SFNet-N: An Improved SFNet Algorithm for Semantic Segmentation of Low-Light Autonomous Driving Road Scenes
Open accessThe DOI matches its Crossref record; 57 of the 57 references it cites were found in a registry or an index.
Abstract
In recent years, considerable progress has been made in semantic segmentation of images with favorable environments. However, the environmental perception of autonomous driving under adverse weather conditions is still very challenging. In particular, the low visibility at nighttime greatly affects driving safety. In this paper, we aim to explore image segmentation in low-light scenarios, thereby expanding the application range of autonomous vehicles. The segmentation algorithms for road scenes based on deep learning are highly dependent on the volume of images with pixel-level annotations. Considering the scarcity of labeled large-scale nighttime data, we performed synthetic data collection and data style transfer using images acquired in daytime based on the autonomous driving simulation platform and generative adversarial network, respectively. In addition, we also proposed a novel nighttime segmentation framework (SFNET-N) to effectively recognize objects in dark environments, aiming at the boundary blurring caused by low semantic contrast in low-illumination images. Specifically, the framework comprises a light enhancement network which introduces semantic information for the first time and a segmentation network with strong feature extraction capability. Extensive experiments with Dark Zurich-test and Nighttime Driving-test datasets show the effectiveness of our method compared with existing state-of-the art approaches, with 56.9% and 57.4% mIoU (mean of category-wise intersection-over-union) respectively. Finally, we also performed real-vehicle verification of the proposed models in road scenes of Zhenjiang city with poor lighting. The datasets are available athttps://github.com/pupu-chenyanyan/semantic-segmentation-on-nightime.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Very Deep Convolutional Networks for Large-Scale Image Recognition
Simonyan, K., & Zisserman, A. (2014) · arXiv · doi:10.48550/arxiv.1409.1556 · Cited by 114,148 · cited 1×
“In order to further reduce the number of parameters in a convolution and improve the computational efficiency, three small 3 × 3 convolution kernels are used to replace one large 7 × 7 convolutional kernel similar to VGG [48].”
MethodImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155 · cited 1×
“For example, AlexNet [11] won the ImageNet image recognition competition.”
U-Net: Convolutional Networks for Biomedical Image Segmentation
Ronneberger, O., Fischer, P., & Brox, T. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-24574-4_28 · Cited by 99,404
ImageNet: A large-scale hierarchical image database
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009) · 2009 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2009.5206848 · Cited by 65,333 · cited 1×
“To improve the segmentation performance, the models were pre-trained using ImgeNet [52].”
MethodFully convolutional networks for semantic segmentation
Long, J., Shelhamer, E., & Darrell, T. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298965 · Cited by 38,834
Squeeze-and-Excitation Networks
Hu, J., Shen, L., & Sun, G. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00745 · Cited by 35,667
Feature Pyramid Networks for Object Detection
Lin, T.-Y., Dollar, P., Girshick, R., He, K., Hariharan, B., & Belongie, S. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.106 · Cited by 30,939 · cited 1×
“Specifically, encoder gradually reduces the image resolution based on the convolution layers, and decoder recovers the details and resolution layer by layer based on deconvolution and comprises different variations of feature pyramid network (FPN) [45].”
BackgroundCBAM: Convolutional Block Attention Module
Woo, S., Park, J., Lee, J.-Y., & Kweon, I. S. (2018) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-01234-2_1 · Cited by 25,853
Unpaired image-to-image translation using cycle-consistent adversarial networks
Zhu, J.-Y., Park, T., Isola, P., & Efros, A. A. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.244 · Cited by 23,602
Chen, L.-C., Papandreou, G., Kokkinos, I., Murphy, K., & Yuille, A. L. (2016) · arXiv · doi:10.48550/arxiv.1606.00915 · Cited by 22,392
Self-citationChen, L.-C., Papandreou, G., Kokkinos, I., Murphy, K., & Yuille, A. L. (2018) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2017.2699184 · Cited by 22,392 · cited 2×
“For example, one study [3] presented the ASPP module, which used parallel dilated convolutions [19] with different dilation rates to aggregate multi-scale context information.”
BackgroundSelf-citationEncoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation
Chen, L.-C., Zhu, Y., Papandreou, G., Schroff, F., & Adam, H. (2018) · arXiv · doi:10.48550/arxiv.1802.02611 · Cited by 20,063 · cited 1×
“Currently, the convolutional neural network (CNN) based algorithms, such as Deeplab [1]–[4] or HRNet-OCR [5], [6], perform segmentation tasks with high accuracy and fast inference speed.”
MethodSelf-citationEncoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation
Chen, L.-C., Zhu, Y., Papandreou, G., Schroff, F., & Adam, H. (2018) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-01234-2_49 · Cited by 20,054
Self-citationSegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation
Badrinarayanan, V., Kendall, A., & Cipolla, R. (2017) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2016.2644615 · Cited by 19,631
Labaca-Castro, R. (2023) · Machine Learning under Malware Attack · doi:10.1007/978-3-658-40442-0_9 · Cited by 19,272
Zhao, H., Shi, J., Qi, X., Wang, X., & Jia, J. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.660 · Cited by 16,438 · cited 2×
“The subsequent semantic segmentation models, such as UNet [13], UNet ++ [14], DenseCRF [15], DLNet [16] and PSPNet [17] are based on FCN.”
BackgroundMethodNonlinear total variation based noise removal algorithms
Rudin, L. I., Osher, S., & Fatemi, E. (1992) · Physica D: Nonlinear Phenomena · doi:10.1016/0167-2789(92)90242-f · Cited by 15,704 · cited 1×
“The total variation loss L t v is widely used in image processing [50], [51] to make the image smoother.”
MethodThe Cityscapes Dataset for Semantic Urban Scene Understanding
Cordts, M., Omran, M., Ramos, S., Rehfeld, T., Enzweiler, M., Benenson, R., Franke, U., Roth, S., & Schiele, B. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.350 · Cited by 12,433 · cited 1×
“Therefore, the models trained using the authoritative daytime datasets, such as CityScapes [7], cannot be directly applied to low-light scenarios.”
MethodWang, X., Girshick, R., Gupta, A., & He, K. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00813 · Cited by 11,596
UNet++: A Nested U-Net Architecture for Medical Image Segmentation
Zhou, Z., Rahman Siddiquee, M. M., Tajbakhsh, N., & Liang, J. (2018) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-00889-5_1 · Cited by 9,610
Multi-scale context aggregation by dilated convolutions
Yu, F., & Koltun, V. (2015) · arXiv · doi:10.48550/arxiv.1511.07122 · Cited by 7,738 · cited 1×
“For example, one study [3] presented the ASPP module, which used parallel dilated convolutions [19] with different dilation rates to aggregate multi-scale context information.”
BackgroundDeformable Convolutional Networks
Dai, J., Qi, H., Xiong, Y., Li, Y., Zhang, G., Hu, H., & Wei, Y. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.89 · Cited by 7,513 · cited 1×
“DCN [42] uses a learnable offset to represent feature direction of the target region (such as an underexposed vehicle), so that the receptive field is no longer limited to a fixed geometry.”
MethodSelf-citationDual Attention Network for Scene Segmentation
Fu, J., Liu, J., Tian, H., Li, Y., Bao, Y., Fang, Z., & Lu, H. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00326 · Cited by 7,351 · cited 1×
“For instance, DANet [22] effectively modeled the features from the spatial and channel perspectives to capture long-range dependencies.”
BackgroundSelf-citationSegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers
Xie, E., Wang, W., Yu, Z., Anandkumar, A., Alvarez, J. M., & Luo, P. (2021) · arXiv · doi:10.48550/arxiv.2105.15203 · Cited by 5,090
Deep High-Resolution Representation Learning for Visual Recognition
Wang, J., Sun, K., Cheng, T., Jiang, B., Deng, C., Zhao, Y., Liu, D., Mu, Y., Tan, M., Wang, X., et al. (2021) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2020.2983686 · Cited by 4,894
Rethinking Semantic Segmentation from a Sequence-to-Sequence Perspective with Transformers
Zheng, S., Lu, J., Zhao, H., Zhu, X., Luo, Z., Wang, Y., Fu, Y., Feng, J., Xiang, T., Torr, P. H. S., et al. (2021) · 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr46437.2021.00681 · Cited by 4,008 · cited 2×
“Recently, there has been a huge upsurge [26]–[29] in studies applying Transformers, originally used for Natural Language Processing (NLP), to visual tasks.”
BackgroundMethodRethinking Semantic Segmentation from a Sequence-to-Sequence Perspective with Transformers
Zheng, S., Lu, J., Zhao, H., Zhu, X., Luo, Z., Wang, Y., Fu, Y., Feng, J., Xiang, T., Torr, P. H. S., et al. (2020) · arXiv · doi:10.48550/arxiv.2012.15840 · Cited by 4,004
CARLA: An open urban driving simulator
Dosovitskiy, A., Ros, G., Codevilla, F., Lopez, A., & Koltun, V. (2017) · arXiv · doi:10.48550/arxiv.1711.03938 · Cited by 3,718 · cited 1×
“…cost required to build an accurate and effective low-illumination dataset comprising real scenes, and the limitation of hardware resources in our laboratory, we used the simulation platform CARLA [37] to collect the low-light dataset required for this experiment and named it Synthesiscarla.”
MethodCCNet: Criss-Cross Attention for Semantic Segmentation
Huang, Z., Wang, X., Huang, L., Huang, C., Wei, Y., & Liu, W. (2019) · 2019 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00069 · Cited by 3,045 · cited 1×
“CCNet [24] proposed a criss-cross attention module to aggregate long-range dependencies from horizontal and vertical directions.”
BackgroundEfficient Inference in Fully Connected CRFs with Gaussian Edge Potentials
Krähenbühl, P., & Koltun, V. (2012) · arXiv · doi:10.48550/arxiv.1210.5644 · Cited by 2,908 · cited 1×
“The subsequent semantic segmentation models, such as UNet [13], UNet ++ [14], DenseCRF [15], DLNet [16] and PSPNet [17] are based on FCN.”
MethodBDD100K: A Diverse Driving Dataset for Heterogeneous Multitask Learning
Yu, F., Chen, H., Wang, X., Xian, W., Chen, Y., Liu, F., Madhavan, V., & Darrell, T. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.00271 · Cited by 2,641
Self-citationZero-Reference Deep Curve Estimation for Low-Light Image Enhancement
Guo, C., Li, C., Guo, J., Loy, C. C., Hou, J., Kwong, S., & Cong, R. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.00185 · Cited by 2,459 · cited 5×
“Similarly, following the previous work [41], we removed the BN layer and down-sampling operations that can damage the pixel relation, so that the resolution remains the same.”
MethodInfluentialLearning to Adapt Structured Output Space for Semantic Segmentation
Tsai, Y.-H., Hung, W.-C., Schulter, S., Sohn, K., Yang, M.-H., & Chandraker, M. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00780 · Cited by 1,430
Object-Contextual Representations for Semantic Segmentation
Yuan, Y., Chen, X., & Wang, J. (2020) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-58539-6_11 · Cited by 1,285
ACDC: The Adverse Conditions Dataset with Correspondences for Semantic Driving Scene Understanding
Sakaridis, C., Dai, D., & Van Gool, L. (2021) · 2021 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.01059 · Cited by 572
Semantic Flow for Fast and Accurate Scene Parsing
Li, X., You, A., Zhu, Z., Zhao, H., Yang, M., Yang, K., Tan, S., & Tong, Y. (2020) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-58452-8_45 · Cited by 416
YOLOv4-5D: An Effective and Efficient Object Detector for Autonomous Driving
Cai, Y., Luan, T., Gao, H., Wang, H., Chen, L., Li, Y., Sotelo, M. A., & Li, Z. (2021) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2021.3065438 · Cited by 340
Self-citationZhang, Y. (2017) · De Gruyter eBooks · doi:10.1515/9783110524116 · Cited by 329
Hierarchical Multi-Scale Attention for Semantic Segmentation
Tao, A., Sapra, K., & Catanzaro, B. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2005.10821 · Cited by 328 · cited 2×
“Different from the hard weighting used by previous work [49] in reasoning, we predicted the mask for each scale of a pair separately, so that the sum of the parameters of scales is not strictly equal to 1.”
MethodLiu, Z., Cai, Y., Wang, H., Chen, L., Gao, H., Jia, Y., & Li, Y. (2022) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2021.3059674 · Cited by 320 · cited 1×
“Currently, the convolutional neural network (CNN) based algorithms, such as Deeplab [1]–[4] or HRNet-OCR [5], [6], perform segmentation tasks with high accuracy and fast inference speed.”
MethodSelf-citationSakaridis, C., Dai, D., & Van Gool, L. (2019) · 2019 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00747 · Cited by 281 · cited 3×
“…above for training, we also evaluated nighttime authoritative datasets with a few low-light images of pixel-level annotations, which are described as follows: 1) Dark Zurich [9]: This dataset contains 201 annotated images with 19 classes, of which 50 are used for validation and 101 for testing.”
MethodOCNet: Object Context for Semantic Segmentation
Yuan, Y., Huang, L., Guo, J., Zhang, C., Chen, X., & Wang, J. (2021) · International Journal of Computer Vision · doi:10.1007/s11263-021-01465-9 · Cited by 259 · cited 1×
“Several works have also been reported [20]–[25] on attention mechanism.”
BackgroundDark Model Adaptation: Semantic Image Segmentation from Daytime to Nighttime
Dai, D., & Gool, L. V. (2018) · 2018 21st International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc.2018.8569387 · Cited by 207
Sakaridis, C., Dai, D., & Van Gool, L. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2020.3045882 · Cited by 186
DANNet: A One-Stage Domain Adaptation Network for Unsupervised Nighttime Semantic Segmentation
Wu, X., Wu, Z., Guo, H., Ju, L., & Wang, S. (2021) · 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr46437.2021.01551 · Cited by 181 · cited 2×
“DANNet [34] proposed a one-stage domain adaptation network, which adopted an adversarial training with labeled daytime data and unlabeled roughly aligned day-night image pairs.”
BackgroundMethodNight-Time Scene Parsing With a Large Real Dataset
Tan, X., Xu, K., Cao, Y., Zhang, Y., Ma, L., & Lau, R. W. H. (2021) · IEEE Transactions on Image Processing · doi:10.1109/tip.2021.3122004 · Cited by 179 · cited 4×
“However, EGNet [36] adopted the fully supervised learning method, which created a real-scene dataset at night with pixel-level semantic annotations.”
BackgroundMethodResultInfluentialBridging the Day and Night Domain Gap for Semantic Segmentation
Romera, E., Bergasa, L. M., Yang, K., Álvarez, J. M., & Barea, R. (2019) · IEEE Intelligent Vehicles Symposium · doi:10.1109/ivs.2019.8813888 · Cited by 126 · cited 1×
“For example, two separate studies [32] and [33] used GANs to learn the mapping of input images to output images, and improved the segmentation performance from two perspectives, including direct inference of night images and real-time online conversion inference of night images through style…”
Zhang, J., Yang, K., Constantinescu, A., Peng, K., Müller, K., & Stiefelhagen, R. (2021) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw54120.2021.00202 · Cited by 102
Han, H.-Y., Chen, Y., Hsiao, P., & Fu, L. (2020) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2019.2962094 · Cited by 101 · cited 1×
“EdgeNet [44] introduced a class-aware edge loss module and a channel-wise attention mechanism in encoder-decoder to improve the accuracy with no harm to inference speed.”
MethodSelf-citationSun, L., Wang, K., Yang, K., & Xiang, K. (2019) · Artificial Intelligence and Machine Learning in Defense Applications · doi:10.1117/12.2532477 · Cited by 90 · cited 1×
“For example, two separate studies [32] and [33] used GANs to learn the mapping of input images to output images, and improved the segmentation performance from two perspectives, including direct inference of night images and real-time online conversion inference of night images through style…”
CDAda: A Curriculum Domain Adaptation for Nighttime Semantic Segmentation
Xu, Q., Ma, Y., Wu, J. B., Long, C., & Huang, X. (2021) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw54120.2021.00331 · Cited by 79 · cited 2×
“CDAda [30] is a curriculum domain adaptation method, which adjusts the model by entropy minimization and pseudo-label self-training method respectively according to the difficulty of the domain.”
MethodDark Model Adaptation: Semantic Image Segmentation from Daytime to Nighttime
Dai, D., & Van Gool, L. (2018) · ETH Zurich · doi:10.3929/ethz-b-000310974 · Cited by 60
Voxel-RCNN-Complex: An Effective 3-D Point Cloud Object Detector for Complex Traffic Conditions
Wang, H., Chen, Z., Cai, Y., Chen, L., Li, Y., Sotelo, M. Á., & Li, Z. (2022) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3165251 · Cited by 58
Self-citationRainy Night Scene Understanding With Near Scene Semantic Adaptation
Di, S., Feng, Q., Li, C.-G., Zhang, M., Zhang, H., Elezovikj, S., Tan, C. C., & Ling, H. (2020) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2020.2972912 · Cited by 42 · cited 1×
“For a variety of adverse conditions, a previous study [35] adapted the daytime model to rainy night scenarios by employing the idea of a pair of near scene cross-weather images sharing similar spatial layout and local context.”
MethodDLnet With Training Task Conversion Stream for Precise Semantic Segmentation in Actual Traffic Scene
Cai, Y., Dai, L., Wang, H., Chen, L., & Li, Y. (2021) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2021.3080261 · Cited by 42 · cited 1×
“The subsequent semantic segmentation models, such as UNet [13], UNet ++ [14], DenseCRF [15], DLNet [16] and PSPNet [17] are based on FCN.”
MethodSelf-citationsEnDec: An Improved Image to Image CNN for Foreground Localization
Akilan, T., & Wu, Q. M. J. (2019) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2019.2940547 · Cited by 36 · cited 1×
“The network reported in previous work [43] is a typical example of this structure.”
BackgroundNighttime Road Scene Parsing by Unsupervised Domain Adaptation
Song, C., Wu, J., Zhu, L., Zhang, M., & Ling, H. (2020) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2020.3033569 · Cited by 29 · cited 1×
“Another study [31] proposed an appearance transferring module to learn the shared feature representation of day-night images, thus improving the performance of night segmentation.”
Background
Metadata
- Authors
- Wang, H., Chen, Y., Cai, Y., Chen, L., Li, Y., Sotelo, M. A., & Li, Z.
- Venue
- IEEE Transactions on Intelligent Transportation Systems
- Published
- 2022-11
- Volume
- 23
- Issue
- 11
- Pages
- 21405-21417
- ISSN
- 1524-9050 (ISSN-L)
- 1558-0016
- OpenAlex
- W4285290430
- Cited by
- 233
- References
- 57
- FWCI
- 16.03
- Type
- journal article
- Language
- en
- Open access
- green
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.