基於全卷積網路之光場影像反射分離

以作者查詢圖書館館藏

、以作者查詢臺灣博碩士

、以作者查詢全國書目

、勘誤回報

、線上人數：94

、訪客IP：18.227.190.228

姓名

張瑞宇(Ruei-Yu Chang) 查詢紙本館藏

畢業系所

通訊工程學系

論文名稱

基於全卷積網路之光場影像反射分離
(Fully Convolutional Networks Based Reflection Separation for Light Field Images)

相關論文

★ 應用於車內視訊之光線適應性視訊壓縮編碼器設計	★ 以粒子濾波法為基礎之改良式頭部追蹤系統
★ 應用於空間與CGS可調性視訊編碼器之快速模式決策演算法	★ 應用於人臉表情辨識之強健式主動外觀模型搜尋演算法
★ 結合Epipolar Geometry為基礎之視角間預測與快速畫面間預測方向決策之多視角視訊編碼	★ 基於改良式可信度傳遞於同質區域之立體視覺匹配演算法
★ 以階層式Boosting演算法為基礎之棒球軌跡辨識	★ 多視角視訊編碼之快速參考畫面方向決策
★ 以線上統計為基礎應用於CGS可調式編碼器之快速模式決策	★ 適用於唇形辨識之改良式主動形狀模型匹配演算法
★ 以運動補償模型為基礎之移動式平台物件追蹤	★ 基於匹配代價之非對稱式立體匹配遮蔽偵測
★ 以動量為基礎之快速多視角視訊編碼模式決策	★ 應用於地點影像辨識之快速局部L-SVMs群體分類器
★ 以高品質合成視角為導向之快速深度視訊編碼模式決策	★ 以運動補償模型為基礎之移動式相機多物件追蹤

檔案

[Endnote RIS 格式]

[Bibtex 格式]

[相關文章]

[文章引用]

[完整記錄]

[館藏目錄]

[檢視]

[下載]

本電子論文使用權限為同意立即開放。
已達開放權限電子全文僅授權使用者為學術研究之目的，進行個人非營利性質之檢索、閱讀、列印。
請遵守中華民國著作權法之相關規定，切勿任意重製、散佈、改作、轉貼、播送，以免觸法。

摘要(中)

現有基於多視角的反射分離(reflection separation)方案，因其並非針對窄基線(narrow baseline)而設計，並不適用於光場影像，又目前僅有的少數光場影像的反射分離方案，皆須先對中心視角估測視差圖(disparity map)。因此，本論文採用基於不含反射之光場影像視差估測所設計之EPINET，對弱反射之混合光場影像進行反射分離，透過訓練以多視角影像堆疊(image stack)為輸入的全卷積網路，以端對端(end-to-end)的方式學習主要方向的多視角背景層之主要卷積特徵，合併後再以全卷積網路，對中央視角直接進行以像素為單位的背景層影像估測。此外，本論文分析光場相機拍攝真實世界中的混合光場影像，其背景層與反射層於各視角間會有不同的位移量，基於此現象建構出滿足真實條件之混合光場影像資料集。實驗結果顯示，採用EPINET輔以本論文提出之混合光場影像資料集，能有效地對合成之混合光場影像重建背景層。

摘要(英)

Existing reflection separation schemes designed for multi-view images cannot be applied to light filed images due to the dense light fields with narrow baselines. In order to improve accuracy of the reconstructed background (i.e., the transmitted layer), most light field data based reflection separation schemes estimate a disparity map before reflection separation. Different from previous work, this thesis uses the existing EPINET based on disparity estimation of light field image without reflection, and separates the mixed light field image of weak reflection. At the training stage, the network takes multi-view images stacks along principle directions of light field data as inputs, and significant convolution features of the background layer are learned in an end-to-end manner. Then the FCN learns to predict pixel-wise gray-scale values of the background layer of the central view. Our experimental results show that the background layer can be reconstructed effectively by using EPINET and the mixed light field image dataset proposed in this thesis.

關鍵字(中)

★ 光場影像
★ 混合影像
★ 反射分離
★ 深度學習
★ 全卷積網路

關鍵字(英)

論文目次

摘要 I
Abstract II
目錄 III
圖目錄 V
表目錄 VI
第一章緒論 1
1.1 前言 1
1.2 研究動機 1
1.3 研究方法 2
1.4 論文架構 2
第二章非光場影像之反射分離方案技術介紹 3
2.1基於非深度學習之反射分離方案 3
2.1.1 基於單影像之反射分離方案 3
2.2基於深度學習之反射分離方案 4
2.2.1 現有反射分離方案採用的網路架構 4
2.2.2 基於深度學習之反射分離技術現況 7
2.3 總結 9
第三章光場影像之反射分離方案技術現況 10
3.1基於非深度學習之光場反射分離方案 10
3.2採用深度學習估測視差值之光場反射分離方案 10
3.3 總結 11
第四章本論文所提之光場影像反射分離方案 12
4.1 現有EPINET之網路架構 12
4.2合成混合光場資料集 18
4.3 總結 20

第五章實驗結果與討論 21
5.1 實驗參數與測試資料集規格 21
5.1.1 測試資料集 21
5.1.2參數設定 27
5.2 反射分離實驗結果 30
5.3 總結 36
第六章結論與未來展望 37
參考文獻 38

參考文獻

[1] A. Levin and Y. Weiss, “User assisted separation of reflections from a single image using a sparsity prior,” in Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), vol. 29, pp.1647-1654, Sept. 2007.
[2] S.-K. Yeung, T.-P. Wu, and C.-K. Tang, “Extracting smooth and transparent layers from a single image,” in Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 1-7, June 2008.
[3] X. Guo, X. Cao, and Y. Ma, “Robust separation of reflection from multiple images,” in Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 2187-2194, June 2014.
[4] Y. Li and M. S. Brown, “Exploiting reflection change for automatic reflection removal, ” in Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 2432-2439, March 2014.
[5] Y. Ni, J. Chen, and L.-P. Chau, “Reflection removal based on single light field capture,” in Proc. IEEE International Symposium on Circuits and Systems (ISSCS), pp. 1-4, May 2017.
[6] T. Li and D.P.-K. Lun, “Robust Reflection Removal Based on Light Field Imaging,” IEEE Transactions on Image Processing, pp.1798-1812, vol. 28, April 2019.
[7] P. Chandramouli, M. Noroozi, and P. Favaro, “ConvNet-Based Depth Estimation, Reflection Separation and Deblurring of Plenoptic Images,” in Proc. Asian Conference on Computer Vision (ACCV), pp. 129-144, 2016.
[8] C. Shin, H.-G. Jeon, Y. Yoon, I. S. Kweon, S. J. Kim, “EPINET: A fully-convolutional neural network using epipolar geometry for depth from light field images,” in Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 4748-4757, June 2018.
[9] H. Farid and E. Adelson, “Separating reflections and lighting using independent components analysis,” in Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), vol. 1, pp. 262-267, June 1999.
[10] A. Agrawal, R. Raskar, S. Nayar, and Y. Li, “Removing photography artifacts using gradient projection and flashexposure sampling,” in ACM Transactions on Graphics, vol. 24, pp. 828-835, July 2005.
[11] Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner, “Gradient-based learning applied to document recognition, ” in Proc. IEEE, vol. 86, no. 11, pp. 2278-2324, Nov. 1998.
[12] J. Long, E. Shelhamer, T. Darrell, “Fully convolutional networks for semantic segmentation, ” in Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 3431-3440, June 2015.
[13] G. E. Hinton and R. Salakhutdinov, “Reducing the dimensionality of data with neural networks,” Science, vol. 313, no. 5786, pp. 504-507, July 2006.
[14] Q. Fan, J. Yang, G. Hua, B. Chen, and D. Wipf. “A generic deep architecture for single image reflection removal and image smoothing,” in Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 22-29, Oct. 2017.
[15] R. Wan, B. Shi, L.Y. Duan, A.H. Tan, and A.C. Kot, “CRNN: Multi-scale guided concurrent reflection removal network,” in Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 4777-4785, June 2018.
[16] P. Wieschollek, O. Gallo, J. Gu, and J. Kautz, “Separating reflection and transmission images in the wild,” in Proc. European Conference on Computer Vision (ECCV), pp. 89-104, Sept. 2018.
[17] Y. Ni, J. Chen, and L. Chau, “Reflection removal on single light field capture using focus manipulation,” IEEE Transactions on Computational Imaging, vol. 4, pp. 562-572, July 2018.
[18] S. Wanner and B. Goldluecke, “Reconstructing reflective and transparent surfaces from epipolar plane images,” in Proc. German Conference on Pattern Recognition (GCPR), vol. 8142, pp. 1-10, Sept. 2013.
[19] Q. Wang, H. Lin, Y. Ma, S. Kang, and J. Yu, “Automatic layer separation using light field imaging,” arXiv preprint arXiv:1506.04721 [cs.CV], June 2015.
[20] O. Johannsen, A. Sulc, and B. Goldluecked, “what sparse light field coding reveals about scene structure,” in Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 3262-3270, June 2016.
[21] H.-G. Jeon, J. Park, G. Choe, J. Park, Y. Bok, Y.-W. Tai, and I. S. Kweon, “Depth from a light field image with learning based matching costs,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 41, pp. 297-310, Jan. 2018.
[22] K. Honauer, O. Johannsen, D. Kondermann, and B. Goldluecke, “A dataset and evaluation methodology for depth estimation on 4D light fields,” Asian Conference on Computer Vision. Lecture Notes in Computer Science, vol. 10113, pp. 19-34, March 2016.
[23] Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli, “Image quality assessment: From error visibility to structural similarity,” IEEE Transactions on Image Processing, vol. 13, no. 4, pp. 600-612, April 2004.

指導教授

唐之瑋(Chih-Wei Tang)

審核日期

2019-7-30

推文