{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,5]],"date-time":"2026-06-05T15:42:55Z","timestamp":1780674175828,"version":"3.54.1"},"reference-count":43,"publisher":"MDPI AG","issue":"16","license":[{"start":{"date-parts":[[2023,8,13]],"date-time":"2023-08-13T00:00:00Z","timestamp":1691884800000},"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":["2022YFB3902400"],"award-info":[{"award-number":["2022YFB3902400"]}],"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":["2023JJ30185"],"award-info":[{"award-number":["2023JJ30185"]}],"id":[{"id":"10.13039\/501100012166","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Natural Science Foundation of Hunan, China","award":["2022YFB3902400"],"award-info":[{"award-number":["2022YFB3902400"]}]},{"name":"Natural Science Foundation of Hunan, China","award":["2023JJ30185"],"award-info":[{"award-number":["2023JJ30185"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Remote Sensing"],"abstract":"<jats:p>Extensive research on deep neural networks for LiDAR point clouds has contributed inexhaustible momentum to the development of computer 3D vision applications. However, storage and energy consumption have always been a challenge for deploying these deep models on mobile devices. Quantization provides a feasible route, of which current primary research is focused on uniform bit-width quantization without considering different layers or filters\u2019 sensitivity to different bit-widths. This article proposes a novel hybrid compression method based on relaxed mixed-precision quantization, relaxed weights pruning, and knowledge distillation to overcome the limitations of uniform quantization illustrated above, while further improving model accuracy and reducing model memory consumption. It employs a differentiable searching method to search for the optimal bit allocation and weight sparsity, while conducting feature distillation, accordingly considering the feature degradation by pooling operation in point cloud deep models. The proposed method combines three compression techniques to balance the trade-off between compression accuracy and model size. Pruning alleviates the increasing memory consumption problem caused by mixed-precision quantization, while distillation improves compression accuracy without increasing model size. The experiments validate that the proposed method outperforms state-of-the-art typical uniform quantization methods in terms of accuracy with an acceptable and relatively competitive compression performance.<\/jats:p>","DOI":"10.3390\/rs15164015","type":"journal-article","created":{"date-parts":[[2023,8,14]],"date-time":"2023-08-14T10:40:31Z","timestamp":1692009631000},"page":"4015","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":5,"title":["Deep Hybrid Compression Network for Lidar Point Cloud Classification and Segmentation"],"prefix":"10.3390","volume":"15","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-0081-3848","authenticated-orcid":false,"given":"Zhi","family":"Zhao","sequence":"first","affiliation":[{"name":"College of Electronic Science, National University of Defence Technology, Changsha 410073, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4752-3370","authenticated-orcid":false,"given":"Yanxin","family":"Ma","sequence":"additional","affiliation":[{"name":"College of Meteorology and Oceanography, National University of Defence Technology, Changsha 410073, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ke","family":"Xu","sequence":"additional","affiliation":[{"name":"College of Electronic Science, National University of Defence Technology, Changsha 410073, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jianwei","family":"Wan","sequence":"additional","affiliation":[{"name":"College of Electronic Science, National University of Defence Technology, Changsha 410073, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2023,8,13]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"4338","DOI":"10.1109\/TPAMI.2020.3005434","article-title":"Deep Learning for 3D Point Clouds: A Survey","volume":"43","author":"Guo","year":"2021","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"485","DOI":"10.1109\/JPROC.2020.2976475","article-title":"Model Compression and Hardware Acceleration for Neural Networks: A Comprehensive Survey","volume":"108","author":"Deng","year":"2020","journal-title":"Proc. IEEE"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"3282","DOI":"10.1109\/TMC.2020.2999956","article-title":"AdaDeep: A Usage-Driven, Automated Deep Model Compression Framework for Enabling Ubiquitous Intelligent Mobiles","volume":"20","author":"Liu","year":"2020","journal-title":"IEEE Trans. Mob. Comput."},{"key":"ref_4","first-page":"886","article-title":"Learning Low Resource Consumption CNN Through Pruning and Quantization","volume":"10","author":"Qi","year":"2021","journal-title":"IEEE Trans. Emerg. Top. Comput."},{"key":"ref_5","unstructured":"Courbariaux, M., Bengio, Y., and David, J.-P. (2015, January 7\u201312). BinaryConnect: Training Deep Neural Networks with binary weights during propagations. Proceedings of the 28th International Conference on Neural Information Processing Systems (NeurIPS), Montreal, QC, Canada."},{"key":"ref_6","unstructured":"Hubara, I., Courbariaux, M., Soudry, D., El-Yaniv, R., and Bengio, Y. (2016, January 5\u201310). Binarized neural networks. Proceedings of the Advances in Neural Information Processing Systems 29 (NeurIPS), Barcelona, Spain."},{"key":"ref_7","unstructured":"Rastegari, M., Ordonez, V., Redmon, J., and Farhadi, A. (2016). European Conference on Computer Vision (ECCV), Springer."},{"key":"ref_8","unstructured":"Choi, J., Wang, Z., Venkataramani, S., Chuang, P.I.-J., Srinivasan, V., and Gopalakrishnan, K. (May, January 30). PACT: Parameterized clipping activation for quantized neural networks. Proceedings of the International Conference on Learning Representations (ICLR), Vancouver, BC, Canada."},{"key":"ref_9","unstructured":"Mishra, A.K., Nurvitadhi, E., Cook, J.J., and Marr, D. (May, January 30). WRPN: Wide reduced-precision networks. Proceedings of the International Conference on Learning Representations (ICLR), Vancouver, BC, Canada."},{"key":"ref_10","unstructured":"Darabi, S., Belbahri, M., Courbariaux, M., and Nia, V. (2018). BNN+: Improved Binary Network Training. arXiv."},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Liu, Z., Wu, B., Luo, W., Yang, X., Liu, W., and Cheng, K. (2018, January 8\u201314). Bi-real net: Enhancing the performance of 1-bit cnns with improved representational capability and advanced training algorithm. Proceedings of the European Conference on Computer Vision (ECCV), Munich, Germany.","DOI":"10.1007\/978-3-030-01267-0_44"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Cai, Z., He, X., Sun, J., and Vasconcelos, N. (2017, January 21\u201326). Deep Learning with Low Precision by Half-Wave Gaussian Quantization. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA.","DOI":"10.1109\/CVPR.2017.574"},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Qin, H., Gong, R., Liu, X., Shen, M., Wei, Z., Yu, F., and Song, J. (2020, January 13\u201319). Forward and Backward Information Retention for Accurate Binary Neural Networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA.","DOI":"10.1109\/CVPR42600.2020.00232"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"14","DOI":"10.1007\/s40687-018-0177-6","article-title":"Blended coarse gradient descent for full quantization of deep neural networks","volume":"6","author":"Yin","year":"2018","journal-title":"Res. Math. Sci."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Cai, Y., Yao, Z., Dong, Z., Gholami, A., Mahoney, M.W., and Keutzer, K. (2020, January 13\u201319). ZeroQ: A Novel Zero Shot Quantization Framework. Proceedings of the 2020 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA.","DOI":"10.1109\/CVPR42600.2020.01318"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Yang, H., Gui, S., Zhu, Y., and Liu, J. (2019, January 13\u201319). Automatic Neural Network Compression by Sparsity-Quantization Joint Learning: A Constrained Optimization-Based Approach. Proceedings of the 2020 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA.","DOI":"10.1109\/CVPR42600.2020.00225"},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Wang, K., Liu, Z., Lin, Y., Lin, J., and Han, S. (2018, January 15\u201320). HAQ: Hardware-Aware Automated Quantization with Mixed Precision. Proceedings of the 2019 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, USA.","DOI":"10.1109\/CVPR.2019.00881"},{"key":"ref_18","unstructured":"Lou, Q., Guo, F., Liu, L., Kim, M., and Jiang, L. (2019). AutoQ: Automated Kernel-Wise Neural Network Quantization. arXiv."},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Cai, Z., and Vasconcelos, N. (2020, January 13\u201319). Rethinking Differentiable Search for Mixed-Precision Neural Networks. Proceedings of the 2020 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA.","DOI":"10.1109\/CVPR42600.2020.00242"},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Yang, L., and Jin, Q. (2020, January 7\u201312). FracBits: Mixed Precision Quantization via Fractional Bit-Widths. Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), New York, NY, USA.","DOI":"10.1609\/aaai.v35i12.17269"},{"key":"ref_21","unstructured":"Song, H., Pool, J., Tran, J., and Dally, W.J. (2015, January 7\u201312). Learning both weights and connections for efficient neural network. Proceedings of the Advances in Neural Information Processing Systems (NeurIPS), Montreal, QC, Canada."},{"key":"ref_22","unstructured":"Wang, C., Wang, C., Zhang, G., and Grosse, R.B. (2020). Picking Winning Tickets Before Training by Preserving Gradient Flow. arXiv."},{"key":"ref_23","unstructured":"Zhou, A., Ma, Y., Zhu, J., Liu, J., Zhang, Z., Yuan, K., Sun, W., and Li, H. (2021). Learning N: M Fine-grained Structured Sparse Neural Networks from Scratch. arXiv."},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Guo, S., Wang, Y., Li, Q., and Yan, J. (2020, January 13\u201319). DMCP: Differentiable Markov Channel Pruning for Neural Networks. Proceedings of the 2020 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA.","DOI":"10.1109\/CVPR42600.2020.00161"},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Lin, M., Ji, R., Wang, Y., Zhang, Y., Zhang, B., Tian, Y., and Shao, L. (2020, January 13\u201319). HRank: Filter Pruning Using High-Rank Feature Map. Proceedings of the 2020 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA.","DOI":"10.1109\/CVPR42600.2020.00160"},{"key":"ref_26","unstructured":"Hinton, G.E., Vinyals, O., and Dean, J. (2015). Distilling the Knowledge in a Neural Network. arXiv."},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Passban, P., Wu, Y., Rezagholizadeh, M., and Liu, Q. (2021, January 2\u20139). ALP-KD: Attention-Based Layer Projection for Knowledge Distillation. Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), Virtually.","DOI":"10.1609\/aaai.v35i15.17610"},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"1275","DOI":"10.1109\/TIP.2020.3026728","article-title":"KT-GAN: Knowledge-Transfer Generative Adversarial Network for Text-to-Image Synthesis","volume":"30","author":"Tan","year":"2021","journal-title":"IEEE Trans. Image Process."},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Yuan, F., Shou, L., Pei, J., Lin, W., Gong, M., Fu, Y., and Jiang, D. (2021, January 2\u20139). Reinforced Multi-Teacher Selection for Knowledge Distillation. Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), Virtually.","DOI":"10.1609\/aaai.v35i16.17680"},{"key":"ref_30","unstructured":"Liu, H., Simonyan, K., and Yang, Y. (2018). DARTS: Differentiable Architecture Search. arXiv."},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Ye, P., Li, B., Li, Y., Chen, T., Fan, J., and Ouyang, W. (2022, January 18\u201324). \u03b2-DARTS: Beta-Decay Regularization for Differentiable Architecture Search. Proceedings of the 2022 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), New Orleans, LA, USA.","DOI":"10.1109\/CVPR52688.2022.01060"},{"key":"ref_32","unstructured":"Zhang, M., Su, S., Pan, S., Chang, X., Abbasnejad, E., and Haffari, R. (2021, January 18\u201324). iDARTS: Differentiable Architecture Search with Stochastic Implicit Gradients. Proceedings of the International Conference on Machine Learning (ICML), Virtual."},{"key":"ref_33","unstructured":"Li, C., Ning, J., Hu, H., and He, K. (2022). Enhancing the Robustness, Efficiency, and Diversity of Differentiable Architecture Search. arXiv."},{"key":"ref_34","unstructured":"Qin, Y., Wang, X., Zhang, Z., and Zhu, W. (2021, January 6\u201314). Graph Differentiable Architecture Search with Structure Learning. Proceedings of the Advances in Neural Information Processing Systems (NeurIPS), Online."},{"key":"ref_35","unstructured":"Qi, C., Su, H., Mo, K., and Guibas, L.J. (2017, January 21\u201326). PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA."},{"key":"ref_36","unstructured":"Wu, Z., Song, S., Khosla, A., Yu, F., Zhang, L., Tang, X., and Xiao, J. (2015, January 7\u201312). 3D ShapeNets: A deep representation for volumetric shapes. Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston, MA, USA."},{"key":"ref_37","unstructured":"Chang, A.X., Funkhouser, T.A., Guibas, L.J., Hanrahan, P., Huang, Q., Li, Z., Savarese, S., Savva, M., Song, S., and Su, H. (2015). ShapeNet: An Information-Rich 3D Model Repository. arXiv."},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Armeni, I., Sener, O., Zamir, A.R., Jiang, H., Brilakis, I.K., Fischer, M., and Savarese, S. (2016, January 27\u201330). 3D Semantic Parsing of Large-Scale Indoor Spaces. Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA.","DOI":"10.1109\/CVPR.2016.170"},{"key":"ref_39","unstructured":"Qi, C., Yi, L., Su, H., and Guibas, L.J. (2017, January 4\u20139). PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space. Proceedings of the Advances in Neural Information Processing Systems (NeurIPS), Long Beach, CA, USA."},{"key":"ref_40","unstructured":"Li, Y., Bu, R., Sun, M., Wu, W., Di, X., and Chen, B. (2018, January 3\u20138). PointCNN: Convolution On X-Transformed Points. Proceedings of the Advances in Neural Information Processing Systems (NeurIPS), Montr\u00e9al, QC, Canada."},{"key":"ref_41","first-page":"1","article-title":"Dynamic Graph CNN for Learning on Point Clouds","volume":"38","author":"Wang","year":"2018","journal-title":"ACM Trans. Graph. TOG"},{"key":"ref_42","doi-asserted-by":"crossref","unstructured":"Liu, Y., Fan, B., Xiang, S., and Pan, C. (2019, January 15\u201320). Relation-Shape Convolutional Neural Network for Point Cloud Analysis. Proceedings of the 2019 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, USA.","DOI":"10.1109\/CVPR.2019.00910"},{"key":"ref_43","doi-asserted-by":"crossref","unstructured":"Thomas, H., Qi, C., Deschaud, J.-E., Marcotegui, B., Goulette, F., and Guibas, L.J. (November, January 27). KPConv: Flexible and Deformable Convolution for Point Clouds. Proceedings of the 2019 IEEE\/CVF International Conference on Computer Vision (ICCV), Seoul, Republic of Korea.","DOI":"10.1109\/ICCV.2019.00651"}],"container-title":["Remote Sensing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2072-4292\/15\/16\/4015\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T20:32:52Z","timestamp":1760128372000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2072-4292\/15\/16\/4015"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,8,13]]},"references-count":43,"journal-issue":{"issue":"16","published-online":{"date-parts":[[2023,8]]}},"alternative-id":["rs15164015"],"URL":"https:\/\/doi.org\/10.3390\/rs15164015","relation":{},"ISSN":["2072-4292"],"issn-type":[{"value":"2072-4292","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,8,13]]}}}