在本文中，我們提出了一種高效率的源感知域增強和適應（SDEA）方法，以提高基於卷積神經網絡（CNN）物件分割方法的準確性。首先，我們先找出源元素，例如落葉，人孔蓋，卷雲和廣告，這些元素通常會導致無效的物件分割。然後，我們使用包含這些源元素的場景創建一個新的類GTA5數據集《俠盜獵車手5》。
此外，我們對創建的類似GTA5數據集執行自適應，以生成逼真的類似GTA5數據集，即GTA5_s^{SDEA}。無需重新標記GTA5_s^{SDEA}的像素註釋，我們將GTA5_s^{SDEA}與實際數據集Camvid結合在一起，構成了一個新的增強數據集，用於訓練現有的基於CNN的物件分割方法，從而實現了較大的分割準確率。 全面的實驗結果表明，通過將我們的SDEA方法應用於FCN（完全卷積網絡），SegNet-basic，AdaptSegNet和Gated-AdaptSegNet上的現有對象分割方法，可以提供實質性的準確性改進，從而提供更可靠的道路，天空和建築物信息 應用於自動駕駛輔助系統（ADAS）。

In this thesis, we propose an effective source-aware domain enhancement and adaptation (SDEA) approach to increase the accuracy of the existing convolutional neural network-based (CNN-based) object segmentation methods. We first scoop out the source elements, such as the falling-leaves, manhole covers, cirrus clouds, and advertisements, which often cause invalid object segmentation. Then, we create a new GTA5-like (Grand Theft Auto V-like) dataset with the scenarios including these source elements. Furthermore, we perform a domain adaptation on the created GTA5-like dataset to generate a photo-realistic GTA5-like dataset, namely GTA5_s^{SDEA}. Without the need to relabel the pixel-annotations of GTA5_s^{SDEA}, we combine GTA5_s^{SDEA} with the realistic dataset, namely Camvid, to constitute a newly enhanced dataset for training the existing CNN-based object segmentation methods, achieving substantial segmentation accuracy improvement. The comprehensive experimental results have demonstrated the substantial accuracy improvement merit by applying our SDEA approach to existing object segmentation methods on FCN (Fully Convolutional Networks), SegNet-basic, AdaptSegNet, and Gated-AdaptSegNet, providing more reliable road, sky, and building information to the applications of automatic driving assistance systems (ADAS).

[1] V. Badrinarayanan, A. Kendall, and R. Cipolla, “SegNet: A deep convolutional encoder-decoder architecture for image segmentation,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39, no. 12, pp. 2481-2495, Dec. 2017.
[2] J. Long, E. Shelhamer, and T. Darrell, “Fully convolutional networks for semantic segmentation,” IEEE Conference on Computer Vision and Pattern Recognition, Boston, MA, USA, 2015, pp. 34313440.
[3] O. Ronneberger, P. Fischer, and T. Brox, “U-net: Convolutional networks for biomedical image segmentation,” International Conference on Medical Image Computing and Computer Assisted Intervention, Munich, Germany, 2015, pp. 234–241.
[4] Y. Tsai, W. Hung, S. Schulter, K. Sohn, M. Yang, and M. Chandraker, “Learning to adapt structured output space for semantic segmentation,” IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA, 2018, pp. 7472-7481.
[5] Y. Lin, D. Tan, W. Cheng, and K. Hua, “Adapting semantic segmentation of urban scenes via maskaware gated discriminator,” IEEE International Conference on Multimedia and Expo, Shanghai, China, 2019, pp. 218-223.
[6] F. S. Saleh, M. S. Aliakbarian, M. Salzmann, L. Petersson, and J. M. Alvarez, “Effective use of synthetic data for urban scene semantic segmentation,” European Conference on Computer Vision, Munich, Germany, 2018, pp. 86-103.
[7] E. Shelhamer, J. Long, and T. Darrell, “Fully convolutional networks for semantic segmentation,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39, no. 4, pp. 640-651, Apr. 2017.
[8] G. J. Brostow, J. Fauqueur, and R. Cipolla, “Semantic object classes in video: A high-definition ground truth database,” Pattern Recognition Letters, vol. 30, no. 2, pp. 88–97, Jan. 2009.
[9] M. Cordts, M. Omran, S. Ramos, T. Rehfeld, M. Enzweiler, R. Benenson, U. Franke, S. Roth, and B. Schiele, “The cityscapes dataset for semantic urban scene understanding,” IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA, 2016, pp. 3213-3223.
[10] A. Geiger, P. Lenz, C. Stiller, and R. Urtasun, “Vision meets robotics: The KITTI dataset,” The International Journal of Robotics Research, vol. 32, no. 11, pp. 1231–1237, Sep. 2013.
[11] B. C. Russell, A. Torralba, K. P. Murphy, and W. T. Freeman, “LabelMe: a database and web-based tool for image annotation,” International Journal of Computer Vision, vol. 77, no. 1-3, pp. 157-173, May 2008.
[12] S. R. Richter, V. Vineet, S. Roth, and V. Koltun, “Playing for data: Ground truth from computer games,” European Conference on Computer Vision, Amsterdam, Netherlands, 2016, pp. 102-118.
[13] Y. Zhang, Z. Qiu, T. Yao, D. Liu, and T. Mei, “Fully convolutional adaptation networks for semantic segmentation,” IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA, 2018, pp. 6810-6818.
[14] D. Ding, C. Lee, and K. Lee, “An adaptive road ROI determination algorithm for lane detection,” IEEE International Conference of IEEE Region 10, Xi’an, China, 2013, pp. 1-4.
[15] C. Lee and J. Moon, “Robust lane detection and tracking for real-time applications,” IEEE Transactions on Intelligent Transportation Systems, vol. 19, no. 12, pp. 4043-4048, Dec. 2018.
[16] W. Song, Y. Yang, M. Fu, Y. Li, and M. Wang, “Lane detection and classification for forward collision warning system based on stereo vision,” IEEE Sensors Journal, vol. 18, no. 12, pp. 5151-5163, Jun.
2018.
[17] C. Wu, L. Wang, and K. Wang, “Ultra-low complexity block-based lane detection and departure warning system,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 29, no. 2, pp.
582-593, Feb. 2019.
[18] C. Chang, J. Zhao, and L. Itti, “DeepVP: deep learning for vanishing point detection on 1 million street view images,” IEEE International Conference on Robotics and Automation, Brisbane, Australia, 2018, pp. 4496-4503.
[19] H. Hwang, G. Yoon, and S. Yoon, “Optimized clustering scheme-based robust vanishing point detection,” IEEE Transactions on Intelligent Transportation Systems, vol. 21, no. 1, pp. 199-208, Jan.
2020.
[20] K. He, G. Gkioxari, P. Dollár, and R. Girshick, “Mask R-CNN,” IEEE International Conference on Computer Vision, Venice, Italy, 2017, pp. 2980-2988.
[21] Y. Li, M. Liu, X. Li, M. Yang, and J. Kautz, “A closed-form solution to photorealistic image stylization”, European Conference on Computer Vision, Munich, Germany, 2018, pp. 453-468.
[22] Execution code. Accessed: 26 Mar. 2020. [Online]. Available: ftp://140.118.175.164/Images.
[23] R. Girshick, J. Donahue, T. Darrell, and J. Malik, “Rich feature hierarchies for accurate object detection and semantic segmentation,” IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, USA, 2014, pp. 580-587.
[24] R. Girshick, ”Fast R-CNN,” IEEE International Conference on Computer Vision, Santiago, Chile, 2015, pp. 1440-1448.
[25] S. Ren, K. He, R. Girshick, and J. Sun, “Faster R-CNN: Towards real-time object detection with region proposal networks,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39, no. 6, pp. 1137-1149, Jun. 2017.