CCF第38届中国计算机应用大会(CCF NCCA 2023)暨2023年人工智能+应用学术会议征文通知.pdfVIP

于 营，等：语义分割评价指标和评价方法综述 2023 ，59（6 ） 69 segmentation of road- driving images[C]//Proceedings of ence on Computer Vision and Pattern Recognition（CVPR ）， 2019 IEEE Conference on Computer Vision and Pattern 2016 ：2574-2582. Recognition （CVPR ），2019 ：12607-12616. [83] PAPERNOT N ，MCDANIEL P ，JHA S ，et al.The limita- [80] KALODIMAS P ，NIKITAKIS A ，PAPAEFSTATHIOU I. tions of deep learning in adversarial settings[C]//Proceed- An open-source high-throughput ，reduced memory foot- ings of the 2016 IEEE European Symposium on Secu- print，face detection，pose estimation and landmark local- rity and Privacy （EuroSP ），2016 ：372-387. ization system[C]//Proceedings of the 2019 22nd Euro- [84] CARLINI N，WAGNER D.Towards evaluating the robust- micro Conference on Digital System Design（DSD），2019. ness of neural networks[C]//Proceedings of the 2017 IEEE [81] CHEN C ，DOU Q ，CHEN H ，et al.Semantic-aware gen- Symposium on Security and Privacy（SP），2017 ：39-57. erative adversarial nets for unsupervised domain adap- [85] KURAKIN A，GOODFELLOW I，BENGIO S.Adversarial tation in chest X-Ray segmentation[C]//Proceedings of the Machine Learning in Medical Imaging.Cham ：Springer machine learning at scale[J].arXiv ：1611.01236 ，2016. International Publishing ，2018. [86] ARNAB A，MIKSIK O，TORR P H S.On the robustness [82] MOOSAVI-DEZFOOLI S，FAWZI A，FROSSARD P.Deep- of semantic segmentation models to adversarial attacks[J]. Fool：a simple and accurate method to fool deep neu- IEEE Transactions on Pattern Analysis and Machine Intelli- ral networks[C]/