一種視頻壓縮感知中兩級多假設(shè)重構(gòu)及實現(xiàn)方法
doi: 10.11999/JEIT161142
-
2.
(華南理工大學(xué)電子與信息學(xué)院 廣州 510640) ②(華為技術(shù)有限公司 深圳 518129)
基金項目:
國家自然科學(xué)基金(61471173)，廣東省自然科學(xué)基金(2016A030313455)
A Two-stage Multi-hypothesis Reconstruction and Two Implementation Schemes for Compressed Video Sensing
-
2.
(School of Electronic and Information Engineering, South China University of Technology, Guangzhou 510640, China)
Funds:
The National Natural Science Foundation of China (61471173), The Natural Science Foundation of Guangdong Province (2016A030313455)
-
摘要: 視頻壓縮感知在采集端資源受限的視頻采集應(yīng)用場景有重要研究意義。重構(gòu)算法是視頻壓縮感知的關(guān)鍵技術(shù)，基于多假設(shè)預(yù)測的預(yù)測-殘差重構(gòu)框架具有良好的重構(gòu)性能。但現(xiàn)有的多假設(shè)預(yù)測算法大多在觀測域提出，這種預(yù)測方法由于受到不重疊分塊的限制，造成了預(yù)測幀的塊效應(yīng)，降低了重構(gòu)質(zhì)量。針對此問題，該文將像素域多假設(shè)預(yù)測與觀測域多假設(shè)預(yù)測相結(jié)合，提出兩級多假設(shè)重構(gòu)思想(2sMHR)，并設(shè)計了基于圖像組(Gw_2sMHR)和基于幀(Fw_2sMHR)的兩種實現(xiàn)方法。仿真結(jié)果表明，所提2sMHR重構(gòu)算法能有效減小塊效應(yīng)，相比于現(xiàn)有最好的多假設(shè)預(yù)測算法具有更低的時間復(fù)雜度和更高的視頻重構(gòu)質(zhì)量。Abstract: Compressed Video Sensing (CVS) has great significance to the scenarios with a resource-deprived video acquisition side. Reconstruction algorithm is the key technique in compressed video sensing. The Multi-Hypothesis (MH) prediction based prediction-residual reconstruction framework has good reconstruction performance. However, most of the existing multi-hypothesis prediction algorithms are proposed in measurement domain, which cause block artifacts in the predicted frames and decrease reconstruction accuracy due to the restriction of non-overlapping block partitioning. To address this issue, this paper proposes a two-stage Multi-Hypothesis Reconstruction (2sMHR) idea by incorporating the measurement-domain MH prediction with pixel-domain MH prediction. Two implementation schemes, GOP-wise (Gw) and Frame-wise (Fw) scheme, are designed for the 2sMHR. Simulation results show that the proposed 2sMHR algorithm can effectively reduce block artifacts and obtain higher video reconstruction accuracy while requiring lower computational complexity than the state-of- the-art CVS prediction methods.
-
Key words:
- Compressed Video Sensing (CVS) /
- Reconstruction /
- Prediction /
- Multi-Hypothesis (MH) /
- Sparsity
-
LIU Y and PADOS D A. Compressed-sensed-domain L1-PCA video surveillance[J]. IEEE Transactions on Multimedia, 2016, 18(3): 351-363. doi: 10.1109/TMM.2016. 2514848. GUO J, SONG B, and DU X. Significance evaluation of video data over media cloud based on compressed sensing[J]. IEEE Transactions on Multimedia, 2016, 18(7): 1297-1304. doi: 10.1109/TMM.2016.2564100. REHMAN A U, SHAH G A, and TAHIR M. Compressed sensing based adaptive video coding for resource constrained devices[C]. IEEE International Wireless Communications and Mobile Computing Conference, Paphos, Cyprus, 2016: 170-175. WANG J, GUPTA M, and SANKARANARAYANAN A C. LiSensA scalable architecture for video compressive sensing[C]. IEEE International Conference on Computational Photography, Houston, TX, 2015: 1-9. LLULL P, LIAO X J, YUAN X, et al. Coded aperture compressive temporal imaging[J]. Optics Express, 2013, 21(9): 10526-10545. doi: 10.1364/OE.21.010526. HOSSEINI M S and PLATANIOTIS K N. High-accuracy total variation with application to compressed video sensing [J]. IEEE Transactions on Image Processing, 2014, 23(9): 3869-3884. doi: 10.1109/TIP.2014.2332755. YANG J B, YUAN X, LIAO X J, et al. Video compressive sensing using Gaussian mixture models[J]. IEEE Transactions on Image Processing, 2014, 23(11): 4863-4878. doi: 10.1109/TIP.2014.2344294. 常侃, 覃團發(fā), 唐振華. 基于聯(lián)合總變分最小化的視頻壓縮感知重建算法[J]. 電子學(xué)報, 2014, 42(12): 2415-2421. doi: 10.3969/j.issn.0372-2112.2014.12.012. CHANG K, QIN T F, and TANG Z H. Reconstruction algorithm for compressed sensing of video based on joint total variation minimization[J]. Acta Electronica Sinica, 2014, 42(12): 2415-2421. doi: 10.3969/j.issn.0372-2112.2014.12.012. ZHAO C, MA S W, ZHANG J, et al. Video compressive sensing reconstruction via reweighted residual sparsity[J]. IEEE Transactions on Circuits Systems for Video Technology, 2016, to be published. doi: 10.1109/TCSVT. 2016.2527181. MUN S and FOWLER J E. Residual reconstruction for block-based compressed sensing of video[C]. IEEE Data Compression Conference, Snowbird, 2011: 183-192. NARAYANAN S and MAKUR A. Compressive coded video compression using measurement domain motion estimation [C]. IEEE International Conference on Electronics, Computing and Communication Technologies, Bangalore, 2014: 1-6. GUO J, SONG B, LIU H X, et al. Motion estimation in measurement domain for compressed video sensing[C]. IEEE International Conference on Computer and Information Technology, Xi,an, 2014: 441-445. DO T T, CHEN Y, NGUYEN D T, et al. Distributed compressed video sensing[C]. IEEE International Conference on Image Processing, Cairo, 2009: 1393-1396. TRAMEL E W and FOWLER J E. Video compressed sensing with multihypothesis[C]. IEEE Data Compression Conference, Snowbird, 2011: 193-202. AZGHANI M, KARIMI M, and MARVASTI F. Multihypothesis compressed video sensing technique[J]. IEEE Transactions on Circuits Systems for Video Technology, 2016, 26(4): 627-635. doi: 10.1109/TCSVT.2015. 2418586. CHEN J, CHEN Y, QIN D, et al. An elastic net-based hybrid hypothesis method for compressed video sensing[J]. Multimedia Tools Applications, 2013, 74(6): 2085-2108. doi: 10.1007/s11042-013-1743-y. KUO Y H, WU K, and CHEN J. A scheme for distributed compressed video sensing based on hypothesis set optimization techniques[J]. Multidimensional Systems and Signal Processing, 2017, 28(1): 129-148. doi: 10.1007/s11045- 015-0337-4. GAN L. Block compressed sensing of natural images[C]. IEEE International Conference on Digital Signal Processing, Cardiff, 2007: 403-406. OU W F, YANG C L, LI W H, et al. A two-stage multi- hypothesis reconstruction scheme in compressed video sensing[C]. IEEE International Conference on Image Processing, Phoenix, AZ, USA, 2016: 2494-2498. 楊春玲, 歐偉楓. CVS中基于多參考幀的最優(yōu)多假設(shè)預(yù)測算法[J]. 華南理工大學(xué)學(xué)報(自然科學(xué)版), 2016, 44(1): 1-8. doi: 10.3969/j.issn.1000-565X.2016.01.001. YANG C L and OU W F. Multi-reference frames-based optimal multi-hypothesis prediction in compressed video sensing[J]. Journal of South China University of Technology (Natural Science Edition), 2016, 44(1): 1-8. doi: 10.3969/ j.issn.1000-565X.2016.01.001. MUN S and FOWLER J E. Block compressed sensing of images using directional transforms[C]. IEEE International Conference on Image Processing, Cairo, 2009: 3021-3024. -
計量
- 文章訪問數(shù): 1300
- HTML全文瀏覽量: 228
- PDF下載量: 301
- 被引次數(shù): 0
下載:
