{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,13]],"date-time":"2026-04-13T08:05:06Z","timestamp":1776067506622,"version":"3.50.1"},"reference-count":44,"publisher":"MDPI AG","issue":"8","license":[{"start":{"date-parts":[[2026,4,11]],"date-time":"2026-04-11T00:00:00Z","timestamp":1775865600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100012166","name":"National Key Research and Development Program of China","doi-asserted-by":"publisher","award":["2024YFE0200700"],"award-info":[{"award-number":["2024YFE0200700"]}],"id":[{"id":"10.13039\/501100012166","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100012166","name":"National Key Research and Development Program of China","doi-asserted-by":"publisher","award":["2024YFE0200703"],"award-info":[{"award-number":["2024YFE0200703"]}],"id":[{"id":"10.13039\/501100012166","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Specific Research Fund of the Innovation Platform for Academicians of Hainan Province","award":["YSPTZX202314"],"award-info":[{"award-number":["YSPTZX202314"]}]},{"name":"Shanghai Key Research Laboratory of NSAI"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Spatial intelligence in autonomous driving requires object-level 3D geometry, yet existing monocular mesh reconstruction methods usually operate with a fixed inference path and a single mesh parameterization, which limits their flexibility under heterogeneous resource constraints. To address this issue, we propose DyPRSI, a dynamic-parameterized framework for monocular vehicle 3D reconstruction that provides multiple predefined accuracy\u2013latency operating points within a single model. DyPRSI inserts two early exits into a shared Res2Net\u2013BiFPN trunk and associates each exit with an exit-specific mesh specification, forming a coarse-to-fine reconstruction hierarchy across network depth. To better match the efficiency requirements of shallow branches, DyPRSI adopts lightweight coordinate-classification keypoint decoding for EE1 and EE2, while retaining a heatmap-regression keypoint head in the Main branch to preserve the upper bound of reconstruction accuracy. Experiments on ApolloCar3D show that DyPRSI-Main achieves competitive reconstruction performance, whereas EE1 and EE2 substantially reduce end-to-end inference latency and provide useful alternatives under different resource requirements. Ablation studies further show that the speedup mainly comes from the lightweight branch-specific keypoint heads, while the exit-specific mesh settings help organize stable coarse-to-fine reconstruction behavior across branches. These results indicate that DyPRSI is a practical monocular vehicle reconstruction framework for resource-aware spatial intelligence.<\/jats:p>","DOI":"10.3390\/s26082355","type":"journal-article","created":{"date-parts":[[2026,4,13]],"date-time":"2026-04-13T07:29:29Z","timestamp":1776065369000},"page":"2355","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Dynamic-Parameterized Reconstruction Model for Resource-Aware Spatial Intelligence"],"prefix":"10.3390","volume":"26","author":[{"ORCID":"https:\/\/orcid.org\/0009-0004-2808-5496","authenticated-orcid":false,"given":"Hongyi","family":"Huang","sequence":"first","affiliation":[{"name":"College of Intelligent Robotics and Advanced Manufacturing, Fudan University, Shanghai 200433, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yanni","family":"Zhang","sequence":"additional","affiliation":[{"name":"Shanghai East-Bund Institute on Networking Systems of AI, Shanghai 202162, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8143-9052","authenticated-orcid":false,"given":"Liang","family":"Song","sequence":"additional","affiliation":[{"name":"College of Intelligent Robotics and Advanced Manufacturing, Fudan University, Shanghai 200433, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhen","family":"Zhao","sequence":"additional","affiliation":[{"name":"China State Construction Engineering (Hong Kong) Limited, Hong Kong, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xiaopeng","family":"Yang","sequence":"additional","affiliation":[{"name":"China State Construction Engineering (Hong Kong) Limited, Hong Kong, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2026,4,11]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Rosinol, A., Abate, M., Chang, Y., and Carlone, L. (2020). Kimera: An open-source library for real-time metric-semantic localization and mapping. Proceedings of the 2020 IEEE International Conference on Robotics and Automation (ICRA), IEEE.","DOI":"10.1109\/ICRA40945.2020.9196885"},{"key":"ref_2","unstructured":"Wang, Z. (2024). 3D representation methods: A survey. arXiv."},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"Mehta, V., Sharma, C., and Thiyagarajan, K. (2025). Large language models and 3D vision for intelligent robotic perception and autonomy. Sensors, 25.","DOI":"10.3390\/s25206394"},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Wu, X., Jiang, L., Wang, P.S., Liu, Z., Liu, X., Qiao, Y., Ouyang, W., He, T., and Zhao, H. (2024, January 16\u201322). Point transformer v3: Simpler faster stronger. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.","DOI":"10.1109\/CVPR52733.2024.00463"},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Chen, Y., Liu, J., Zhang, X., Qi, X., and Jia, J. (2023, January 17\u201324). Voxelnext: Fully sparse voxelnet for 3d object detection and tracking. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Vancouver, BC, Canada.","DOI":"10.1109\/CVPR52729.2023.02076"},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Li, Z., M\u00fcller, T., Evans, A., Taylor, R.H., Unberath, M., Liu, M.Y., and Lin, C.H. (2023, January 17\u201324). Neuralangelo: High-fidelity neural surface reconstruction. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Vancouver, BC, Canada.","DOI":"10.1109\/CVPR52729.2023.00817"},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Zhang, C., Delitzas, A., Wang, F., Zhang, R., Ji, X., Pollefeys, M., and Engelmann, F. (2025, January 11\u201315). Open-vocabulary functional 3d scene graphs for real-world indoor spaces. Proceedings of the Computer Vision and Pattern Recognition Conference, Nashville, TN, USA.","DOI":"10.1109\/CVPR52734.2025.01807"},{"key":"ref_8","unstructured":"Renz, K., Chitta, K., Mercea, O.B., Koepke, A., Akata, Z., and Geiger, A. (2022). Plant: Explainable planning transformers via object-level representations. arXiv."},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Jiang, B., Chen, S., Xu, Q., Liao, B., Chen, J., Zhou, H., Zhang, Q., Liu, W., Huang, C., and Wang, X. (2023, January 1\u20136). Vad: Vectorized scene representation for efficient autonomous driving. Proceedings of the IEEE\/CVF International Conference on Computer Vision, Paris, France.","DOI":"10.1109\/ICCV51070.2023.00766"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"100279","DOI":"10.1016\/j.geits.2025.100279","article-title":"Multi-objective autonomous eco-driving strategy: A pathway to future green mobility","volume":"4","author":"Tong","year":"2025","journal-title":"Green Energy Intell. Transp."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"100156","DOI":"10.1016\/j.geits.2024.100156","article-title":"A review on reinforcement learning-based highway autonomous vehicle control","volume":"3","author":"Irshayyid","year":"2024","journal-title":"Green Energy Intell. Transp."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"100288","DOI":"10.1016\/j.geits.2025.100288","article-title":"The investigation of reinforcement learning-based end-to-end decision-making algorithms for autonomous driving on the road with consecutive sharp turns","volume":"4","author":"Li","year":"2025","journal-title":"Green Energy Intell. Transp."},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Jia, D., Ruan, X., Xia, K., Zou, Z., Wang, L., and Tang, W. (2024). Analysis-by-synthesis transformer for single-view 3d reconstruction. Proceedings of the European Conference on Computer Vision, Springer.","DOI":"10.1007\/978-3-031-72664-4_15"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"3746","DOI":"10.1109\/TIP.2023.3279661","article-title":"Single-view 3D mesh reconstruction for seen and unseen categories","volume":"32","author":"Yang","year":"2023","journal-title":"IEEE Trans. Image Process."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Long, X., Guo, Y.C., Lin, C., Liu, Y., Dou, Z., Liu, L., Ma, Y., Zhang, S.H., Habermann, M., and Theobalt, C. (2024, January 16\u201322). Wonder3d: Single image to 3d using cross-domain diffusion. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.","DOI":"10.1109\/CVPR52733.2024.00951"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Chabot, F., Chaouch, M., Rabarisoa, J., Teuliere, C., and Chateau, T. (2017, January 21\u201326). Deep manta: A coarse-to-fine many-task network for joint 2d and 3d vehicle analysis from monocular image. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA.","DOI":"10.1109\/CVPR.2017.198"},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Ke, L., Li, S., Sun, Y., Tai, Y.W., and Tang, C.K. (2020). Gsnet: Joint vehicle pose and shape reconstruction with geometrical and scene-aware supervision. Proceedings of the European Conference on Computer Vision, Springer.","DOI":"10.1007\/978-3-030-58555-6_31"},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Lee, H.J., Kim, H., Choi, S.M., Jeong, S.G., and Koh, Y.J. (2023, January 17\u201324). BAAM: Monocular 3D pose and shape reconstruction with bi-contextual attention module and attention-guided modeling. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Vancouver, BC, Canada.","DOI":"10.1109\/CVPR52729.2023.00870"},{"key":"ref_19","first-page":"3812","article-title":"Learning anytime predictions in neural networks via adaptive loss balancing","volume":"Volume 33","author":"Hu","year":"2019","journal-title":"Proceedings of the AAAI Conference on Artificial Intelligence, Honolulu, HI, USA, 27 January\u20131 February 2019"},{"key":"ref_20","unstructured":"Lee, H., and Shin, J. (2018). Anytime neural prediction via slicing networks vertically. arXiv."},{"key":"ref_21","first-page":"27631","article-title":"Improving anytime prediction with parallel cascaded networks and a temporal-difference loss","volume":"34","author":"Iuzzolino","year":"2021","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_22","unstructured":"Huang, G., Chen, D., Li, T., Wu, F., Van Der Maaten, L., and Weinberger, K.Q. (2017). Multi-scale dense networks for resource efficient image classification. arXiv."},{"key":"ref_23","first-page":"18330","article-title":"Bert loses patience: Fast and robust inference with early exit","volume":"33","author":"Zhou","year":"2020","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Panda, P., Sengupta, A., and Roy, K. (2016). Conditional deep learning for energy-efficient and enhanced pattern recognition. Proceedings of the 2016 Design, Automation & Test in Europe Conference & Exhibition (DATE), IEEE.","DOI":"10.3850\/9783981537079_0819"},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Teerapittayanon, S., McDanel, B., and Kung, H.T. (2016). Branchynet: Fast inference via early exiting from deep neural networks. Proceedings of the 2016 23rd International Conference on Pattern Recognition (ICPR), IEEE.","DOI":"10.1109\/ICPR.2016.7900006"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"2278","DOI":"10.1109\/5.726791","article-title":"Gradient-based learning applied to document recognition","volume":"86","author":"LeCun","year":"2002","journal-title":"Proc. IEEE"},{"key":"ref_27","first-page":"1097","article-title":"Imagenet classification with deep convolutional neural networks","volume":"25","author":"Krizhevsky","year":"2012","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_28","unstructured":"Liu, D., Kan, M., Shan, S., and Chen, X. (2023, January 1\u20135). A simple romance between multi-exit vision transformer and token reduction. Proceedings of the Twelfth International Conference on Learning Representations, Kigali, Rwanda."},{"key":"ref_29","unstructured":"Yoon, J.W., Woo, B.J., and Kim, N.S. (2022). Hubert-ee: Early exiting hubert for efficient speech recognition. arXiv."},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Lasbordes, M., Falavigna, D., and Brutti, A. (2025). Splitformer: An improved early-exit architecture for automatic speech recognition on edge devices. arXiv.","DOI":"10.23919\/EUSIPCO63237.2025.11226356"},{"key":"ref_31","unstructured":"Di Francesco, A.G., Bucarelli, M.S., Nardini, F.M., Perego, R., Tonellotto, N., and Silvestri, F. (2025). Early-Exit Graph Neural Networks. arXiv."},{"key":"ref_32","unstructured":"Yu, J., Yang, L., Xu, N., Yang, J., and Huang, T. (2018). Slimmable neural networks. arXiv."},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Li, C., Wang, G., Wang, B., Liang, X., Li, Z., and Chang, X. (2021, January 19\u201325). Dynamic slimmable network. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Virtual.","DOI":"10.1109\/CVPR46437.2021.00850"},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Chen, Y., Dai, X., Liu, M., Chen, D., Yuan, L., and Liu, Z. (2020, January 14\u201319). Dynamic convolution: Attention over convolution kernels. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.","DOI":"10.1109\/CVPR42600.2020.01104"},{"key":"ref_35","doi-asserted-by":"crossref","unstructured":"Pan, Z., Cai, J., and Zhuang, B. (2023, January 17\u201324). Stitchable neural networks. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Vancouver, BC, Canada.","DOI":"10.1109\/CVPR52729.2023.01545"},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"He, K., Gkioxari, G., Doll\u00e1r, P., and Girshick, R. (2017, January 22\u201329). Mask r-cnn. Proceedings of the IEEE International Conference on Computer Vision, Venice, Italy.","DOI":"10.1109\/ICCV.2017.322"},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"652","DOI":"10.1109\/TPAMI.2019.2938758","article-title":"Res2net: A new multi-scale backbone architecture","volume":"43","author":"Gao","year":"2019","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Tan, M., Pang, R., and Le, Q.V. (2020, January 14\u201319). Efficientdet: Scalable and efficient object detection. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.","DOI":"10.1109\/CVPR42600.2020.01079"},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"Song, X., Wang, P., Zhou, D., Zhu, R., Guan, C., Dai, Y., Su, H., Li, H., and Yang, R. (2019, January 16\u201320). Apollocar3d: A large 3d car instance understanding benchmark for autonomous driving. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, CA, USA.","DOI":"10.1109\/CVPR.2019.00560"},{"key":"ref_40","unstructured":"Liu, S., Li, T., Chen, W., and Li, H. (November, January 27). Soft rasterizer: A differentiable renderer for image-based 3d reasoning. Proceedings of the IEEE\/CVF International Conference on Computer Vision, Seoul, Republic of Korea."},{"key":"ref_41","doi-asserted-by":"crossref","unstructured":"Li, Y., Yang, S., Liu, P., Zhang, S., Wang, Y., Wang, Z., Yang, W., and Xia, S.T. (2022). Simcc: A simple coordinate classification perspective for human pose estimation. Proceedings of the European Conference on Computer Vision, Springer.","DOI":"10.1007\/978-3-031-20068-7_6"},{"key":"ref_42","doi-asserted-by":"crossref","unstructured":"Chen, Y., Tai, L., Sun, K., and Li, M. (2020, January 14\u201319). Monopair: Monocular 3d object detection using pairwise spatial relationships. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.","DOI":"10.1109\/CVPR42600.2020.01211"},{"key":"ref_43","doi-asserted-by":"crossref","unstructured":"Lu, Y., Ma, X., Yang, L., Zhang, T., Liu, Y., Chu, Q., Yan, J., and Ouyang, W. (2021, January 11\u201317). Geometry uncertainty projection network for monocular 3d object detection. Proceedings of the IEEE\/CVF International Conference on Computer Vision, Montreal, QC, Canada.","DOI":"10.1109\/ICCV48922.2021.00310"},{"key":"ref_44","doi-asserted-by":"crossref","unstructured":"Lin, T.Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Doll\u00e1r, P., and Zitnick, C.L. (2014). Microsoft coco: Common objects in context. Proceedings of the European Conference on Computer Vision, Springer.","DOI":"10.1007\/978-3-319-10602-1_48"}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/8\/2355\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,13]],"date-time":"2026-04-13T07:37:44Z","timestamp":1776065864000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/8\/2355"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,11]]},"references-count":44,"journal-issue":{"issue":"8","published-online":{"date-parts":[[2026,4]]}},"alternative-id":["s26082355"],"URL":"https:\/\/doi.org\/10.3390\/s26082355","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,11]]}}}