrinsic Calibration Based on the Common Features of Radar and Camera. Proceedings of the NAECON 2023\u2014IEEE National Aerospace and Electronics Conference, Dayton, OH, USA.","DOI":"10.1109\/NAECON58068.2023.10366051"},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Liu, X., Deng, Z., and Zhang, G. (2025). Targetless Radar-Camera Extrinsic Parameter Calibration Using Track-to-Track Association. Sensors, 25.","DOI":"10.3390\/s25030949"},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"15502","DOI":"10.1109\/JSEN.2025.3548274","article-title":"Trajectory-Driven Automatic Extrinsic Calibration for Roadside Radar-Camera Fusion","volume":"25","author":"Zhu","year":"2025","journal-title":"IEEE Sens. J."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"1885","DOI":"10.1109\/JSEN.2024.3493952","article-title":"3-D Multiple Extended Object Tracking by Fusing Roadside Radar and Camera Sensors","volume":"25","author":"Deng","year":"2025","journal-title":"IEEE Sens. J."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"1295","DOI":"10.1109\/TITS.2025.3624716","article-title":"Radar-Camera Fused Multi-Object Tracking: Online Calibration and Common Feature","volume":"27","author":"Cheng","year":"2026","journal-title":"IEEE Trans. Intell. Transp. Syst."},{"key":"ref_32","unstructured":"Ren, T., Liu, S., Zeng, A., Lin, J., Li, K., Cao, H., Chen, J., Huang, X., Chen, Y., and Yan, F. (2024). Grounded sam: Assembling open-world models for diverse visual tasks. arXiv."},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Geiger, A., Lenz, P., and Urtasun, R. (2012, January 16\u201321). Are we ready for autonomous driving? the kitti vision benchmark suite. Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recognition, Providence, RI, USA.","DOI":"10.1109\/CVPR.2012.6248074"},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"567","DOI":"10.1109\/34.24792","article-title":"Principal warps: Thin-plate splines and the decomposition of deformations","volume":"11","author":"Bookstein","year":"2002","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_35","doi-asserted-by":"crossref","unstructured":"Yang, L., Zhang, X., Li, J., Wang, C., Ma, J., Song, Z., Zhao, T., Song, Z., Wang, L., and Zhou, M. (2024). V2x-radar: A multi-modal dataset with 4d radar for cooperative perception. arXiv.","DOI":"10.1038\/s41597-025-04698-2"},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Yang, L., Yu, K., Tang, T., Li, J., Yuan, K., Wang, L., Zhang, X., and Chen, P. (2023, January 17\u201324). Bevheight: A robust framework for vision-based roadside 3d object detection. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Vancouver, BC, Canada.","DOI":"10.1109\/CVPR52729.2023.02070"}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/10\/2956\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,21]],"date-time":"2026-05-21T04:18:06Z","timestamp":1779337086000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/10\/2956"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,8]]},"references-count":36,"journal-issue":{"issue":"10","published-online":{"date-parts":[[2026,5]]}},"alternative-id":["s26102956"],"URL":"https:\/\/doi.org\/10.3390\/s26102956","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,8]]}}}