浏览全部资源
扫码关注微信
网络出版日期： 2024-12-30 ，
移动端阅览
陈晓雷,杜泽龙,张学功等.畸变自适应与位置感知的360度全景图像显著目标检测网络[J].中国图象图形学报,
Chen Xiaolei,Du Zelong,Zhang Xuegong,et al.Distortion-adaptive and position-aware network for salient object detection in 360° omnidirectional image[J].Journal of Image and Graphics,
目的
2
现有360°全景图像显著目标检测方法一定程度上解决了360°全景图像投影后的几何畸变问题，但是这些方法面对复杂场景或是前景与背景对比度较低的场景时，容易受到背景干扰，导致检测效果不佳。为了同时解决几何畸变和背景干扰，本文提出了一种畸变自适应与位置感知的360°全景图像显著目标检测网络（distortion-adaptive and position-aware network，DPNet）。
方法
2
提出了两个对畸变和位置敏感的自适应检测模块：畸变自适应模块（distortion-adaptive module，DAM）和位置感知模块（position-aware module，PAM）。它们可以帮助模型根据等矩形投影的特点和具体图像来决定该关注图像的哪些区域。在此基础上，本文进一步提出了一个显著信息增强模块（salient information enhancement module，SIEM），该模块用高级特征来指导低级特征，过滤其中的非显著信息，防止背景干扰对360°显著目标检测效果的影响。
结果
2
实验在2个公开数据集（360-SOD，360-SSOD）上与最新的13种方法进行了客观指标和主观结果的比较，在8个评价指标上的综合性能优于最新的13种方法。并且本文还设置了泛化性实验，采用交叉验证的方式证明了本文模型优秀的泛化性能。
结论
2
本文所提出的360°全景图像显著目标检测模型DPNet，同时考虑了360°全景图像投影后的几何畸变问题和复杂场景下的背景干扰问题，能够有效地、完全自适应地检测显著目标。
Objective
2
Salient object detection （SOD） in the field of computer vision originates from the study of human visual attention mechanisms. Its goal is to emulate the human ability to focus on specific objects or areas in complex scenes that human eyes are most interested in. Serving as a foundational research area in computer vision， SOD holds significance for various downstream tasks， such as object tracking， semantic segmentation， person re-identification， camouflaged object detection， and image retrieval， etc. In recent years， spurred by advancements in VR and AR technologies， SOD is no longer limited to traditional 2D images， it has progressively expanded to encompass 360° omnidirectional images （or panoramic images）. The application of 360° SOD serves as a crucial preprocessing step for enhancing the efficiency of subsequent advanced visual tasks. These tasks include coding， editing， stitching， quality assessment and transmission of 360° omnidirectional images. In contrast to traditional 2D images， 360° omnidirectional images exhibit the following core differences： 360° omnidirectional images are spherical. Since there is no encoder specifically for spherical images， it is necessary to project 360° omnidirectional images into 2D images for further processing. Common projection methods include equirectangular projection （ERP）， cube-map projection （CMP）， octahedron projection （OHP）， and so on. Regardless of the projection method used， geometric distortion is inevitable. This geometric distortion severely impacts the performance of SOD， resulting in a significant decline in performance when traditional 2D SOD methods are directly applied to 360° omnidirectional images. Therefore， addressing the challenge of geometric distortion generated by 360° omnidirectional image projection is the core problem in the field of salient object detection in 360° omnidirectional image （360° SOD）. In recent years， some 360° SOD methods have tried to solve the problem of geometric distortion caused by projection， and have achieved good detection results to a certain extent. But their approaches are either limited in effectiveness or rely on artificially designed features， restricting the model’s ability to detect salient objects in 360° omnidirectional images. Meanwhile， most of the models have poor detection results when facing complex scenes or scenes with low contrast between foreground and background， which are easily interfered with by the background. To solve the above problems， this paper introduces a distortion-adaptive and position-aware network （DPNet） for 360° SOD， aiming to further solve the problem of background interference in complex scenes by considering geometric distortion of ERP image， so as to better detect salient objects in 360° omnidirectional images.
Method
2
DPNet combines vision transformer （ViT） and convolutional neural networks （CNN） to build the basic framework of the network. It uses ViT and CNN to design the encoder and constructs a combination decoder based on U-Net architecture to decode the features from the two encoders step by step， so as to combine the global coding advantages of ViT and the local coding advantages of CNN. It should be noted that compared with previous dual parallel structures， the two encoder backbones of our network are not only parallel， but the ViT backbone also plays a guiding role for the CNN backbone， i.e. the CNN backbone can complement the detail information based on the semantic features extracted by the ViT backbone. On the one hand， to solve the geometric distortion problem caused by ERP， this paper proposes two distortion-adaptive detection modules： distortion-adaptive module （DAM） and position-aware module （PAM）. DAM models geometric distortion in feature maps through channel-by-channel deformable convolution. PAM calculates spatial weights along the latitude and longitude， directing the network to adaptively focus on salient regions in the image. Specifically， the global features extracted by the ViT backbone are processed by the DAM to model the geometric distortion， and then two branches are extracted， one branch is sent to the decoder， and the other branch is sent to PAM to provide position prior information. PAM is placed in the shallow layer of CNN backbone and is responsible for fusing the position prior information with the information extracted in the shallow layer of CNN backbone to guide the subsequent feature extraction. In this way， DPNet can decide which regions of the 360° omnidirectional images should be focused on according to the characteristics of ERP and specific input images. On the other hand， to further solve the problem of background interference in complex scenes， a salient information enhancement module （SIEM） is proposed. Currently， most SOD methods use structures such as U-Net to simply aggregate feature maps at different scales， which inevitably treats a large amount of non-salient information contained in the low-level features as useful information， leading to poor detection results. To address this， SIEM uses high-level features to guide low-level features， filters non-salient information， and prevents the influence of background interference on the effectiveness of 360° SOD.
Result
2
We compared our model with 13 state-of-the-art methods on 2 public datasets， namely， 360-SOD and 360-SSOD， and the overall performance of 8 evaluation metrics was better than the latest 13 methods. In addition， the generalization experiment is also set up in this paper， and the excellent generalization performance of the model is proved by cross-validation. And then an ablation experiment is set up to verify the performance of the proposed module. Finally， a set of complexity comparison experiments prove that the proposed model DPNet achieves a good balance in terms of detection accuracy and model complexity.
Conclusion
2
The existing 360° SOD methods cannot effectively address the geometric distortion problem after projection and the background interference problem in complex scenes， so we propose a distortion-adaptive and position-aware 360° SOD network （DPNet） based on the ViT and CNN. The proposed DAM and PAM play a pivotal role in guiding the network to focus on areas requiring attention based on the distinctive characteristics of ERP and specific input images. In addition， the proposed SIEM works to guide low-level features with high-level features， effectively filtering out non-salient information present in low-level features and enhancing the salient information， which can help the model effectively deal with the background interference problem in complex scenes. Through an extensive set of experiments， we demonstrate that our method outperforms 13 state-of-the-art SOD methods， establishing its superiority in 360° SOD applications.
360°全景图像显著目标检测畸变自适应位置感知抗背景干扰
360°omnidirectional imagesalient object detectiondistortion-adaptiveposition-awareanti-background interference
Achanta R， Hemami S， Estrada F， Susstrunk S. 2009. Frequency-tuned salient region detection//2009 IEEE conference on computer vision and pattern recognition. IEEE： 1597-1604 ［DOI： 10.1109/CVPR.2009.5206596http://dx.doi.org/10.1109/CVPR.2009.5206596］
Chen G， Shao F， Chai X， Jaing Q， Ho Y. 2023. Multi-Stage Salient Object Detection in 360° Omnidirectional Image Using Complementary Object-Level Semantic Information. IEEE Transactions on Emerging Topics in Computational Intelligence ［DOI： 10.1109/TETCI.2023.3259433http://dx.doi.org/10.1109/TETCI.2023.3259433］
Chen X L， Zhang P C， Lu Y B， and Cao B N. 2023. Saliency Detection of Panoramic Images Based on Robust Vision Transformer and Multiple Attention. Journal of Electronics & Information Technology， 45（6）： 2246-2255
陈晓雷， 张鹏程， 卢禹冰， 曹宝宁. 2023. 基于鲁棒视觉变换和多注意力的全景图像显著性检测.电子与信息学报，45（06）：2246-2255 ［doi： 10.11999/JEIT220684http://dx.doi.org/10.11999/JEIT220684］
Chen Z， Li Y， Zhang Y. 2018. Recent advances in omnidirectional video coding for virtual reality： Projection and evaluation. Signal Processing， 146： 66-78 ［DOI： https：//doi.org/10.1016/j.sigpro.2018.01.004http://dx.doi.org/https：//doi.org/10.1016/j.sigpro.2018.01.004］
Cong R， Huang K， Lei J， Zhou X， Meur O L. Multi-Projection Fusion and Refinement Network for Salient Object Detection in 360 degree Omnidirectional Image. IEEE Transactions on Neural Networks and Learning Systems， 2023.
Cong R， Zhang Y， Fang L， Li J， Zhao Y， and Kwong S. 2021. RRNet： Relational reasoning network with parallel multiscale attention for salient object detection in optical remote sensing images. IEEE Transactions on Geoscience and Remote Sensing， 60： 1-11 ［DOI： 10.1109/TGRS.2021.3123984http://dx.doi.org/10.1109/TGRS.2021.3123984］
Dai H， Bao L， Shen K， Zhou X， and Zhang J. 2023. 360∘ Omnidirectional Salient Object Detection with Multi-scale Interaction and Densely-Connected Prediction//International Conference on Image and Graphics. Cham： Springer Nature Switzerland： 427-438 ［DOI： https：//doi.org/10.1007/978-3-031-46305-1_35http://dx.doi.org/https：//doi.org/10.1007/978-3-031-46305-1_35］
De Boer P T， Kroese D P， Mannor S， Rubinstein R Y. 2005. A tutorial on the cross-entropy method. Annals of operations research， 134： 19-67 ［DOI： 10.1007/s10479-005-5724-zhttp://dx.doi.org/10.1007/s10479-005-5724-z］
Dejun F， Hongyu C， Suning L， Liao Z， Shen X， Xie Y， and Zhu J. 2023. Boundary-semantic collaborative guidance network with dual-stream feedback mechanism for salient object detection in optical remote sensing imagery. IEEE Transactions on Geoscience and Remote Sensing ［doi： 10.1109/TGRS.2023.3332282http://dx.doi.org/10.1109/TGRS.2023.3332282］
Fan D P， Cheng M M， Liu Y， Li T， and Borji A. 2017. Structure-measure： A new way to evaluate foreground maps//Proceedings of the IEEE international conference on computer vision： 4548-4557 ［DOI： 10.1109/ICCV.2017.487http://dx.doi.org/10.1109/ICCV.2017.487］
Fan D P， Gong C， Cao Y， Ren B， Cheng M M， and Borji A. 2018. Enhanced-alignment measure for binary foreground map evaluation［J］. arXiv preprint arXiv：1805.10421 ［DOI： https：//doi.org/10.48550/arXiv.1805.10421http://dx.doi.org/https：//doi.org/10.48550/arXiv.1805.10421］
FAN D P， JI G P， SUN G， Cheng M M， Shen J， and Shao L. 2020. Camouflaged object detection// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle， WA， USA： IEEE： 2777-2787 ［DOI： 10.1109/CVPR42600.2020.00285http://dx.doi.org/10.1109/CVPR42600.2020.00285］
He K， Zhang X， Ren S， and Sun J. 2016. Deep residual learning for image recognition//Proceedings of the IEEE conference on computer vision and pattern recognition： 770-778 ［DOI： 10.1109/CVPR.2016.90http://dx.doi.org/10.1109/CVPR.2016.90］
He W， Pan C. 2020. The salient object detection based on attention-guided network. Journal of Image and Graphics， 47（7）： 1575-1583
何伟， 潘晨. 2022.注意力引导网络的显著性目标检测［J］.中国图象图形学报，27（04）： 1176-1190［DOI：10.11834/jig.200658http://dx.doi.org/10.11834/jig.200658］
He Z， Shao F， Chen G， Chai X， and Ho Y. 2023. SCFANet： Semantics and Context Feature Aggregation Network for 360° Salient Object Detection. IEEE Transactions on Multimedia ［DOI： 10.1109/TMM.2023.3293994http://dx.doi.org/10.1109/TMM.2023.3293994］
HONG S， YOU T， KWAK S， and Han B. 2015. Online tracking by learning discriminative saliency map with convolutional neural network//Proceedings of International Conference on Machine Learning. ［S.l.］： PMLR：597-606 ［DOI： https：//doi.org/10.48550/arXiv.1502.06796http://dx.doi.org/https：//doi.org/10.48550/arXiv.1502.06796］
HOYER L， MUNOZ M， KATIYAR P， Khoreva A， and Fischer V. 2019. Grid saliency for context explanations of semantic segmentation. Advances in Neural Information Processing Systems， 32： 6462-6473 ［DOI： https：//doi.org/10.48550/arXiv.1907.13054http://dx.doi.org/https：//doi.org/10.48550/arXiv.1907.13054］
Huang M， Li G， Liu Z， and Zhu L. 2023. Lightweight Distortion-aware Network for Salient Object Detection in Omnidirectional Images. IEEE Transactions on Circuits and Systems for Video Technology ［DOI： 10.1109/TCSVT.2023.3253685http://dx.doi.org/10.1109/TCSVT.2023.3253685］
Huang M， Liu Z， Li G， et al. 2020. FANet： Features Adaptation Network for 360 degree Omnidirectional Salient Object Detection［J］. IEEE Signal Processing Letters， 27： 1819-1823 ［DOI： 10.1109/LSP.2020.3028192http://dx.doi.org/10.1109/LSP.2020.3028192］
Jiao J， Tang Y M， Lin K Y， Gao Y， Ma A J， Wang Y， amd Zhang W S. 2023. Dilateformer： Multi-scale dilated transformer for visual recognition. IEEE Transactions on Multimedia ［DOI： 10.1109/TMM.2023.3243616http://dx.doi.org/10.1109/TMM.2023.3243616］
Li G， Bai Z， Liu Z. 2023. Texture-semantic collaboration network for ORSI salient object detection. IEEE Transactions on Circuits and Systems II： Express Briefs ［DOI： https：//doi.org/10.48550/arXiv.2312.03548http://dx.doi.org/https：//doi.org/10.48550/arXiv.2312.03548］
Li G， Liu Z， Bai Z， Lin W， and Ling H. 2023. Lightweight salient object detection in optical remote-sensing images via semantic matching and edge alignment. IEEE Transactions on Geoscience and Remote Sensing， 61： 1-11 ［DOI： 10.1109/TGRS.2022.3145483http://dx.doi.org/10.1109/TGRS.2022.3145483］
Li J， Su J， Xia C， and Tian Y. 2019. Distortion-Adaptive Salient Object Detection in 360 degree Omnidirectional Images. IEEE Journal of Selected Topics in Signal Processing， 14（1）： 38-48 ［DOI： 10.1109/JSTSP.2019.2957982http://dx.doi.org/10.1109/JSTSP.2019.2957982］
Li J， Zhao Y， Ye W， Yu K， and Ge S. 2019. Attentive Deep Stitching and Quality Assessment for 360 degree Omnidirectional Images. IEEE Journal of Selected Topics in Signal Processing， 14（1）： 209-221 ［DOI： 10.1109/JSTSP.2019.2953950http://dx.doi.org/10.1109/JSTSP.2019.2953950］
Lin Y， Sun H， Liu N， Bian Y， Cen J， and Zhou H. 2022. A lightweight multi-scale context network for salient object detection in optical remote sensing images//2022 26th international conference on pattern recognition （ICPR）. IEEE： 238-244 ［DOI： https：//doi.org/10.48550/arXiv.2205.08959http://dx.doi.org/https：//doi.org/10.48550/arXiv.2205.08959］
Lin Y， Sun H， Liu N， Bian Y， Cen J， and Zhou H. 2022. Attention guided network for salient object detection in optical remote sensing images//International Conference on Artificial Neural Networks. Cham： Springer International Publishing： 25-36 ［https：//doi.org/10.48550/arXiv.2207.01755https://doi.org/10.48550/arXiv.2207.01755］
Liu G， Fan D. 2013. A model of visual attention for natural image retrieval//Proceedings of the IEEE Conference on Information Science and Cloud Computing Companion： 728-733 ［DOI： 10.1109/ISCC-C.2013.21http://dx.doi.org/10.1109/ISCC-C.2013.21］
Liu N， Zhang N， Wan K， Shao L， and Han J. 2021. Visual saliency transformer//Proceedings of the IEEE/CVF international conference on computer vision： 4722-4732 ［DOI： https：//doi.org/10.48550/arXiv.2104.12099http://dx.doi.org/https：//doi.org/10.48550/arXiv.2104.12099］
Luz G， Ascenso J， Brites C， and Pereira F. 2017. Saliency-driven omnidirectional imaging adaptive coding： Modeling and assessment//2017 IEEE 19th international workshop on multimedia signal processing （MMSP）. IEEE： 1-6 ［DOI： 10.1109/MMSP.2017.8122228http://dx.doi.org/10.1109/MMSP.2017.8122228］
Ma G， Li S， Chen C， Hao A， and Qin H. 2020. Stage-wise salient object detection in 360 omnidirectional image via object-level semantical saliency ranking. IEEE Transactions on Visualization and Computer Graphics， 26（12）： 3535-3545 ［DOI： 10.1109/TVCG.2020.3023636http://dx.doi.org/10.1109/TVCG.2020.3023636］
Máttyus G， Luo W， Urtasun R. 2017. Deeproadmapper： Extracting road topology from aerial images//Proceedings of the IEEE international conference on computer vision： 3438-3446 ［DOI： 10.1109/ICCV.2017.372http://dx.doi.org/10.1109/ICCV.2017.372］
Maugey T， Le Meur O， Liu Z. 2017. Saliency-based navigation in omnidirectional image//2017 IEEE 19th International Workshop on Multimedia Signal Processing （MMSP）. IEEE： 1-6 ［DOI： 10.1109/MMSP.2017.8122229http://dx.doi.org/10.1109/MMSP.2017.8122229］
Perazzi F， Krähenbühl P， Pritch Y， Hornung A. 2012. Saliency filters： Contrast based filtering for salient region detection//2012 IEEE conference on computer vision and pattern recognition. IEEE： 733-740 ［DOI： 10.1109/CVPR.2012.6247743http://dx.doi.org/10.1109/CVPR.2012.6247743］
Serrano A， Sitzmann V， Ruiz-Borau J， Wetzstein G， and Masia B. 2017. Movie editing and cognitive event segmentation in virtual reality video. ACM Transactions on Graphics （TOG）， 36（4）： 1-12 ［DOI： https：//doi.org/10.1145/3072959.3073668http://dx.doi.org/https：//doi.org/10.1145/3072959.3073668］
Tang B， Liu Z， Tan Y， and He Q. 2022. HRTransNet： HRFormer-driven two-modality salient object detection. IEEE Transactions on Circuits and Systems for Video Technology， 33（2）： 728-742 ［DOI： 10.1109/TCSVT.2022.3202563http://dx.doi.org/10.1109/TCSVT.2022.3202563］
Wei J， Wang S， and Huang Q. 2020. F³Net： fusion， feedback and focus for salient object detection//Proceedings of the AAAI conference on artificial intelligence， 34（07）： 12321-12328 ［DOI： https：//doi.org/10.48550/arXiv.1911.11445http://dx.doi.org/https：//doi.org/10.48550/arXiv.1911.11445］
Wei J， Wang S， Wu Z， Su C， Huang Q， and Tian Q. 2020. Label decoupling framework for salient object detection//Proceedings of the IEEE/CVF conference on computer vision and pattern recognition： 13025-13034 ［DOI： 10.1109/CVPR42600.2020.01304http://dx.doi.org/10.1109/CVPR42600.2020.01304］
Wu J， Xia C， Yu T， and Li J. 2022. View-aware Salient Object Detection for 360 ｛\deg｝ Omnidirectional Image. arXiv preprint arXiv：2209.13222 ［DOI： https：//doi.org/10.48550/arXiv.2209.13222http://dx.doi.org/https：//doi.org/10.48550/arXiv.2209.13222］
Xie C， Xia C， Ma M， Zhao Z， Chen X， and Li J. 2022. Pyramid grafting network for one-stage high resolution saliency detection//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition： 11717-11726 ［DOI： 10.1109/CVPR52688.2022.01142http://dx.doi.org/10.1109/CVPR52688.2022.01142］
Yao C， Feng L， Kong Y， Xiao L， and Chen T. 2023. Transformers and CNNs fusion network for salient object detection. Neurocomputing， 520： 342-355 ［DOI： https：//doi.org/10.1016/j.neucom.2022.10.081http://dx.doi.org/https：//doi.org/10.1016/j.neucom.2022.10.081］
Ye L， Liu Z， Li L， Shen L， Bai C， and Wang Y. 2017. Salient object segmentation via effective integration of saliency and objectness［J］. IEEE Transactions on Multimedia， 19（8）： 1742-1756 ［DOI： 10.1109/TMM.2017.2693022http://dx.doi.org/10.1109/TMM.2017.2693022］
Ye X Y， Zhu L， Wang W W and Fu Y. 2024. RGB_D salient object detection algorithm based on complementary information interaction. Journal of Image and Graphics，29（05）：1252-1264
叶欣悦， 朱磊， 王文武， 付云. 2024. 互补特征交互融合的RGB＿D实时显著目标检测.中国图象图形学报，29（05）：1252-1264［DOI：10.11834/jig.230583http://dx.doi.org/10.11834/jig.230583］
Yuan J， Zhu A， Xu Q， Wattanachote K， and Gong Y. 2023. CTIF-Net： A CNN-Transformer Iterative Fusion Network for Salient Object Detection. IEEE Transactions on Circuits and Systems for Video Technology ［DOI： 10.1109/TCSVT.2023.3321190http://dx.doi.org/10.1109/TCSVT.2023.3321190］
Yuan Y， Gao P， Tan X Y. 2023. M3Net： Multilevel， Mixed and Multistage Attention Network for Salient Object Detection. arXiv preprint arXiv：2309.08365 ［DOI： https：//doi.org/10.48550/arXiv.2309.08365http://dx.doi.org/https：//doi.org/10.48550/arXiv.2309.08365］
Yun Y K， Lin W. 2022. Selfreformer： Self-refined network with transformer for salient object detection. arXiv preprint arXiv：2205.11283 ［DOI： https：//doi.org/10.48550/arXiv.2205.11283http://dx.doi.org/https：//doi.org/10.48550/arXiv.2205.11283］
Zhang J， Zhang Q， Shen X， Wang X. 2023. Salient Object Detection on 360° Omnidirectional Image with Bi-Branch Hybrid Projection Network//2023 IEEE 25th International Workshop on Multimedia Signal Processing （MMSP）. IEEE： 1-5 ［DOI： 10.1109/MMSP59012.2023.10337695http://dx.doi.org/10.1109/MMSP59012.2023.10337695］
Zhang Y， Hamidouche W， Deforges O. 2022. Channel-Spatial Mutual Attention Network for 360° Salient Object Detection//2022 26th International Conference on Pattern Recognition （ICPR）. IEEE： 3436-3442 ［DOI： 10.1109/ICPR56361.2022.9956354http://dx.doi.org/10.1109/ICPR56361.2022.9956354］
Zhao J X， Liu J J， Fan D P， Cao Y， Yang J， and Cheng M M. EGNet： Edge guidance network for salient object detection//Proceedings of the IEEE/CVF international conference on computer vision. 2019： 8779-8788 ［DOI： 10.1109/ICCV.2019.00887http://dx.doi.org/10.1109/ICCV.2019.00887］
ZHAO R， OYANG W， WANG X. 2017. Person re-identification by saliency learning. IEEE Transactions on Pattern Analysis and Machine Intelligence， 39（2）： 356-370 ［DOI： 10.1109/TPAMI.2016.2544310http://dx.doi.org/10.1109/TPAMI.2016.2544310］
Zhao X， Pang Y， Zhang L， Lu H， and Zhang L. 2020. Suppress and balance： A simple gated network for salient object detection//Computer Vision–ECCV 2020： 16th European Conference， Glasgow， UK， August 23–28， 2020， Proceedings， Part II 16. Springer International Publishing： 35-51 ［DOI： https：//doi.org/10.48550/arXiv.2007.08074http://dx.doi.org/https：//doi.org/10.48550/arXiv.2007.08074］
Zhao Y， Zhao L， Yu Q， Sheng L， Zhang J， and Xu D. 2023. Distortion-aware Transformer in 360° Salient Object Detection//Proceedings of the 31st ACM International Conference on Multimedia： 499-508 ［DOI： https：//doi.org/10.1145/3581783.3612025http://dx.doi.org/https：//doi.org/10.1145/3581783.3612025］
Zhou X， Zhang Y， Li N， Wang X， Zhou Y， and Ho Y. 2021. Projection invariant feature and visual saliency-based stereoscopic omnidirectional image quality assessment. IEEE Transactions on Broadcasting， 67（2）： 512-523 ［DOI： 10.1109/TBC.2021.3056231http://dx.doi.org/10.1109/TBC.2021.3056231］
相关作者
相关机构
京公网安备11010802024621