{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,17]],"date-time":"2026-01-17T19:33:27Z","timestamp":1768678407865,"version":"3.49.0"},"reference-count":51,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2023,2,9]],"date-time":"2023-02-09T00:00:00Z","timestamp":1675900800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Neurorobot."],"abstract":"<jats:p>Lightweight semantic segmentation promotes the application of semantic segmentation in tiny devices. The existing lightweight semantic segmentation network (LSNet) has the problems of low precision and a large number of parameters. In response to the above problems, we designed a full 1D convolutional LSNet. The tremendous success of this network is attributed to the following three modules: 1D multi-layer space module (1D-MS), 1D multi-layer channel module (1D-MC), and flow alignment module (FA). The 1D-MS and the 1D-MC add global feature extraction operations based on the multi-layer perceptron (MLP) idea. This module uses 1D convolutional coding, which is more flexible than MLP. It increases the global information operation, improving features\u2019 coding ability. The FA module fuses high-level and low-level semantic information, which solves the problem of precision loss caused by the misalignment of features. We designed a 1D-mixer encoder based on the transformer structure. It performed fusion encoding of the feature space information extracted by the 1D-MS module and the channel information extracted by the 1D-MC module. 1D-mixer obtains high-quality encoded features with very few parameters, which is the key to the network\u2019s success. The attention pyramid with FA (AP-FA) uses an AP to decode features and adds a FA module to solve the problem of feature misalignment. Our network requires no pre-training and only needs a 1080Ti GPU for training. It achieved 72.6 mIoU and 95.6 FPS on the Cityscapes dataset and 70.5 mIoU and 122 FPS on the CamVid dataset. We ported the network trained on the ADE2K dataset to mobile devices, and the latency of 224 ms proves the application value of the network on mobile devices. The results on the three datasets prove that the network generalization ability we designed is powerful. Compared to state-of-the-art lightweight semantic segmentation algorithms, our designed network achieves the best balance between segmentation accuracy and parameters. The parameters of LSNet are only 0.62 M, which is currently the network with the highest segmentation accuracy within 1 M parameters.<\/jats:p>","DOI":"10.3389\/fnbot.2023.1119231","type":"journal-article","created":{"date-parts":[[2023,2,9]],"date-time":"2023-02-09T09:18:02Z","timestamp":1675934282000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":2,"title":["Rethinking 1D convolution for lightweight semantic segmentation"],"prefix":"10.3389","volume":"17","author":[{"given":"Chunyu","family":"Zhang","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Fang","family":"Xu","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Chengdong","family":"Wu","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Chenglong","family":"Xu","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2023,2,9]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"2481","DOI":"10.1109\/TPAMI.2016.2644615","article-title":"SegNet: A deep convolutional encoder-decoder architecture for image segmentation.","volume":"39","author":"Badrinarayanan","year":"2017","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"B2","doi-asserted-by":"publisher","first-page":"4175","DOI":"10.1109\/CVPRW56347.2022.00462","article-title":"Denoising pretraining for semantic segmentation","author":"Brempong","year":"2022","journal-title":"Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition"},{"key":"B3","doi-asserted-by":"publisher","first-page":"44","DOI":"10.1007\/978-3-540-88682-2_5","article-title":"Segmentation and recognition using structure from motion point clouds","author":"Brostow","year":"2008","journal-title":"European conference on computer vision"},{"key":"B4","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/VCIP.2017.8305148","article-title":"LinkNet: Exploiting encoder representations for efficient semantic segmentation","author":"Chaurasia","year":"2017","journal-title":"Proceedings of the 2017 IEEE visual communications and image processing (VCIP)"},{"key":"B5","doi-asserted-by":"publisher","first-page":"801","DOI":"10.1007\/978-3-030-01234-2_49","article-title":"Encoder-decoder with atrous separable convolution for semantic image segmentation","author":"Chen","year":"2018","journal-title":"Proceedings of the European conference on computer vision (ECCV)"},{"key":"B6","article-title":"Conditional positional encodings for vision transformers.","author":"Chu","year":"2021","journal-title":"arXiv"},{"key":"B7","doi-asserted-by":"publisher","first-page":"3213","DOI":"10.1109\/CVPR.2016.350","article-title":"The cityscapes dataset for semantic urban scene understanding","author":"Cordts","year":"2016","journal-title":"Proceedings of the IEEE conference on computer vision and pattern recognition"},{"key":"B8","author":"Dosovitskiy","year":"2020","journal-title":"An image is worth 16x16 words: Transformers for image recognition at scale."},{"key":"B9","doi-asserted-by":"publisher","first-page":"9716","DOI":"10.1109\/CVPR46437.2021.00959","article-title":"Rethinking BiSeNet for real-time semantic segmentation","author":"Fan","year":"2021","journal-title":"Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition"},{"key":"B10","article-title":"Visual attention network.","author":"Guo","year":"","journal-title":"arXiv"},{"key":"B11","doi-asserted-by":"publisher","first-page":"331","DOI":"10.1007\/s41095-022-0271-y","article-title":"Attention mechanisms in computer vision: A survey.","volume":"8","author":"Guo","year":"","journal-title":"Comput. Vis. Media"},{"key":"B12","article-title":"A survey on vision transformer","author":"Han","year":"2022","journal-title":"Proceedings of the IEEE transactions on pattern analysis and machine intelligence"},{"key":"B13","first-page":"15908","article-title":"Transformer in transformer.","volume":"34","author":"Han","year":"2021","journal-title":"Adv. Neural Inf. Proc. Syst."},{"key":"B14","doi-asserted-by":"publisher","first-page":"770","DOI":"10.1109\/CVPR.2016.90","article-title":"Deep residual learning for image recognition","author":"He","year":"2016","journal-title":"Proceedings of the IEEE conference on computer vision and pattern recognition"},{"key":"B15","doi-asserted-by":"publisher","first-page":"1314","DOI":"10.1109\/ICCV.2019.00140","article-title":"Searching for MobileNetV3","author":"Howard","year":"2019","journal-title":"Proceedings of the IEEE\/CVF international conference on computer vision"},{"key":"B16","article-title":"SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and< 0.5 MB model size.","author":"Iandola","year":"2016","journal-title":"arXiv"},{"key":"B17","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/ICMEW46912.2020.9106038","article-title":"Lrnnet: A light-weighted network with efficient reduced non-local operation for real-time semantic segmentation","author":"Jiang","year":"2020","journal-title":"Proceedings of the 2020 IEEE international conference on multimedia &amp; expo workshops (ICMEW)"},{"key":"B18","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3505244","article-title":"Transformers in vision: A survey.","volume":"54","author":"Khan","year":"2022","journal-title":"ACM Comput. Surv."},{"key":"B19","doi-asserted-by":"publisher","first-page":"6399","DOI":"10.1109\/CVPR.2019.00656","article-title":"Panoptic feature pyramid networks","author":"Kirillov","year":"2019","journal-title":"Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition"},{"key":"B20","article-title":"DabNet: Depth-wise asymmetric bottleneck for real-time semantic segmentation.","author":"Li","year":"2019","journal-title":"arXiv"},{"key":"B21","doi-asserted-by":"publisher","first-page":"9522","DOI":"10.1109\/CVPR.2019.00975","article-title":"DFANet: Deep feature aggregation for real-time semantic segmentation","author":"Li","year":"2019","journal-title":"Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition"},{"key":"B22","doi-asserted-by":"publisher","first-page":"775","DOI":"10.1007\/978-3-030-58452-8_45","article-title":"Semantic flow for fast and accurate scene parsing","author":"Li","year":"2020","journal-title":"Proceedings of the European conference on computer vision"},{"key":"B23","article-title":"Localvit: Bringing locality to vision transformers.","author":"Li","year":"2021","journal-title":"arXiv"},{"key":"B24","doi-asserted-by":"publisher","first-page":"2373","DOI":"10.1109\/ICASSP40776.2020.9053838","article-title":"FDDWNet: A lightweight convolutional neural network for real-time semantic segmentation","author":"Liu","year":"2020","journal-title":"Proceedings of the ICASSP 2020-2020 IEEE international conference on acoustics, speech and signal processing (ICASSP)"},{"key":"B25","doi-asserted-by":"publisher","first-page":"1855","DOI":"10.1109\/ICIP.2019.8803132","article-title":"Esnet: Edge-based segmentation network for real-time semantic segmentation in traffic scenes","author":"Lyu","year":"2019","journal-title":"Proceedings of the 2019 IEEE international conference on image processing (ICIP)"},{"key":"B26","doi-asserted-by":"publisher","first-page":"116","DOI":"10.1007\/978-3-030-01264-9_8","article-title":"ShuffleNet V2: Practical guidelines for efficient CNN architecture design","author":"Ma","year":"2018","journal-title":"Proceedings of the European conference on computer vision (ECCV)"},{"key":"B27","doi-asserted-by":"publisher","first-page":"552","DOI":"10.1007\/978-3-030-01249-6_34","article-title":"ESPNet: Efficient spatial pyramid of dilated convolutions for semantic segmentation","author":"Mehta","year":"2018","journal-title":"Proceedings of the European conference on computer vision (ECCV)"},{"key":"B28","doi-asserted-by":"publisher","first-page":"9190","DOI":"10.1109\/CVPR.2019.00941","article-title":"ESPNetv2: A light-weight, power efficient, and general purpose convolutional neural network","author":"Mehta","year":"2019","journal-title":"Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition"},{"key":"B29","doi-asserted-by":"publisher","first-page":"626","DOI":"10.1016\/j.neucom.2022.01.005","article-title":"Review the state-of-the-art technologies of semantic segmentation based on deep learning.","volume":"493","author":"Mo","year":"2022","journal-title":"Neurocomputing"},{"key":"B30","doi-asserted-by":"publisher","first-page":"4061","DOI":"10.1109\/CVPR46437.2021.00405","article-title":"Hyperseg: Patch-wise hypernetwork for real-time semantic segmentation","author":"Nirkin","year":"2021","journal-title":"Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition"},{"key":"B31","doi-asserted-by":"publisher","first-page":"1520","DOI":"10.1109\/ICCV.2015.178","article-title":"Learning deconvolution network for semantic segmentation","author":"Noh","year":"2015","journal-title":"Proceedings of the IEEE international conference on computer vision"},{"key":"B32","article-title":"ENet: A deep neural network architecture for real-time semantic segmentation.","author":"Paszke","year":"2016","journal-title":"arXiv"},{"key":"B33","article-title":"Fast-SCNN: Fast semantic segmentation network.","author":"Poudel","year":"2019","journal-title":"arXiv"},{"key":"B34","doi-asserted-by":"publisher","first-page":"263","DOI":"10.1109\/TITS.2017.2750080","article-title":"ERFNet: Efficient residual factorized convNet for real-time semantic segmentation.","volume":"19","author":"Romera","year":"2017","journal-title":"IEEE Trans. Intell. Transp. Syst."},{"key":"B35","doi-asserted-by":"publisher","first-page":"4510","DOI":"10.1109\/CVPR.2018.00474","article-title":"MobileNetV2: Inverted residuals and linear bottlenecks","author":"Sandler","year":"2018","journal-title":"Proceedings of the IEEE conference on computer vision and pattern recognition"},{"key":"B36","doi-asserted-by":"publisher","first-page":"7880","DOI":"10.1109\/TCSVT.2022.3187664","article-title":"UrbanLF: A comprehensive light field dataset for semantic segmentation of urban scenes.","volume":"32","author":"Sheng","year":"2022","journal-title":"IEEE Trans. Circuits Syst. Video Technol."},{"key":"B37","first-page":"6105","article-title":"EfficientNet: Rethinking model scaling for convolutional neural networks","author":"Tan","year":"2019","journal-title":"Proceedings of the 36th international conference on machine learning, ICML 2019"},{"key":"B38","author":"Treml","year":"2016","journal-title":"Speeding up semantic segmentation for autonomous driving."},{"key":"B39","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1080\/08839514.2022.2032924","article-title":"A survey on deep learning-based architectures for semantic segmentation on 2D images.","author":"Ulku","year":"2022","journal-title":"Appl. Artif. Intell."},{"key":"B40","doi-asserted-by":"publisher","first-page":"1860","DOI":"10.1109\/ICIP.2019.8803154","article-title":"LEDNet: A lightweight encoder-decoder network for real-time semantic segmentation","author":"Wang","year":"2019","journal-title":"Proceedings of the 2019 IEEE international conference on image processing (ICIP)"},{"key":"B41","doi-asserted-by":"publisher","first-page":"3","DOI":"10.1007\/978-3-030-01234-2_1","article-title":"CBAM: Convolutional block attention module","author":"Woo","year":"2018","journal-title":"Proceedings of the European conference on computer vision (ECCV)"},{"key":"B42","doi-asserted-by":"publisher","first-page":"1169","DOI":"10.1109\/TIP.2020.3042065","article-title":"CGNet: A light-weight context guided network for semantic segmentation.","volume":"30","author":"Wu","year":"2020","journal-title":"IEEE Trans. Image Process."},{"key":"B43","first-page":"12077","article-title":"SegFormer: Simple and efficient design for semantic segmentation with transformers.","volume":"34","author":"Xie","year":"2021","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"B44","article-title":"PIDNet: A real-time semantic segmentation network inspired from PID controller.","author":"Xu","year":"2022","journal-title":"arXiv"},{"key":"B45","article-title":"Dense dual-path network for real-time semantic segmentation","author":"Yang","year":"2020","journal-title":"Proceedings of the Asian conference on computer vision"},{"key":"B46","doi-asserted-by":"publisher","first-page":"3051","DOI":"10.1007\/s11263-021-01515-2","article-title":"BiSeNet V2: Bilateral network with guided aggregation for real-time semantic segmentation.","volume":"129","author":"Yu","year":"2021","journal-title":"Int. J. Comput. Vis."},{"key":"B47","doi-asserted-by":"publisher","first-page":"325","DOI":"10.1007\/978-3-030-01261-8_20","article-title":"BiSeNet: Bilateral segmentation network for real-time semantic segmentation","author":"Yu","year":"2018","journal-title":"Proceedings of the European conference on computer vision (ECCV)"},{"key":"B48","doi-asserted-by":"publisher","first-page":"564","DOI":"10.1007\/s10489-021-02437-9","article-title":"Lightweight and efficient asymmetric network design for real-time semantic segmentation.","volume":"52","author":"Zhang","year":"2022","journal-title":"Appl. Intell."},{"key":"B49","doi-asserted-by":"publisher","first-page":"405","DOI":"10.1007\/978-3-030-01219-9_25","article-title":"ICNet for real-time semantic segmentation on high-resolution images","author":"Zhao","year":"2018","journal-title":"Proceedings of the European conference on computer vision (ECCV)"},{"key":"B50","doi-asserted-by":"publisher","first-page":"2881","DOI":"10.1109\/CVPR.2017.660","article-title":"Pyramid scene parsing network","author":"Zhao","year":"2017","journal-title":"Proceedings of the IEEE conference on computer vision and pattern recognition"},{"key":"B51","doi-asserted-by":"publisher","first-page":"6881","DOI":"10.1109\/CVPR46437.2021.00681","article-title":"Rethinking semantic segmentation from a sequence-to-sequence perspective with transformers","author":"Zheng","year":"2021","journal-title":"Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition"}],"container-title":["Frontiers in Neurorobotics"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnbot.2023.1119231\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,2,9]],"date-time":"2023-02-09T09:18:12Z","timestamp":1675934292000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnbot.2023.1119231\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,2,9]]},"references-count":51,"alternative-id":["10.3389\/fnbot.2023.1119231"],"URL":"https:\/\/doi.org\/10.3389\/fnbot.2023.1119231","relation":{},"ISSN":["1662-5218"],"issn-type":[{"value":"1662-5218","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,2,9]]},"article-number":"1119231"}}