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 401–500 of 1,002
Reinoso-Rondinel, R., Rempel, M., Schultze, M., & Trömel, S. (2022) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2022.3144342 · Cited by 33
Motion-Focused Contrastive Learning of Video Representations
Li, R., Zhang, Y., Qiu, Z., Yao, T., Liu, D., & Mei, T. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00211 · Cited by 33
“Given a video of N frames with the resolution of H × W , we first extract the optical flow from each pair of consecutive frames as [53] algo-rithm.”
Event detection in surveillance videos: a review
Karbalaie, A., Abtahi, F., & Sjöström, M. (2022) · Multimedia Tools and Applications · doi:10.1007/s11042-021-11864-2 · Cited by 35
“The TVL1 [96] optical flow algorithm is used to extract optical flow features.”
MethodSemi-supervised intelligent surveillance system for secure environments
Fookes, C., Denman, S., Lakemond, R., Ryan, D., Sridharan, S., & Piccardi, M. (2010) · IEEE International Symposium on Industrial Electronics (ISIE) · doi:10.1109/isie.2010.5636922 · Cited by 33
“motion segmentation [12] or optical flow [13], [ 14 ]) in the scene.”
BackgroundNacci, A. A., Rana, V., Bruschi, F., Sciuto, D., Beretta, I., & Atienza, D. (2013) · ACM/IEEE Design Automation Conference (DAC) · doi:10.1145/2463209.2488797 · Cited by 32
“For example, the architectures in [3], [22] and [23]), are unable to reach the real-time threshold (i.”
BackgroundOptimization Strategies for High-Performance Computing of Optical-Flow in General-Purpose Processors
Anguita, M., Díaz, J., Ros, E., & Fernández, F. J. R. (2009) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2009.2026821 · Cited by 32
“[25]) based on a GTX 280 graphics card 2 (release date: Q2’2008, TDP: 183 W).”
BackgroundMethodOnline Video Stream Abstraction and Stylization
Zhang, S.-H., Li, X.-Y., Hu, S.-M., & Martin, R. R. (2011) · IEEE transactions on multimedia · doi:10.1109/tmm.2011.2165052 · Cited by 34
“Typically we can process a CIF video stream with default parameter settings at about 9‐12 frames per second using the CPU for everything but optical flow (we directly use Zach’s GPU optical flow code [ 36 ]).”
MethodSign Language Recognition via Skeleton-Aware Multi-Model EnsemblePreprint
Jiang, S., Sun, B., Wang, L., Bai, Y., Li, K., & Fu, Y. (2021) · arXiv.org · doi:10.48550/arxiv.2110.06161 · Cited by 60
“Besides, we use the Denseflow API implemented with OpenCV and CUDA that provided by OpenMMLab [70] to obtain the optical flows using the TVL1 algorithm [71].”
MethodLee, J. T., Park, E., & Jung, T.-D. (2019) · Sensors · doi:10.3390/s19183873 · Cited by 30
“First, we applied a TV-L1 optical flow algorithm [31,32] on gray-scale images.”
MethodStepNet: Spatial-temporal Part-aware Network for Isolated Sign Language Recognition
Shen, X.-L., Zheng, Z., & Yang, Y. (2022) · ACM Trans. Multim. Comput. Commun. Appl. · doi:10.1145/3656046 · Cited by 33
“” Note that all these datasets employ the signer-independent setting, which means no signer overlap between the training and testing splits.”
BackgroundConvex Relaxation for Multilabel Problems with Product Label Spaces
Goldlüecke, B., & Cremers, D. (2010) · Lecture notes in computer science · doi:10.1007/978-3-642-15555-0_17 · Cited by 31
Monitoring systems for the support of home care
Hein, A., Winkelbach, S., Martens, B., Wilken, O., Eichelberg, M., Spehr, J., Gietzelt, M., Wolf, K.-H., Büsching, F., Hülsken-Giesler, M., et al. (2010) · Informatics for Health and Social Care · doi:10.3109/17538157.2010.528637 · Cited by 31
“Fortunately, there are real-time implementations [52] accelerated by modern graphics processing units.”
BackgroundA Continuous Occlusion Model for Road Scene Understanding
Dhiman, V., Tran, Q.-H., Corso, J. J., & Chandraker, M. K. (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.2016.469 · Cited by 31
“We then apply [32] to extract point tracks.”
MethodSchaub, A., Baumgartner, D., & Burschka, D. (2016) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2016.2633292 · Cited by 32
“optical flow, where the flow vectors are evenly distributed over the entire image [28], ii) sparse optical flow, where”
BackgroundFully convolutional networks for action recognition
Yu, S., Cheng, Y., Xie, L., & Li, S. (2017) · IET Computer Vision · doi:10.1049/iet-cvi.2017.0005 · Cited by 32
Multiscale Deep Alternative Neural Network for Large-Scale Video Classification
Wang, J., Wang, W., & Gao, W. (2018) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2018.2855081 · Cited by 31
TapLab: A Fast Framework for Semantic Video Segmentation Tapping Into Compressed-Domain Knowledge
Feng, J., Li, S., Li, X., Wu, F. Q., Tian, Q., Yang, M., & Ling, H. (2020) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2020.3024646 · Cited by 33
Optical Flow Estimation by Matching Time Surface with Event-Based Cameras
Nagata, J., Sekikawa, Y., & Aoki, Y. (2021) · Sensors · doi:10.3390/s21041150 · Cited by 34
“The Graphics Processing Unit (GPU) was used only as a method of Reconstruction.”
MethodInfluentialSLIC: Self-Supervised Learning with Iterative Clustering for Human Action Videos
Khorasgani, S. H., Chen, Y., & Shkurti, F. (2022) · Computer Vision and Pattern Recognition · doi:10.1109/cvpr52688.2022.01562 · Cited by 34
“For multi-view positive sampling, optical flow (u) is computed using the unsupervised TV-L1 algorithm [64], with the same pre-processing procedure as in [11].”
MethodEvaluation of Scan-Line Optimization for 3D Medical Image Registration
Hermann, S. (2014) · 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.2014.393 · Cited by 31
“[30] (cTVL1) and Brox et al.”
Geodesic Active Fields—A Geometric Framework for Image Registration
Zosso, D. P., Bresson, X., & Thiran, J. (2010) · IEEE Transactions on Image Processing · doi:10.1109/tip.2010.2093904 · Cited by 31
“For nonsmooth deformation fields, e.g., observed in optical flow-based image registration, the -norm may perform better as data term [ 44 ].”
MethodPAN: Towards Fast Action Recognition via Learning Persistence of AppearancePreprint
Zhang, C., Zou, Y., Chen, G., & Gan, L. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2008.03462 · Cited by 37
“We speculate that it is because the regularization term, penalizing high variations to obtain smooth displacement fields, is not applied for efficiency concerns, while it is used in conventional optical flow method [26].”
BackgroundMethodInfluentialDomain-Specific Priors and Meta Learning for Few-Shot First-Person Action Recognition
Coskun, H., Zia, M. Z., Tekin, B., Bogo, F., Navab, N., Tombari, F., & Sawhney, H. S. (2021) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2021.3058606 · Cited by 36
“We pre-compute optical flow by using the TV-L1 algorithm proposed in [68], with the same parameters used in [9].”
MethodInfluentialHistograms of Motion Gradients for real-time video classification
Duţă, I. C., Uijlings, J., Nguyen, T. A., Aizawa, K., Hauptmann, A. G., Ionescu, B. E., & Sebe, N. (2016) · International Workshop on Content-Based Multimedia Indexing (CBMI) · doi:10.1109/cbmi.2016.7500260 · Cited by 30
“from relatively old methods, such as [15, 8] to relatively recent approaches like [7, 3, 31, 4], which use complex algorithms to compute the motion information.”
MethodIllumination-Invariance Optical Flow Estimation Using Weighted Regularization Transform
Mei, L., Lai, J., Xie, X., Zhu, J., & Chen, J. (2019) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2019.2890861 · Cited by 30
“Therefore, we exploit a duality-based approach [27] and reformulate equation (16) as”
A Spatiotemporal Heterogeneous Two-Stream Network for Action Recognition
Chen, E., Bai, X., Gao, L., Tinega, H. C., & Ding, Y. (2019) · IEEE Access · doi:10.1109/access.2019.2910604 · Cited by 30
“The method of [40] is used to calculate the optical flows, and all experiments are implemented based on the Caffe platform [41].”
MethodDeep Insights into Convolutional Networks for Video Recognition
Feichtenhofer, C., Pinz, A. J., Wildes, R. P., & Zisserman, A. (2019) · International Journal of Computer Vision · doi:10.1007/s11263-019-01225-w · Cited by 30
“The models are taken from the Temporal Segment Networks (TSN) approach [53], which pretrains the motion stream on TVL1 optical flow [56], IDT-flow [52] (termed warped flow in [53]) and difference images of UCF101.”
MethodPurely Attention Based Local Feature Integration for Video Classification
Long, X., de Melo, G., He, D., Li, F., Chi, Z., Wen, S., & Gan, C. (2020) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2020.3029554 · Cited by 30
Short-Term Forecasting of Large-Scale Clouds Impact on Downwelling Surface Solar Irradiation
Kosmopoulos, P., Kouroutsidis, D., Papachristopoulou, K., Raptis, P. I., Masoom, A., Saint‐Drenan, Y., Blanc, P., Kontoes, C., & Kazadzis, S. (2020) · Energies · doi:10.3390/en13246555 · Cited by 30
“[39], based on the minimization of a function containing a data term using the L1 norm and a regularization term using the total variation of the flow.”
MethodDeep Neural Networks Using Capsule Networks and Skeleton-Based Attentions for Action Recognition
Ha, M.-H., & Chen, O. T.-C. (2021) · IEEE Access · doi:10.1109/access.2020.3048741 · Cited by 30
“On the other hand, the Optical-Flow (OF) channel frames from the RGB frames are correspondingly computed by the popular method of the Total Variation regularization and the robust L 1 norm (TVL1) [33].”
MethodRobust Trajectory-Space TV-L1 Optical Flow for Non-rigid Sequences
Garg, R., Roussos, A., & Agapito, L. (2011) · Energy Minimization Methods in Computer Vision and Pattern Recognition · doi:10.1007/978-3-642-23094-3_22 · Cited by 36
“Formulating the manifold constraint as a soft constraint using variational principles leads to an energy with a quadratic relaxation term that allows us to adopt a decoupling scheme, similar to the one described above [13, 9], for its efficient optimization.”
BackgroundMethodInfluentialContinuous Energy Minimization Via Repeated Binary Fusion
Trobin, W., Pock, T., Cremers, D., & Bischof, H. (2008) · Lecture notes in computer science · doi:10.1007/978-3-540-88693-8_50 · Cited by 30
Self-citationAdversarial Attacks for Optical Flow-Based Action Recognition ClassifiersPreprint
Inkawhich, N., Inkawhich, M. J., Chen, Y., & Li, H. H. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1811.11875 · Cited by 33
“Before any training, we generate TV-L1 (Zach et al., 2007) and Farneback (Farnebäck, 2003) optical flow datasets using the OpenCV implementations with default parameters.”
MethodCorrelation Net: Spatiotemporal multimodal deep learning for action recognition
Yudistira, N., & Kurita, T. (2019) · Signal Processing Image Communication · doi:10.1016/j.image.2019.115731 · Cited by 29
“The flow modality is an optical flow that has been extracted using the TVL1 algorithm [15].”
Few-Shot Learning of Video Action Recognition Only Based on Video Contents
Yang, B., Lu, Y., & He, W. (2020) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv45572.2020.9093481 · Cited by 29
“The optical flow images [44] are pre-computed and stored as JPEG images (with displacement vectors > 20 pixels clipped).”
MethodJoint Learning of Blind Video Denoising and Optical Flow Estimation
Yu, S., Park, B., Park, J., & Jeong, J. (2020) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw50498.2020.00258 · Cited by 29
“F2F training employs the TV-L1 method [22] to obtain optical flows in a noisy sequence owing to its high speed and robustness to noise [6].”
MethodLearning SpatioTemporal and Motion Features in a Unified 2D Network for Action Recognition
Wang, M., Xing, J., Su, J., Chen, J., & Yong, L. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3173658 · Cited by 29
“However, calculating optical flow [40] is expensive in both time and space.”
BackgroundReal-time anomaly detection on surveillance video with two-stream spatio-temporal generative model
Liu, W., Cao, J., Zhu, Y., Liu, B., & Zhu, X. (2022) · Multimedia Systems · doi:10.1007/s00530-022-00979-7 · Cited by 29
“Compared with the traditional TV-L1 algorithm [31], it has the significant advantages of similar optical flow extraction quality but faster calculation.”
MethodHigh Accuracy Optical Flow for 3D Medical Image Registration Using the Census Cost Function
Hermann, S., & Werner, R. (2014) · Lecture notes in computer science · doi:10.1007/978-3-642-53842-1_3 · Cited by 29
Ehrhart, M., & Lienhart, W. (2015) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2084221 · Cited by 29
“For our work, we evaluated Farnebäck’s algorithm([38]) and the Dual TV L1 approach([39]).”
MethodA Novel Hardware Architecture of the Lucas–Kanade Optical Flow for Reduced Frame Memory Access
Seong, H.-S., Rhee, C. E., & Lee, H.-J. (2015) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2015.2437077 · Cited by 29
“unit acceleration takes longer than 1 s to calculate the optical flow for a 640 × 480 video [11], [12].”
BackgroundHosseini, A., & Lienkamp, M. (2016) · IEEE Intelligent Vehicles Symposium · doi:10.1109/ivs.2016.7535381 · Cited by 29
“Fusion of the stereo disparity and optical flow maps using EKF [21] [22].”
MethodDeep Video Color PropagationPreprint
Schaub-Meyer, S., Cornillère, V., Djelouah, A., Schroers, C., & Groß, M. H. (2018) · arXiv (Cornell University) · doi:10.3929/ethz-b-000319608 · Cited by 29
A multiscale image segmentation method
Li, Y., & Feng, X. (2015) · Pattern Recognition · doi:10.1016/j.patcog.2015.10.004 · Cited by 28
“The TV-L1 model has been used in many fields including computer vision [57], medical imaging [58] and optical flow [59], etc.”
MethodManhattan and Piecewise-Planar Constraints for Dense Monocular Mapping
Concha, A., Hussain, W., Montano, L., & Civera, J. (2014) · Robotics: Science and Systems · doi:10.15607/rss.2014.x.016 · Cited by 28
“Our approach addresses the more challenging case of RGB monocular sequences, being [20, 15, 17] the first works that used the total variational framework of [24] to achieve dense and real-time multiview 3D reconstructions.”
MethodOcclusion handling using semantic segmentation and visibility-based rendering for mixed reality
Roxas, M., Hori, T., Fukiage, T., Okamoto, Y., & Oishi, T. (2018) · ACM Symposium on Virtual Reality Software and Technology (VRST) · doi:10.1145/3281505.3281546 · Cited by 28
Learning Coupled Convolutional Networks Fusion for Video Saliency Prediction
Wu, Z., Su, L., & Huang, Q. (2018) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2018.2870954 · Cited by 29
“The optical flow is extracted by the TVL1 optical flow algorithm [40] due to its ability to balance accuracy and efficiency.”
MethodTavakolian, M., López, M. B., & Liu, L. (2020) · Pattern Recognition Letters · doi:10.1016/j.patrec.2020.09.012 · Cited by 31
“For the computation of the optical flow, we use TLV1 optical flow algorithm (Zach et al. (2007)), which is implemented in OpenCV with CUDA.”
MethodMeta-MMFNet: Meta-learning-based Multi-model Fusion Network for Micro-expression Recognition
Gong, W., Zhang, Y., Wang, W., Cheng, P., & Gonzàlez, J. (2022) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3539576 · Cited by 28
“In order to capture facial motions, optical low feature [44] [57] was also introduced.”
BackgroundLook&listen: Multi-Modal Correlation Learning for Active Speaker Detection and Speech Enhancement
Xiong, J., Yu, Z., Zhang, P., Xie, L., Huang, W., & Zha, Y. (2022) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2022.3199109 · Cited by 28
AU-aware graph convolutional network for Macroand Micro-expression spotting
Yin, S., Wu, S., Xu, T., Liu, S., Zhao, S., & Chen, E. (2023) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme55011.2023.00047 · Cited by 28
“After that, we adopt TV-L1 algorithm [27] to compute coarse optical-flow features, denoted as O = {on}lv−1 n=1 ∈ RH×W×(lv−1)×2, where on is the n-th opticalflow map calculated from xn and xn+1.”
MethodOPTICAL FLOW FOR GLACIER MOTION ESTIMATION
Vogel, C., Bauder, A., & Schindler, K. (2012) · ISPRS annals of the photogrammetry, remote sensing and spatial information sciences · doi:10.5194/isprsannals-i-3-359-2012 · Cited by 28
“In (Zach et al., 2007) it was proposed to add additional auxiliary variables v to the energy function, to simplify the optimization:”
BackgroundCrowd Escape Behavior Detection and Localization Based on Divergent Centers
Chen, C., & Shao, Y. (2015) · IEEE Sensors Journal · doi:10.1109/jsen.2014.2381260 · Cited by 28
“In this paper, the Horn-Schunck optical flow method which assumes the change of the optical flow is smooth in the whole image [31] is employed.”
MethodAdvanced Data Terms for Variational Optic Flow Estimation
Steinbr, F., Pock, T., & Cremers, D. (2009) · Vision Modeling and Visualization · Cited by 28
Self-citationLocal abnormal behavior detection based on optical flow and spatio-temporal gradient
Zhu, S., Hu, J., & Shi, Z. (2015) · Multimedia Tools and Applications · doi:10.1007/s11042-015-3122-3 · Cited by 28
“Firstly, bag of words [21] is adopted to describe the optical flow information of each blob, and the semi-parametric basedmodel [4] is adopted to detect suspicious local abnormal blobs; then, suspicious abnormal blobs are divided into rectangular cells equally, and themaximum optical flow energy method [27] is adopted…”
MethodMulti-modal learning for affective content analysis in movies
Yi, Y., & Wang, H. (2018) · Multimedia Tools and Applications · doi:10.1007/s11042-018-5662-9 · Cited by 28
“Meanwhile, the temporal ConvNet, which stacks optical flow [70] fields, captures the motion information in videos.”
Recognition of driver distractions using deep learning
Valeriano, L. C., Napoletano, P., & Schettini, R. (2018) · IEEE International Conference on Consumer Electronics - Berlin (ICCE-Berlin) · doi:10.1109/icce-berlin.2018.8576183 · Cited by 29
“flow using the TV-L1 algorithm [34], just like in the original work.”
MethodLearning Optical Flow and Scene Flow With Bidirectional Camera-LiDAR Fusion
Liu, H., Lu, T., Xu, Y., Liu, J., & Wang, L. (2023) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2023.3330866 · Cited by 28
“Traditional Methods: Traditional methods [7], [9], [10], [11], [20], [63], [74] formulate optical flow estimation as an energy minimization problem.”
MethodMulti-hypothesis transform domain Wyner-Ziv video coding including optical flow
Huang, X., Rakêt, L. L., Luong, H. V., Nielsen, M., Lauze, F., & Forchhammer, S. O. (2011) · IEEE International Workshop on Multimedia Signal Processing (MMSP) · doi:10.1109/mmsp.2011.6093771 · Cited by 27
“The minimizing flow is calculated using a highly efficient method, relying on tools from non-smooth convex analysis in a coarse-to-fine pyramid setup, which was introduced in [15], and further improved in [16].”
MethodExploring hybrid spatio-temporal convolutional networks for human action recognition
Wang, H., Yang, Y., Yang, E., & Deng, C. D. (2017) · Multimedia Tools and Applications · doi:10.1007/s11042-017-4514-3 · Cited by 26
Learning multi-temporal-scale deep information for action recognition
Yao, G., Leí, T., Zhong, J., & Jiang, P. (2018) · Applied Intelligence · doi:10.1007/s10489-018-1347-3 · Cited by 28
“Optical flow: We use the TVL1 algorithm [36] implemented in OpenCV with CUDA, and we create 3-channel optical flow described in Section 3.”
MethodSun, W., Su, R., Yu, Q., & Xu, D. (2022) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2022.3201540 · Cited by 27
“The optical flow maps are generated by using the TVL1 algorithm [64].”
MethodPose-Appearance Relational Modeling for Video Action Recognition
Cui, M., Wang, W., Zhang, K., Sun, Z., & Wang, L. (2022) · IEEE Transactions on Image Processing · doi:10.1109/tip.2022.3228156 · Cited by 29
“Two-stream methods [4], [6], [53] usually adopt TV-L1 algorithm [59] to extract optic flow.”
MethodNicolas, A., Zentgraf, F., Linne, M., Dreizler, A. M., & Peterson, B. (2023) · Experiments in Fluids · doi:10.1007/s00348-023-03594-y · Cited by 27
Explicit Motion Disentangling for Efficient Optical Flow Estimation
Deng, C., Luo, A., Huang, H., Ma, S., Liu, J., & Liu, S. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00873 · Cited by 27
“Although some better methods are proposed to enhance feature similarity and motion smoothness [4, 37, 27], limited by the handcrafted features, traditional methods are still hard to handle the cases in complex environments.”
MethodFast Optical Flow Estimation Based on the Split Bregman Method
Chen, J., Cai, Z., Lai, J., & Xie, X.-H. (2018) · IEEE transactions on circuits and systems for video technology (Print) · doi:10.1109/tcsvt.2016.2615324 · Cited by 24
“[20] proposed another efficient algorithm based on an isotropic TV-L1 regularization.”
MethodInfluentialAffective Video Content Analysis With Adaptive Fusion Recurrent Network
Yi, Y., Wang, H., & Li, Q. (2019) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2019.2955300 · Cited by 28
“, audio, RGB frame and optical flow [38].”
MethodProgressive modality cooperation for multi-modality domain adaptation
Zhang, W., Xu, D., Zhang, J., & Ouyang, W. (2021) · IEEE Transactions on Image Processing · doi:10.1109/tip.2021.3052083 · Cited by 25
“To capture more temporal information in videos, we further exploit the optical flow maps between two consecutive RGB/depth frames to form the additional modalities ( i.e. , Flow/DepthFlow), which are both generated by using the TVL1 optical flow extraction algorithm [77].”
A novel multi-stream method for violent interaction detection using deep learning
Hong-chang, L., Wang, J., Han, J., Zhang, J., Yang, Y., & Zhao, Y. (2020) · Measurement and Control · doi:10.1177/0020294020902788 · Cited by 26
“The difference in feature representation is large, so it can be fused to each other to obtain better recognition performance similar to the flow module in TSN. 14 For a given video S , we first employ TVL-1 algo-rithm 44 in OpenCV 45 to extract optical flow of the video.”
MethodLiu, N., Liu, X., Zhang, Z., Xu, X., & Chen, T. (2020) · International Conference on Intelligent Informatics and Biomedical Sciences (ICIIBMS) · doi:10.1109/iciibms50712.2020.9336412 · Cited by 27
“In the experiment, we used the TV-L1 method to calculate the optical flow of the micro-expression onset and apex frame as well as apex and offset frame[20].”
MethodEnhanced bare-bones particle swarm optimization based evolving deep neural networks
Li, Z., Lim, C. P., & Liu, C. (2023) · Expert Systems with Applications · doi:10.1016/j.eswa.2023.120642 · Cited by 26
Achromatic Imaging Systems with Flat Lenses Enabled by Deep Learning
Maman, R., Mualem, E., Mazurski, N., Engelberg, J., & Levy, U. (2023) · ACS Photonics · doi:10.1021/acsphotonics.3c01349 · Cited by 26
Real versus realistically rendered scenes for optical flow evaluation
Meister, S., & Kondermann, D. (2011) · Cited by 26
Motion estimation of crowd flow using optical flow techniques: A review
Kajo, I., Malik, A., & Kamel, N. (2015) · International Conference on Signal Processing and Communication Systems · doi:10.1109/icspcs.2015.7391778 · Cited by 25
“[32] introduced a new method to resolve the TVL1 formulation that may maintain discontinuities in the field of motion vectors in addition to enhance robustness against lighting changes, occlusions as well as noise.”
MethodA vision based traffic accident detection method using extreme learning machine
Chen, Y., Yu, Y., & Li, T. (2016) · International Conference on Advanced Robotics and Mechatronics (ICARM) · doi:10.1109/icarm.2016.7606983 · Cited by 26
“Then, we compute dense optical flow using the duality based total variation with an L1 norm (TV-L1) algorithm [22],”
MethodA novel recurrent hybrid network for feature fusion in action recognition
Yu, S., Cheng, Y., Xie, L., Luo, Z., Huang, M., & Li, S. (2017) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2017.09.007 · Cited by 26
Real-time obstacle detection using dense stereo vision and dense optical flow
Pantilie, C. D., Bota, S., Haller, I., & Nedevschi, S. (2010) · IEEE International Conference on Intelligent Computer Communication and Processing (ICCP) · doi:10.1109/iccp.2010.5606441 · Cited by 25
“Sequential implementations of increased accuracy algorithms, like the one presented in [14], cannot reach real-time performance.”
BackgroundThe parallelization of video processing
Lin, D. J., Huang, X., Nguyen, Q. H., Blackburn, J., Rodrigues, C. I., Huang, T. S., Do, M. N., Patel, S. J., & Hwu, W. (2009) · IEEE Signal Processing Magazine · doi:10.1109/msp.2009.934116 · Cited by 25
“We have already incorporated the GPU-based optic flow library (OFLib) [ 5 ].”
MethodRe-estimation of Motion and Reconstruction for Distributed Video Coding
Luong, H. V., Rakêt, L. L., & Forchhammer, S. O. (2014) · IEEE Transactions on Image Processing · doi:10.1109/tip.2014.2320364 · Cited by 25
“We minimize (2) using a duality based method [20], [21], and refer to [13] for implementation and parameter settings.”
BackgroundMethodSelf-Occlusions and Disocclusions in Causal Video Object Segmentation
Yang, Y., Sundaramoorthi, G., & Soatto, S. (2015) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.501 · Cited by 25
“(1) is reminiscent of many optical flow estimation algorithms [15, 6, 7, 39], but there are important differences: First, each warp is restricted to a subset Ri ⊂ D with no compatibility condition or relation among the different warps.”
BackgroundAnyFlow: Arbitrary Scale Optical Flow with Implicit Neural Representation
Jung, H., Hui, Z., Luo, L., Yang, H., Liu, F., Yoo, S., Ranjan, R., & Demandolx, D. (2023) · 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/cvpr52729.2023.00528 · Cited by 27
Individual honey bee tracking in a beehive environment using deep learning and Kalman filter
Kongsilp, P., Taetragool, U., & Duangphakdee, O. (2024) · Scientific Reports · doi:10.1038/s41598-023-44718-y · Cited by 24
“It offers users the flexibility to choose from multiple tracking algorithms, including RMOT 14 , Optical Flow DualTVL1 15 , K-means 16 , and GroupTracker 17 .”
MethodMatrix-Free Convex Optimization Modeling
Diamond, S., & Boyd, S. (2016) · Springer optimization and its applications · doi:10.1007/978-3-319-42056-1_7 · Cited by 24
The Moreau envelope approach for the L1/TV image denoising model
Chen, F., Shen, L., Xu, Y., & Zeng, X. (2014) · Inverse Problems and Imaging · doi:10.3934/ipi.2014.8.53 · Cited by 25
On illumination-invariant variational optical flow for weakly textured scenes
Trinh, D. H., & Daul, C. (2018) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2018.11.004 · Cited by 25
Two-Stream Video Classification with Cross-Modality Attention
Chi, L., Tian, G., Mu, Y., & Tian, Q. (2019) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw.2019.00552 · Cited by 28
“We choose the TVL1 optical flow algorithm [41] to extract optical flow for the Flow branch, based on the GPU version from the OpenCV toolbox.”
Zhang, H., Yin, L., Zhang, H., & Wu, X. (2024) · Signal Image and Video Processing · doi:10.1007/s11760-024-03039-x · Cited by 25
“Then, we use TV-L1 [25] to obtain the optical flow ( u , v) between F onset and F apex : where u and v are the horizontal and vertical components of the optical flow, respectively.”
A Filtering-Based Framework for Optical Flow Estimation
Chen, J., Cai, Z., Lai, J., & Xie, X. (2018) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2018.2805101 · Cited by 24
“, the Horn-Schunck model [1], PDE-based models [8]–[11], TV-based models [14], [15], and nonlocal-based models [16], [18], [35], [36], [40]) can be reinterpreted and the new optical flow algorithms can be easily derived by adopting suitable filters for the regularizers.”
BackgroundMethodInfluentialGhosting-free DCT based multi-exposure image fusion
Martorell, O., Sbert, C., & Buades, A. (2019) · Signal Processing Image Communication · doi:10.1016/j.image.2019.07.020 · Cited by 24
Sarabu, A., & Santra, A. K. (2020) · Data · doi:10.3390/data5040104 · Cited by 24
“TVL1 optical flow algorithm [37] is used to extract optical flow frames from videos.”
MethodNesterov Accelerated ADMM for Fast Diffeomorphic Image Registration
Thorley, A., Jia, X., Chang, H. J., Liu, B., Bunting, K. V., Stoll, V., de Marvao, A., O’Regan, D. P., Gkoutos, G. V., Kotecha, D., et al. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-87202-1_15 · Cited by 24
“However, if we set s = 1, the data term will be the sum absolute differences (SAD) [27] which is more robust to outliers in images.”
BackgroundWAS-VTON: Warping Architecture Search for Virtual Try-on Network
Xie, Z., Zhang, X., Zhao, F., Dong, H., Kampffmeyer, M., Yan, H., & Liang, X. (2021) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3474085.3475490 · Cited by 24
“Finally, we introduce the TV loss [52] LTV of the estimated flow to avoid excessive distortion in the warped clothing,”
MethodSecond-Order Ultrasound Elastography With L1-Norm Spatial Regularization
Ashikuzzaman, M., & Rivaz, H. J. (2022) · IEEE Transactions on Ultrasonics Ferroelectrics and Frequency Control · doi:10.1109/tuffc.2022.3141686 · Cited by 24
“Advanced techniques such as alternating direction method of multipliers (ADMM)- and duality-based [46] approaches can be adapted for an improved L1-norm optimization.”
MethodMao, Q., Zhou, L., Zheng, W., Shao, X., & Huang, X. (2022) · IEEE Transactions on Affective Computing · doi:10.1109/taffc.2022.3197785 · Cited by 28
“In the preprocessing, for each micro-expression sequence, we calculate the TV-L1 [51] optical flow from the onset and apex frames.”
MethodInfluentialWeakly-Supervised Online Action Segmentation in Multi-View Instructional Videos
Ghoddoosian, R., Dwivedi, I., Agarwal, N., Choi, C., & Dariush, B. (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/cvpr52688.2022.01341 · Cited by 24
“We extracted I3D features [3] for the IAD dataset using TV-L1 optical flow [57] on a moving window of 16 frames.”
MethodRunia, T. F. H., Snoek, C. G. M., & Smeulders, A. W. M. (2019) · International Journal of Computer Vision · doi:10.1007/s11263-019-01194-0 · Cited by 24
“We also report the repetition count results using TV-L1 (Zach et al. 2007) and EpicFlow (Revaud et al.”
InfluentialFine-grained action recognition using multi-view attentions
Zhu, Y., & Liu, G. (2019) · The Visual Computer · doi:10.1007/s00371-019-01770-y · Cited by 24
“The optical flows are computed using the method of [36].”
MethodDeep Alternative Neural Network: Exploring Contexts as Early as Possible for Action Recognition
Wang, J., Wang, W., Chen, X., Wang, R., & Gao, W. (2016) · neural information processing systems · Cited by 23
Attention‐based spatial–temporal hierarchical ConvLSTM network for action recognition in videos
Xue, F., Ji, H., Zhang, W., & Cao, Y. (2019) · IET Computer Vision · doi:10.1049/iet-cvi.2018.5830 · Cited by 24
“The optical flow used in our paper is extracted by the TVL1 optical flow algorithm [45], which is implemented in OpenCV with CUDA.”
MethodBRAFT: Recurrent All-Pairs Field Transforms for Optical Flow Based on Correlation Blocks
Jia, D., Wang, K., Luo, S., Liu, T., & Liu, Y. (2021) · IEEE Signal Processing Letters · doi:10.1109/lsp.2021.3099076 · Cited by 23
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
