The DOI matches its Crossref record; 27 of the 27 references it cites were found in a registry or an index.
A Duality Based Approach for Realtime TV-L 1 Optical Flow
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“A reasonable explanation is that the temporal network is actually designed to capture different visual aspects of videos compared to the spatial network, which indicates that the feature distributions are more diverse between those fields (e.g., optical flow and RGB) [20], [24], [25].”
From Unsupervised Deep Video Hashing via Balanced Code for Large-Scale Video Retrieval (2019) · cites this paper as [24]
Citations
Cites
Nonlinear total variation based noise removal algorithms
Rudin, L., Osher, S., & Fatemi, E. (1992) · Physica D Nonlinear Phenomena · doi:10.1016/0167-2789(92)90242-f · Cited by 17,214
Horn, B. K. P., & Schunck, B. G. (1981) · Artificial Intelligence · doi:10.1016/0004-3702(81)90024-2 · Cited by 10,312
Horn, B. K. P., & Schunck, B. G. (1981) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.965761 · Cited by 7,443
chambolle (2004) · doi:10.1023/b:jmiv.0000011321.19549.88 · Cited by 3,227
High Accuracy Optical Flow Estimation Based on a Theory for Warping
Brox, T., Bruhn, A., Papenberg, N., & Weickert, J. (2004) · Lecture notes in computer science · doi:10.1007/978-3-540-24673-2_3 · Cited by 2,634
A computational framework and an algorithm for the measurement of visual motion
Anandan, P. (1989) · International Journal of Computer Vision · doi:10.1007/bf00158167 · Cited by 1,073
Thormaehlen, T., Broszio, H., & Weißenfeld, A. (2004) · Lecture notes in computer science · doi:10.1007/b97865 · Cited by 901
Fast Global Minimization of the Active Contour/Snake Model
Bresson, X., Esedoḡlu, S., Vandergheynst, P., Thiran, J., & Osher, S. (2007) · Journal of Mathematical Imaging and Vision · doi:10.1007/s10851-007-0002-0 · Cited by 857
A Nonlinear Primal-Dual Method for Total Variation-Based Image Restoration
Chan, T. F.-C., Golub, G. H., & Mulet, P. (1999) · SIAM Journal on Scientific Computing · doi:10.1137/s1064827596299767 · Cited by 849
An Algorithm for Total Variation Minimization and Applications
Chambolle, A. (2004) · Journal of Mathematical Imaging and Vision · doi:10.1023/b:jmiv.0000011325.36760.1e · Cited by 835
Cited by
Sorted by citations
Items 101–200 of 1,002
Dual-Inception Network for Cross-Database Micro-Expression Recognition
Zhou, L. H., Mao, Q., & Xue, L. (2019) · International Conference on Automatic Face and Gesture Recognition · doi:10.1109/fg.2019.8756579 · Cited by 171
“For each micro-expression sample, there are twodimensional optical flow feature images named horizontal and vertical components which are generated from the TVL1 method.”
MethodInfluentialUMATracker: an intuitive image-based tracking platform
Yamanaka, O., & Takeuchi, R. (2018) · Journal of Experimental Biology · doi:10.1242/jeb.182469 · Cited by 176
“The proposed software supports the implementation of four tracking algorithms: RMOT (Yoon et al., 2015), Optical Flow DualTVL1 (Zach et al., 2007; Sánchez Pérez et al., 2013), K-means (Lloyd, 1982; Arthur and Vassilvitskii, 2007) and GroupTracker (Fukunaga et al., 2015).”
MethodStereoscopic Scene Flow Computation for 3D Motion Understanding
Wedel, A., Brox, T., Vaudrey, T., Rabe, C., Franke, U., & Cremers, D. (2011) · International Journal of Computer Vision · doi:10.1007/s11263-010-0404-0 · Cited by 197
“Realtime methods have been proposed in Bruhn et al. (2005), Zach et al. (2007).”
MethodWho's Better? Who's Best? Pairwise Deep Ranking for Skill Determination
Doughty, H., Damen, D., & Mayol-Cuevas, W. W. (2018) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2018.00634 · Cited by 172
“To extract the optical flow frames for the temporal network we use the TV − L 1 algorithm [34].”
MethodLate Temporal Modeling in 3D CNN Architectures with BERT for Action Recognition
Kalfaoglu, M. E., Kalkan, S., & Alatan, A. A. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-68238-5_48 · Cited by 170
Convolutional Relational Machine for Group Activity Recognition
Azar, S. M., Atigh, M. G., Nickabadi, A., & Alahi, A. (2019) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2019.00808 · Cited by 158
“We use TVL1 algorithm [28] to compute these optical flow fields.”
MethodDMC-Net: Generating Discriminative Motion Cues for Fast Compressed Video Action Recognition
Shou, Z., Lin, X., Kalantidis, Y., Sevilla-Lara, L., Rohrbach, M., Chang, S., & Yan, Z. (2019) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2019.00136 · Cited by 156
Tao, R., Pan, Z., Das, R. K., Qian, X., Shou, M. Z., & Li, H. (2021) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3474085.3475587 · Cited by 176
Is Someone Speaking? Exploring Long-term Temporal Features for Audio-visual Active Speaker Detection
Tao, R., Pan, Z., Das, R. K., Qian, X., Shou, M. Z., & Li, H. (2021) · arXiv · doi:10.48550/arxiv.2107.06592 · Cited by 156
Video Modeling With Correlation Networks
Wang, H., Tran, D., Torresani, L., & Feiszli, M. (2020) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr42600.2020.00043 · Cited by 158
Robust Real-Time Visual Odometry with a Single Camera and an IMU
Kneip, L., Chli, M., & Siegwart, R. (2011) · British Machine Vision Conference (BMVC) · doi:10.5244/c.25.16 · Cited by 152
“More recently, the computer vision literature has seen some very powerful, yet computationally demanding methods for optical flow as for example [24].”
Temporal Action Localization by Structured Maximal Sums
Yuan, Z., Stroud, J., Lü, T., & Deng, J. (2017) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2017.342 · Cited by 148
“The inputs to Motion-CNN are dense optical flow channels computed by the TVL1 optical flow algorithm [30].”
MethodThe Role of Machine Vision for Intelligent Vehicles
Ranft, B., & Stiller, C. (2016) · IEEE Transactions on Intelligent Vehicles · doi:10.1109/tiv.2016.2551553 · Cited by 153
“The reference method [56] successfully mitigates the above issue through a different solver compatible with the L1 norm.”
MethodA survey of variational and CNN-based optical flow techniques
Tu, Z., Xie, W. C., Zhang, D., Poppe, R. W., Veltkamp, R. C., Li, B., & Yuan, J. (2018) · Signal Processing Image Communication · doi:10.1016/j.image.2018.12.002 · Cited by 155
“Additionally, the variational method can generally be implemented to run in real-time using modern numerical approaches [48], or advanced computation techniques (FPGA [78] or GPU [49,79]).”
BackgroundMethodInfluentialSF-Net: Single-Frame Supervision for Temporal Action Localization
Ma, F., Zhu, L., Yang, Y., Zha, S., Kundu, G., Feiszli, M., & Shou, Z. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58548-8_25 · Cited by 152
Motion-Aware Feature for Improved Video Anomaly DetectionPreprint
Zhu, Y., & Newsam, S. D. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1907.10211 · Cited by 150
“ah} 2018{Sultani, Chen, and Shah} 2018{Hasan, Choi, Neumann, Roy-Chowdhury, and Davis} 2016{Lu, Shi, and Jia} 2013{Sultani, Chen, and Shah} 2018YI ZHU, SHAWN NEWSAM: 9 Method Speed (fps) AUC (%) TVL1 [25] 102 71:7 FlowFields [1] 3 70:6 FlowNet2 [6] 185 73:2 PWCNet [16] 767 72:1 Motion Vector >1000 51:8 Video Salien…”
MethodUnsupervised Deep Video Hashing via Balanced Code for Large-Scale Video Retrieval
Wu, G., Han, J., Guo, Y., Liu, L., Ding, G., Ni, Q., & Shao, L. (2019) · IEEE Transactions on Image Processing · doi:10.1109/tip.2018.2882155 · Cited by 148
“Particularly in this case, we utilize the original RGB frames and extract their optical flow8 [66] images from videos in above benchmark datasets by OpenCV as the”
MethodA Unified Video Segmentation Benchmark: Annotation, Metrics and Analysis
Galasso, F., Nagaraja, N. S., Cardenas, T. J., Brox, T., & Schiele, B. (2013) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2013.438 · Cited by 149
“The result of [1] (across the hierarchy) at the central frame of the video sequences are propagated to other frames in the video with optical flow [30] and used to label corresponding image segments (across the hierarchy) by maximum voting.”
MethodAction Recognition Using Sparse Representation on Covariance Manifolds of Optical Flow
Guo, K., Ishwar, P., & Konrad, J. (2010) · IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS) · doi:10.1109/avss.2010.71 · Cited by 148
“Here we use a variant of the HornSchunck method, which optimizes a functional based on residuals from the intensity constraint and a smoothness regularization term [21].”
MethodMMNet: A Model-Based Multimodal Network for Human Action Recognition in RGB-D Videos
Yu, B. X. B., Liu, Y., Zhang, X., Zhong, S.-H., & Chan, K. C. C. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3177813 · Cited by 150
“flow stream (extracted by the TV-L1 algorithm [41]), were applied to vision-based HAR for the two-stream model in [5].”
MethodMotionSqueeze: Neural Motion Feature Learning for Video Understanding
Kwon, H., Kim, M., Kwak, S., & Cho, M. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58517-4_21 · Cited by 147
Rolling-Unrolling LSTMs for Action Anticipation from First-Person Video
Furnari, A., & Farinella, G. M. (2020) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2020.2992889 · Cited by 155
“Optical flow fields have been computed on all datasets using the TVL1 algorithm [71].”
MethodEnable Deep Learning on Mobile Devices: Methods, Systems, and Applications
Cai, H., Lin, J., Lin, Y., Liu, Z., Tang, H., Wang, H., Zhu, L., & Han, S. (2022) · ACM Transactions on Design Automation of Electronic Systems · doi:10.1145/3486618 · Cited by 162
“Using optical flow as an extra input modality can improve the video recognition accuracy [343], which is also called “two-stream” method.”
BackgroundMethodCombined Region and Motion-Based 3D Tracking of Rigid and Articulated Objects
Brox, T., Rosenhahn, B., Gall, J., & Cremers, D. (2009) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2009.32 · Cited by 143
Action Recognition From Video Using Feature Covariance Matrices
Guo, K., Ishwar, P., & Konrad, J. (2013) · IEEE Transactions on Image Processing · doi:10.1109/tip.2013.2252622 · Cited by 142
Weakly-Supervised Action Localization with Expectation-Maximization Multi-Instance Learning
Luo, Z., Guillory, D., Shi, B., Ke, W., Wan, F., Darrell, T., & Xu, H. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58526-6_43 · Cited by 140
Label propagation in video sequences
Badrinarayanan, V., Galasso, F., & Cipolla, R. (2010) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2010.5540054 · Cited by 141
“The first experiment compares the accuracy of propagated labels against known ground truth, using: pixel wise correspondences as delivered by an optic flow (OF) algorithm [13], rectangular image patch (IP) matches [5], and pre-optimised mappings of semantically consistent regions (SR) [1] (see Section 3).”
MethodA deep learning framework for quality assessment and restoration in video endoscopy
Ali, S., Zhou, F., Bailey, A. A., Braden, B., East, J. E., Lü, X., & Rittscher, J. (2020) · Medical Image Analysis · doi:10.1016/j.media.2020.101900 · Cited by 146
Horn-Schunck Optical Flow with a Multi-Scale Strategy
Meinhardt-Llopis, E., Pérez, J. S., & Kondermann, D. (2013) · Image Processing On Line · doi:10.5201/ipol.2013.20 · Cited by 138
“The use of L(1) functionals, for instance, like in [6, 13, 12], has shown to be a better approach.”
BackgroundNon-local Total Generalized Variation for Optical Flow Estimation
Ranftl, R., Bredies, K., & Pock, T. (2014) · Lecture notes in computer science · doi:10.1007/978-3-319-10590-1_29 · Cited by 138
“In contrast to the approach of linearizing the matching image [29], which leads to the classical optical flow constraint, this strategy allows to incorporate complex data terms into the model.”
MethodResultSelf-citationLarge displacement optical flow computation withoutwarping
Steinbrücker, F., Pock, T., & Cremers, D. (2009) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2009.5459364 · Cited by 137
“To circumvent this combinatorial explosion of permissible solutions, researchers have reverted to coarse-to-fine strategies of estimation [1, 8, 10, 11].”
BackgroundMethodSelf-citationExploring global diverse attention via pairwise temporal relation for video summarization
Li, P., Ye, Q., Zhang, L., Yuan, L., Xu, X., & Shao, L. (2020) · Pattern Recognition · doi:10.1016/j.patcog.2020.107677 · Cited by 143
Unified Spatio-Temporal Attention Networks for Action Recognition in Videos
Li, D., Yao, T., Duan, L., Mei, T., & Rui, Y. (2018) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2018.2862341 · Cited by 136
“Here we exploit TVL1 optical flow algorithm [47] to extract optical flow and discretize the optical flow fields in the range of [0, 255]”
MethodAccurate Detection and Recognition of Dirty Vehicle Plate Numbers for High-Speed Applications
Panahi, R., & Gholampour, I. (2016) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2016.2586520 · Cited by 139
“On the other hand, optical flow based algorithms such as variational methods [44] are great for calculating speed vectors”
MethodEfficient representation of traffic scenes by means of dynamic stixels
Pfeiffer, D., & Franke, U. (2010) · IEEE Intelligent Vehicles Symposium · doi:10.1109/ivs.2010.5548114 · Cited by 131
“[14] that is based upon the variational approach of Horn and Schunk [15].”
MethodRobust Monocular Epipolar Flow Estimation
Yamaguchi, K., McAllester, D. A., & Urtasun, R. (2013) · 2013 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2013.243 · Cited by 136
“3 px DB-TV-L1 [41] 33.”
BackgroundExploiting Temporal Consistency for Real-Time Video Depth Estimation
Zhang, H., Li, Y., Cao, Y., Liu, Y., Shen, C., & Yan, Y. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00181 · Cited by 143
“oflow denotes real time TV −L(1) optical flow [46].”
BackgroundProcedural Generation of Videos to Train Deep Action Recognition Networks
de Souza, C. R., Gaidon, A., Cabon, Y., & López, A. M. (2017) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2017.278 · Cited by 135
“Note that although we could use our generated ground truth flow for the PHAV samples in the motion stream, we use the same fast optical flow estimation algorithm as [75] (TVL1 [79]) for all samples in order to fairly estimate the usefulness of our generated videos.”
MethodTwo-Stream Collaborative Learning With Spatial-Temporal Attention for Video Classification
Peng, Y., Zhao, Y., & Zhang, J. (2018) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2018.2808685 · Cited by 131
Violent Interaction Detection in Video Based on Deep Learning
Zhou, P., Ding, Q., Luo, H., & Hou, X. (2017) · Journal of Physics Conference Series · doi:10.1088/1742-6596/844/1/012044 · Cited by 131
Multimodal Keyless Attention Fusion for Video Classification
Long, X., Gan, C., de Melo, G., Liu, X., Li, Y., Li, F., & Wen, S. (2018) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v32i1.12319 · Cited by 126
“To overcome the shortcomings of 2D CNNs and better account for spatio-temporal information, the optical flow method (Zach, Pock, and Bischof 2007) was proposed to consider the variation in the surrounding frames.”
MethodTotal variation regularization of local-global optical flow
Drulea, M., & Nedevschi, S. (2011) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc.2011.6082986 · Cited by 125
Massively parallel Lucas Kanade optical flow for real-time video processing applications
Plyer, A., Besnerais, G. L., & Champagnat, F. (2014) · Journal of Real-Time Image Processing · doi:10.1007/s11554-014-0423-0 · Cited by 121
“2.4, and we use the CUDA code available on OpenCV's website for PyramLK [9].”
BackgroundMethodInfluentialViolence detection in surveillance video using low-level features
Zhou, P., Ding, Q., Luo, H., & Hou, X. (2018) · PLoS ONE · doi:10.1371/journal.pone.0203668 · Cited by 121
A Comprehensive Study of Deep Video Action RecognitionPreprint
Zhu, Y., Li, X., Liu, C., Zolfaghari, M., Xiong, Y., Wu, C., Zhang, Z., Tighe, J., Manmatha, R., & Li, M. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2012.06567 · Cited by 212
“They derive such features directly from the definition of optical flow [255].”
BackgroundRobust Local Optical Flow for Feature Tracking
Senst, T., Eiselein, V., & Sikora, T. (2012) · IEEE transactions on circuits and systems for video technology (Print) · doi:10.1109/tcsvt.2012.2202070 · Cited by 129
“More robust norms are the L(1) norm [8], [9], [32] and Huber’s minmax norm [7], [34] because their influence functions are limited.”
BackgroundEye blink based fatigue detection for prevention of Computer Vision Syndrome
Divjak, M., & Bischof, H. (2009) · Machine Vision Applications · Cited by 117
Self-citationCreating Autonomous Vehicle Systems
Liu, S., Li, L., Tang, J., Wu, S., & Gaudiot, J. (2017) · Synthesis lectures on computer science · doi:10.2200/s00787ed1v01y201707csl009 · Cited by 114
The State of the Art in HDR Deghosting: A Survey and Evaluation
Tursun, O. T., Akyüz, A. O., Erdem, A., & Erdem, E. (2015) · Computer Graphics Forum · doi:10.1111/cgf.12593 · Cited by 115
Differences between stereo and motion behaviour on synthetic and real-world stereo sequences
Vaudrey, T., Rabe, C., Klette, R., & Milburn, J. (2008) · International Conference on Image and Vision Computing New Zealand (IVCNZ) · doi:10.1109/ivcnz.2008.4762133 · Cited by 116
“Examples of accurate methods are Semi-Global Matching (SGM) [12] for stereo, and the duality based optical ow technique from [ 25 ].”
MethodA Variational Approach to Video Registration with Subspace Constraints
Garg, R., Roussos, A., & Agapito, L. (2013) · International Journal of Computer Vision · doi:10.1007/s11263-012-0607-7 · Cited by 116
Optical Flow based CNN for detection of unlearnt deepfake manipulations
Caldelli, R., Galteri, L., Amerini, I., & Del Bimbo, A. (2021) · Pattern Recognition Letters · doi:10.1016/j.patrec.2021.03.005 · Cited by 115
JOLO-GCN: Mining Joint-Centered Light-Weight Information for Skeleton-Based Action Recognition
Cai, J., Jiang, N.-J., Han, X., Jia, K., & Lu, J.-B. (2020) · IEEE Workshop/Winter Conference on Applications of Computer Vision · doi:10.1109/wacv48630.2021.00278 · Cited by 110
“In this work, we use the classic TV-L1 algorithm [37] for the optical flow estimation due to its simplicity and effectiveness.”
MethodWeakly Supervised Audio-Visual Violence Detection
Wu, P., Liu, X., & Liu, J. (2022) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2022.3147369 · Cited by 125
“Optical flow is computed by the GPU version of TV-L1 [48] with great rapidity.”
MethodAdversarial Background-Aware Loss for Weakly-Supervised Temporal Activity Localization
Min, K., & Corso, J. J. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58568-6_17 · Cited by 113
Brigot, G., Colin‐Koeniguer, E., Plyer, A., & Janez, F. (2016) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2016.2578362 · Cited by 112
“Therefore, global methods can produce piecewise constant flow fields, as for example in the case of TV regularization [26].”
MethodRAFT: Recurrent All-Pairs Field Transforms for Optical Flow (Extended Abstract)
Teed, Z., & Deng, J. (2021) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2021/662 · Cited by 112
Temporal Action Segmentation: An Analysis of Modern Techniques
Ding, G., Şener, F., & Yao, A. (2023) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2023.3327284 · Cited by 150
“For feature extraction, the optical flow of the input video is computed by the TV-L1 algorithm [86].”
MethodOn a PCA-based lung motion model
Li, R., Lewis, J. H., Jia, X., Zhao, T., Liu, W., Wuenschel, S., Lamb, J., Yang, D., Low, D. A., & Jiang, S. B. (2011) · Physics in Medicine and Biology · doi:10.1088/0031-9155/56/18/015 · Cited by 110
“The regularization parameter λ and the relaxation parameter θ (Zach et al 2007) were set empirically at 70 and 0.25 respectively in this study.”
MethodResultInfluentialSeparable Flow: Learning Motion Cost Volumes for Optical Flow Estimation
Zhang, F., Woodford, O. J., Prisacariu, V. A., & Torr, P. H. S. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.01063 · Cited by 110
“The second usually optimizes a global energy function that consists of a local matching cost data term and an MRF-based smoothness regularization term, using gradient-based solvers [5,6, 19,45,47,57,66].”
MethodUncertainty Guided Collaborative Training for Weakly Supervised Temporal Action Detection
Yang, W., Zhang, T., Yu, X., Tian, Q. C., Zhang, Y., & FengWu, F. (2021) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr46437.2021.00012 · Cited by 110
“And TV-L1 optical flow [52] is used to generate optical frames for the FLOW stream.”
MethodA Survey on Driver Behavior Analysis From In-Vehicle Cameras
Wang, J., Chai, W., Venkatachalapathy, A., Tan, K. L., Haghighat, A. K., Velipasalar, S., Adu-Gyamfi, Y. O., & Sharma, A. (2021) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2021.3126231 · Cited by 115
Video Denoising via Empirical Bayesian Estimation of Space-Time Patches
Arias, P., & Morel, J. (2017) · Journal of Mathematical Imaging and Vision · doi:10.1007/s10851-017-0742-4 · Cited by 107
Actionness Estimation Using Hybrid Fully Convolutional Networks
Wang, L., Qiao, Y., Tang, X., & Gool, L. V. (2016) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2016.296 · Cited by 103
“We choose the TVL1 optical flow algorithm [48] and use OpenCV implementation, due to its balance between accuracy and efficiency.”
MethodDense, Robust, and Accurate Motion Field Estimation from Stereo Image Sequences in Real-Time
Rabe, C., Müller, T., Wedel, A., & Franke, U. (2010) · Lecture notes in computer science · doi:10.1007/978-3-642-15561-1_42 · Cited by 103
“in [12],[13].”
BackgroundSenst, T., Eiselein, V., Kühn, A., & Sikora, T. (2017) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2017.2725820 · Cited by 103
“The Lagrangian fields are estimated based on the Dual TV-L1 optical flow [38].”
MethodARBEE: Towards Automated Recognition of Bodily Expression of Emotion in the Wild
Luo, Y., Ye, J., Adams, R. B., Li, J., Newman, M. G., & Wang, J. Z. (2019) · International Journal of Computer Vision · doi:10.1007/s11263-019-01215-y · Cited by 107
“Optical flow was computed via TVL1 optical flow algorithm (Zach et al. 2007).”
End-to-end Video-level Representation Learning for Action Recognition
Zhu, J., Zhu, Z. D., & Zou, W. (2018) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2018.8545710 · Cited by 102
Global Matching with Overlapping Attention for Optical Flow Estimation
Zhao, S., Zhao, L., Zhang, Z., Zhou, E., & Metaxas, D. (2022) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52688.2022.01707 · Cited by 101
“Later research made further improvements by using better regularization terms [6, 34, 53] or additional robust optimization terms [5, 8].”
MethodA Duality Based Algorithm for TV-L 1-Optical-Flow Image Registration
Pock, T., Urschler, M., Zach, C., Beichel, R., & Bischof, H. (2007) · Springer eBooks · doi:10.1007/978-3-540-75759-7_62 · Cited by 101
“A proof of this proposition is presented in [17].”
BackgroundSelf-citationVideo Segmentation with Superpixels
Galasso, F., Cipolla, R., & Schiele, B. (2013) · Lecture notes in computer science · doi:10.1007/978-3-642-37331-2_57 · Cited by 100
3D scene flow estimation with a rigid motion prior
Vogel, C., Schindler, K., & Roth, S. (2011) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2011.6126381 · Cited by 100
“To enable efficient energy minimization we employ a relaxation similar to the one introduced in [24] for optical flow computation.”
MethodPatch-Based Video Denoising With Optical Flow Estimation
Buades, A., Lisani, J.-L., & Miladinović, M. (2016) · IEEE Transactions on Image Processing · doi:10.1109/tip.2016.2551639 · Cited by 99
“Collaborative filtering is realized by transforming each group through a decorrelating 4D separable transform and then by shrinkage and inverse transformation.”
BackgroundMethodZeng, R., Gan, C., Chen, P., Huang, W., Wu, Q., & Tan, M. (2019) · IEEE Transactions on Image Processing · doi:10.1109/tip.2019.2922108 · Cited by 96
“The input for the RGB and Flow stream is 1-frame RGB image and 5-frame stacks of TV-L1 [43] optical flows, respectively.”
MethodEPIC-KITCHENS-100Preprint
Damen, D., Doughty, H., Farinella, G. M., Furnari, A., Kazakos, E., Ma, J., Moltisanti, D., Munro, J., Perrett, T., Price, W., et al. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2006.13256 · Cited by 153
Proximity algorithms for the L1/TV image denoising model
Micchelli, C. A., Shen, L., Xu, Y., & Zeng, X. (2011) · Advances in Computational Mathematics · doi:10.1007/s10444-011-9243-y · Cited by 96
“Applications of the L1/TV model include computer vision [15], biomedical imaging [37], optical flow and object tracking [40], and shape denoising [40].”
MethodDense, accurate optical flow estimation with piecewise parametric model
Yang, J., & Li, H. (2015) · Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2015.7298704 · Cited by 128
“Remarkable progresses have been made in the past decades, with high performance optical flow algorithms available nowadays [11, 48, 37, 46, 25].”
BackgroundEvaluating fusion of RGB-D and inertial sensors for multimodal human action recognition
Imran, J., & Raman, B. (2019) · Journal of Ambient Intelligence and Humanized Computing · doi:10.1007/s12652-019-01239-9 · Cited by 97
“The horizontal and vertical optical flow components computed in Step 1 can be obtained using any one of the commonly used methods like Brox (Brox et al. 2004), TV-L1 (Zach et al. 2007) or Farneback (Farnebäck 2003) algorithm.”
MethodLearning a Confidence Measure for Optical Flow
Aodha, O. M., Humayun, A., Pollefeys, M., & Brostow, G. (2012) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2012.171 · Cited by 93
Broaden Your Views for Self-Supervised Video Learning
Recasens, A., Luc, P., Alayrac, J.-B., Wang, L., Strub, F., Tallec, C., Malinowski, M., Pătrăucean, V., Altché, F., Valko, M., et al. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00129 · Cited by 98
“For flow extraction, we use the TVL1 [88] algorithm.”
Recognize Actions by Disentangling Components of Dynamics
Zhao, Y., Xiong, Y., & Lin, D. (2018) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2018.00687 · Cited by 90
“The most widely used optical flow method for action recognition is TV-L1 [34].”
MethodInfluentialUnderstanding More About Human and Machine Attention in Deep Neural Networks
Lai, Q., Khan, S. H., Nie, Y., Sun, H., Shen, J., & Shao, L. (2020) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2020.3007321 · Cited by 102
“To get training data for the two-stream architecture, we extract the frames and compensated optical flows images [86], and build the file list of video snippets.”
MethodReAct: Temporal Action Detection with Relational Queries
Shi, D., Zhong, Y., Cao, Q., Zhang, J., Ma, L., Li, J., & Tao, D. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-20080-9_7 · Cited by 94
Graph Distillation for Action Detection with Privileged Modalities
Luo, Z., Hsieh, J.-T., Jiang, L., Niebles, J. C., & Fei-Fei, L. (2018) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-01264-9_11 · Cited by 111
“The optical flow is calculated on the RGB videos using the dual TV-L1 method [62].”
MethodModel-Blind Video Denoising via Frame-To-Frame Training
Ehret, T., Davy, A., Morel, J., Facciolo, G., & Arias, P. (2019) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2019.01163 · Cited by 87
“We use the TV-L1 optical flow [46] with an implementation available in [37].”
MethodLearning Optical Flow from a Few Matches
Jiang, S., Lu, Y., Li, H., & Hartley, R. I. (2021) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr46437.2021.01632 · Cited by 89
“Various subsequent papers have worked on on improving robustness [2] and energy term [35], incorporating descriptor matching into energy minimization [4] and improving regularizations [26].”
Multi-modal and Multi-spectral Registration for Natural Images
Shen, X.-Y., Xu, L., Zhang, Q., & Jia, J.-Y. (2014) · European Conference on Computer Vision · doi:10.1007/978-3-319-10593-2_21 · Cited by 100
“Robust functions, such as L1 norm and Charbonnier function, were used by [5,3,31,39] in regularization.”
MethodPatchBatch: A Batch Augmented Loss for Optical Flow
Gadot, D., & Wolf, L. (2016) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2016.459 · Cited by 86
“28.50% 116s DeepFlow [38] 29.18% 17s HS [35] 42.18% 2.6m DB-TV-L1 [40] 47.97% 16s HAOF [6] 50.29% 16.2s PolyExpand [14] 53.32% 1s for 32D. Based on these results, further improvements of our method’s accuracy are expected with larger patch and representation sizes.”
MethodInfluentialFine-Grained Activity Recognition in Baseball Videos
Piergiovanni, A., & Ryoo, M. S. (2018) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2018.00226 · Cited by 85
High-Resolution Optical Flow from 1D Attention and Correlation
Xu, H., Yang, J., Cai, J., Zhang, J., & Tong, X. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.01033 · Cited by 86
“By further improving the architectures and training strategies, deep learning based methods [18, 34, 35, 43, 37] have demonstrated stronger performance and faster inference speed compared with traditional optimization based approaches [13, 46, 33, 29, 44, 3].”
MethodExtreme Low Resolution Activity Recognition With Multi-Siamese Embedding Learning
Ryoo, M. S., Kim, K., & Yang, H. J. (2018) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v32i1.12299 · Cited by 85
Iterative Semi-Global Matching for Robust Driver Assistance Systems
Hermann, S., & Klette, R. (2013) · Lecture notes in computer science · doi:10.1007/978-3-642-37431-9_36 · Cited by 84
Multimodal Multi-Stream Deep Learning for Egocentric Activity Recognition
Song, S., Chandrasekhar, V., Mandal, B., Li, L., Lim, J., Babu, G. S. D., San, P. P., & Cheung, N. M. (2016) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2016.54 · Cited by 84
“TVL1 optical flow algorithm [26] is chosen with OpenCV implementation, since it makes a good balance between efficiency and accuracy.”
TSM: Temporal Shift Module for Efficient and Scalable Video Understanding on Edge Devices
Lin, J., Gan, C., Wang, K., & Han, S. (2020) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2020.3029799 · Cited by 88
“[46] designed a two-stream CNN for RGB input (spatial stream) and optical flow [65] input (temporal stream) respectively.”
Activity Recognition for Ambient Assisted Living with Videos, Inertial Units and Ambient Sensors
Ranieri, C. M., MacLeod, S., Dragone, M., Vargas, P. A., & Romero, R. A. F. (2021) · Sensors · doi:10.3390/s21030768 · Cited by 84
“Those maps were generated with OpenCV implementation of the TVL1 algorithm [31] on each pair of successive frames on the RGB videos previously converted to grayscale images.”
MethodGoing deeper with two-stream ConvNets for action recognition in video surveillance
Han, Y., Zhang, P., Zhuo, T., Huang, W., & Zhang, Y. (2017) · Pattern Recognition Letters · doi:10.1016/j.patrec.2017.08.015 · Cited by 81
Trajectory Convolution for Action Recognition
Zhao, Y., Xiong, Y., & Lin, D. (2018) · neural information processing systems · Cited by 82
Metric Learning for Image Registration
Niethammer, M., Kwitt, R., & Vialard, F. (2019) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2019.00866 · Cited by 81
“Initial models based on deep learning [13, 24] were proposed to speed-up optical flow computations [22, 3, 8, 7, 49, 40].”
MethodDeep learning for vision‐based fall detection system: Enhanced optical dynamic flow
Chhetri, S., Alsadoon, A., Al‐Dala'in, T., Prasad, P. W. C., Rashid, T. A., & Maag, A. (2020) · Computational Intelligence · doi:10.1111/coin.12428 · Cited by 81
“In the final step of transfer learning, the weight of the convolution layer was frozen to make the weight unaltered during the training stage [18].”
BackgroundInfluentialRajalakshmi, E., Elakkiya, R., Subramaniyaswamy, V., Alexey, L. P., Grif, M. G., Bakaev, M., Kotecha, K. V., Gabralla, L. A., & Abraham, A. (2023) · IEEE Access · doi:10.1109/access.2022.3233671 · Cited by 81
Metadata
- Authors
- Zach, C., Pock, T., & Bischof, H.
- Affiliations
- VRVis GmbH (Austria) (AT)
- Graz University of Technology (AT)
- Venue
- Lecture Notes in Computer Science
- Publisher
- Springer Berlin Heidelberg
- Published
- 2007-09-21
- Pages
- 214-223
- OpenAlex
- W1578985305
- Cited by
- 1,658
- References
- 27
- Citations by year
- 2026: 44 · 2025: 62 · 2024: 83 · 2023: 111 · 2022: 107 · 2021: 188 · 2020: 152 · 2019: 149 · 2018: 112 · 2017: 98 · 2016: 88 · 2015: 99 · 2014: 70 · 2013: 59 · 2012: 56
- FWCI
- 27.93
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- Springer eBooks (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Vision and Imaging
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Video Surveillance and Tracking Methods
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Image and Signal Denoising Methods
- Keywords
- optical flow, total variation regularization, GPU acceleration, real-time video processing, variational methods, discontinuity preservation
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
