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 201–300 of 1,002
Video Super Resolution Using Duality Based TV-L 1 Optical Flow
Mitzel, D., Pock, T., Schoenemann, T., & Cremers, D. (2009) · Lecture notes in computer science · doi:10.1007/978-3-642-03798-6_44 · Cited by 81
“[1, 13] The solution for Eq.”
BackgroundMethodInfluentialSelf-citationReal-time photometric registration from arbitrary geometry
Gruber, L., Richter-Trummer, T., & Schmalstieg, D. (2012) · Proceedings - International Symposium on Mixed and Augmented Reality, ISMAR · doi:10.1109/ismar.2012.6402548 · Cited by 81
“To remove camera noise and high frequency texture parts, we run a TVL1 de-noising algorithm [20] on the Luminance channel.”
MethodVision-Infused Deep Audio Inpainting
Zhou, H., Liu, Z., Xu, X., Luo, P., & Wang, X. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00037 · Cited by 80
“Optical flows are extracted by using TV-L-1 algorithm [50] and bounded to be maximum 20 pixels.”
MethodReal-Time Online Video Detection with Temporal Smoothing Transformers
Zhao, Y., & Krähenbühl, P. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-19830-4_28 · Cited by 96
Pedestrian intention recognition using Latent-dynamic Conditional Random Fields
Schulz, A., & Stiefelhagen, R. (2015) · IEEE Intelligent Vehicles Symposium · doi:10.1109/ivs.2015.7225754 · Cited by 80
“We use a public available version of the TV-L1 flow [22],[17].”
MethodDeep Manifold Learning Combined With Convolutional Neural Networks for Action Recognition
Chen, X., Weng, J., Lu, W., Xu, J., & Weng, J. (2018) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2017.2740318 · Cited by 80
“We choose the TVL1 optical flow algorithm [56] and use the OpenCV implementation due to its balance between accuracy and efficiency.”
MethodPramono, R. R. A., Chen, Y., & Fang, W. H. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58452-8_5 · Cited by 79
“The two-stream I3D is fine-tuned with a volume of 64 RGB and optical flow images [58] using Adam and a pre-trained Kinetics with a learning rate of 0.”
MethodEye Blink Detection Using Variance of Motion Vectors
Drutarovsky, T., & Fogelton, A. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-16199-0_31 · Cited by 76
“[5] computes the optical flow of face region proposed in [18] and compensate the face movement.”
BackgroundLin, B., Sun, Y., Qian, X., Goldgof, D., Gitlin, R., & You, Y. (2016) · The international journal of medical robotics + computer assisted surgery : MRCAS · doi:10.1002/rcs.1661 · Cited by 87
“In (123), the Total Variation regularization with L1 norm (TV-L1) optical fl ow (127) was applied to establish the correspondences between synthesized and real images.”
MethodUnsupervised Deep Epipolar Flow for Stationary or Dynamic Scenes
Zhong, Y., Ji, P., Wang, J., Dai, Y., & Li, H. (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.2019.01237 · Cited by 77
“Since Horn and Schunck’s seminal work [14], various methods have been developed using variational optimization [2, 43, 5], energy minimization [19, 24, 32, 40], or deep learning [7, 8, 25, 33].”
An Evaluation of Data Costs for Optical Flow
Vogel, C., Roth, S., & Schindler, K. (2013) · Lecture notes in computer science · doi:10.1007/978-3-642-40602-7_37 · Cited by 76
“(iii) Sophisticated optimization schemes [7, 8, 29] have made flow estimation much more efficient.”
BackgroundMethodShared Cross-Modal Trajectory Prediction for Autonomous Driving
Choi, C., Choi, J. H., Li, J., & Malla, S. (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/cvpr46437.2021.00031 · Cited by 76
Motion detail preserving optical flow estimation
Xu, L., Jia, J.-Y., & Matsushita, Y. (2010) · 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2010.5539820 · Cited by 193
“Steinbrücker and Pock [20] extended the numerical scheme of [25] and searched over all possible values for the large displacement flow.”
BackgroundTemporal Action Detection with Structured Segment Networks
Zhao, Y., Xiong, Y., Wang, L., Wu, Z., Tang, X., & Lin, D. (2019) · International Journal of Computer Vision · doi:10.1007/s11263-019-01211-2 · Cited by 91
Joint SFM and detection cues for monocular 3D localization in road scenes
Song, S., & Chandraker, M. K. (2015) · 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.2015.7298997 · Cited by 75
“Instead, we use a combination of dense intensity-aligned pose estimation [9], an epipolar-guided extension [16] to TV-L1 optical flow of [22] and consistency checks in the spirit of sparse SFM, to extract 3D points for our joint optimization framework.”
MethodHierarchical Attention Network for Action Recognition in VideosPreprint
Wang, Y., Wang, S., Tang, J., O’Hare, N., Chang, Y., & Li, B. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1607.06416 · Cited by 82
“Optical flow: The optical flow is computed by the off-the-shelf OpenCV toolbox with GPU implementation of [23].”
MethodMimetics: Towards Understanding Human Actions Out of Context
Weinzaepfel, P., & Rogez, G. (2021) · International Journal of Computer Vision · doi:10.1007/s11263-021-01446-y · Cited by 74
“For optical flow input, we use the TV-L1 algorithm [53].”
MethodMotion Estimation Using the Correlation Transform
Drulea, M., & Nedevschi, S. (2013) · IEEE Transactions on Image Processing · doi:10.1109/tip.2013.2263149 · Cited by 72
Multi-Stream Deep Neural Networks for RGB-D Egocentric Action Recognition
Tang, Y., Wang, Z., Lu, J., Feng, J., & Zhou, J. (2018) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2018.2875441 · Cited by 71
“We employed optical flow based on TVL1 algorithm [84], which achieves a good trade-off between accuracy and”
MethodInfluentialZero-Shot Depth Estimation From Light Field Using A Convolutional Neural Network
Peng, J., Xiong, Z., Wang, Y., Zhang, Y., & Liu, D. (2020) · IEEE Transactions on Computational Imaging · doi:10.1109/tci.2020.2967148 · Cited by 71
“target frame using optical flow computed with a conventional method [49].”
MethodFastFlowNet: A Lightweight Network for Fast Optical Flow Estimation
Kong, L., Shen, C., & Yang, J. (2021) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48506.2021.9560800 · Cited by 71
Spatio-Temporal Vector of Locally Max Pooled Features for Action Recognition in Videos
Duţă, I. C., Ionescu, B. E., Aizawa, K., & Sebe, N. (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.2017.341 · Cited by 71
“To extract optical flow fields we use the OpenCV implementation of the TVL1 algorithm [53].”
MethodLearning Long-Term Dependencies for Action Recognition with a Biologically-Inspired Deep Network
Shi, Y., Tian, Y., Wang, Y., Zeng, W., & Huang, T. (2017) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.84 · Cited by 72
“To extract optical flow, we choose the TVL1 optical flow algorithm [45] and use the OpenCV GPU implementation.”
MethodP-ODN: Prototype-based Open Deep Network for Open Set Recognition
Shu, Y., Shi, Y., Wang, Y., Huang, T., & Tian, Y. (2020) · Scientific Reports · doi:10.1038/s41598-020-63649-6 · Cited by 73
“For the extraction of optical flow, we choose the TVL1 optical flow algorithm 23 implemented in OpenCV with CUDA.”
MethodA Nonlocal Structure Tensor-Based Approach for Multicomponent Image Recovery Problems
Chierchia, G., Pustelnik, N., Pesquet‐Popescu, B., & Pesquet, J.-C. (2014) · IEEE Transactions on Image Processing · doi:10.1109/tip.2014.2364141 · Cited by 70
“A first approach consists of computing TV channel-by-channel and then summing up the resulting smoothness measures [23, 24,25,26].”
MethodD2-Net: Weakly-Supervised Action Localization via Discriminative Embeddings and Denoised Activations
Narayan, S., Cholakkal, H., Hayat, M., Khan, F. S., Yang, M., & Shao, L. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.01335 · Cited by 70
“Implementation details: The optical flow frames of a video are generated using the TV-L1 flow [36].”
MethodDetector-Free Weakly Supervised Group Activity Recognition
Kim, D.-K., Lee, J., Cho, M., & Kwak, S. (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/cvpr52688.2022.01945 · Cited by 69
“To incorporate motion information into the features without using computationally heavy 3D CNNs [9, 47] or optical flows [46, 60], our model computes local correlation between two adjacent intermediate feature maps, F ( t ) and F ( t +1) , and encodes the correlations into frame-wise motion…”
MethodSchulz, A., & Stiefelhagen, R. (2015) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc.2015.37 · Cited by 69
“We use a public available version of the TV-L1 flow [17] and the method of [6] for the calculation of stereo disparity maps.”
MethodMultimodal Fusion via Teacher-Student Network for Indoor Action Recognition
Yu, B. X. B., Liu, Y., & Chan, K. (2021) · AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v35i4.16430 · Cited by 84
“The optical flow data is usually extracted from the RGB modality by using the TV-L 1 algorithm (Zach, Pock, and Bischof 2007).”
MethodSign Language Translation with Hierarchical Spatio-Temporal Graph Neural Network
Kan, J., Hu, K., Hagenbuchner, M., Tsoi, A. C., Bennamoun, M., & Wang, Z. (2022) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv51458.2022.00219 · Cited by 68
Proposal-Based Multiple Instance Learning for Weakly-Supervised Temporal Action Localization
Ren, H., Yang, W., Zhang, T., & Zhang, Y. (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/cvpr52729.2023.00237 · Cited by 69
“Optical flow frames are extracted from RGB frames using the TV-L1 [58] algorithm.”
Illumination-Robust Optical Flow Using a Local Directional Pattern
Mohamed, M. A., Rashwan, H. A., Mertsching, B., García, M. Á., & Puig, D. (2014) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2014.2308628 · Cited by 68
“Furthermore, an adapted variational optical flow model based on [32] to be applicable with binary descriptors”
MethodInfluentialTechnical Foundation and Calibration Methods for Time-of-Flight Cameras
Lefloch, D., Nair, R., Lenzen, F., Schäfer, H., Streeter, L. V., Cree, M. J., Koch, R., & Kolb, A. (2013) · Lecture notes in computer science · doi:10.1007/978-3-642-44964-2_1 · Cited by 68
“[48] on a 2009 machine.”
AssembleNet: Searching for Multi-Stream Neural Connectivity in Video ArchitecturesPreprint
Ryoo, M. S., Piergiovanni, A., Tan, M., & Angelova, A. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1905.13209 · Cited by 90
A non-local structure tensor based approach for multicomponent image recovery problemsPreprint
Chierchia, G., & Pustelnik, N. (2016) · Cited by 67
A Super-Resolution Framework for High-Accuracy Multiview Reconstruction
Goldlüecke, B., Aubry, M., Kolev, K., & Cremers, D. (2013) · International Journal of Computer Vision · doi:10.1007/s11263-013-0654-8 · Cited by 66
“It was successfully employed in previous work for the optical flow functional [46], a method we also use again in later sections.”
Two-Stream Flow-Guided Convolutional Attention Networks for Action Recognition
Tran, A., & Cheong, L. (2017) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw.2017.368 · Cited by 68
Manipulation-Skill Assessment from Videos with Spatial Attention Network
Li, Z., Huang, Y., Cai, M., & Sato, Y. (2019) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw.2019.00539 · Cited by 65
“The optical flow images for motion input are extracted by TVL(1) algorithm [54].”
MethodTemporal Gaussian Mixture Layer for VideosPreprint
Piergiovanni, A., & Ryoo, M. S. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1803.06316 · Cited by 67
“We extracted frames from the videos at 25 fps, computed TVL1 (Zach et al., 2007) optical flow, clipped to [ − 20 , 20] .”
MethodA General Dynamic Knowledge Distillation Method for Visual Analytics
Tu, Z., Liu, X., & Xiao, X. (2022) · IEEE Transactions on Image Processing · doi:10.1109/tip.2022.3212905 · Cited by 67
“We utilize the TV-L1 algorithm [54] to compute optical flow with default OpenCV parameter settings.”
MethodSelfME: Self-Supervised Motion Learning for Micro-Expression Recognition
Fan, X., Chen, X., Jiang, M., Shahid, A. R., & Yan, H. (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/cvpr52729.2023.01329 · Cited by 69
Improved Simultaneous Computation of Motion Detection and Optical Flow for Object Tracking
Denman, S., Fookes, C., & Sridharan, S. (2009) · International Conference on Digital Image Computing: Techniques and Applications (DICTA) · doi:10.1109/dicta.2009.35 · Cited by 65
General Dynamic Scene Reconstruction from Multiple View Video
Mustafa, A., Kim, H., Guillemaut, J., & Hilton, A. (2015) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.109 · Cited by 61
“Dynamic elements of the scene are identified by performing optical flow [37] on consecutive frames for a single view of each cluster.”
BackgroundUnsupervised Deep Video Denoising
Sheth, D. Y., Mohan, S., Vincent, J., Manzorro, R., Crozier, P. A., Khapra, M. M., Simoncelli, E. P., & Fernandez‐Granda, C. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00178 · Cited by 73
“The idea is to register contiguous frames using the optical flow (obtained from TV-L1 [44]), and treat them as noisy realizations of the same clean image.”
BackgroundCompensation of Motion Artifacts for Time-of-Flight Cameras
Lindner, M., & Kolb, A. (2009) · Lecture notes in computer science · doi:10.1007/978-3-642-03778-8_2 · Cited by 64
“al [11].”
Fast gain-adaptive KLT tracking on the GPU
Zach, C., Gallup, D., & Frahm, J. (2008) · 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops · doi:10.1109/cvprw.2008.4563089 · Cited by 64
Self-citationLightPIVNet: An Effective Convolutional Neural Network for Particle Image Velocimetry
Yu, C., Bi, X., Fan, Y., Han, Y., & Kuai, Y.-F. (2021) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2021.3082313 · Cited by 63
Human Action Recognition Based on Transfer Learning Approach
AbdulAzeem, Y., Balaha, H. M., Bahgat, W. M., & Badawy, M. (2021) · IEEE Access · doi:10.1109/access.2021.3086668 · Cited by 63
“It is also used to generate OF frames fed to the temporal stream of the CNN. Ma et al. [41] showed that extracting OF images using TV-L1 [42] is better than using Brox [43].”
MethodSelf-supervised Video Representation Learning by Context and Motion Decoupling
Huang, L., Yu, L., Wang, B., Pan, P., Xu, Y., & Jin, R. (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/cvpr46437.2021.01367 · Cited by 63
“Considering the scalability of our framework on larger datasets, we endeavor to avoid the use of computationally expensive features such as optical flow [59, 6] and dense trajectories [47].”
MethodSpatio-Temporal Person Retrieval via Natural Language Queries
Yamaguchi, M., Saito, K., Ushiku, Y., & Harada, T. (2017) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.162 · Cited by 63
“Following the CUHK team, we extract the optical flow maps by using the TV-L1 optical flow algorithm [48] and input the optical flow maps corresponding to the five frames preceding the target timestep and the five frames following the target timestep into the model.”
MethodSequential fusion of facial appearance and dynamics for depression recognition
Chen, Q., Chaturvedi, I., Ji, S., & Cambria, E. (2021) · Pattern Recognition Letters · doi:10.1016/j.patrec.2021.07.005 · Cited by 61
“We compute the optical flow with the duality-based approach [52] , which is a decent method with sufficiently fast speed.”
MethodEAN: Event Adaptive Network for Enhanced Action Recognition
Tian, Y., Yan, Y., Zhai, G., Guo, G., & Gao, Z. (2022) · International Journal of Computer Vision · doi:10.1007/s11263-022-01661-1 · Cited by 64
“Both conventional or CNN-based optical flow estimation methods (Zach et al., 2007; Ilg et al., 2017; Ranjan&Black, 2017; Sun et al., 2018) can be adopted in this framework.”
MethodMotion-aware Contrastive Video Representation Learning via Foreground-background Merging
Ding, S., Li, M., Yang, T., Qian, R., Xu, H., Chen, Q., Wang, J., & Xiong, H. (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/cvpr52688.2022.00949 · Cited by 63
“Apart from frame difference, we also consider other methods to convey motion information like optical flow [65].”
MethodZhang, L., Lim, C. P., & Yu, Y. (2021) · Knowledge-Based Systems · doi:10.1016/j.knosys.2021.106918 · Cited by 59
Two-Stream Action Recognition-Oriented Video Super-Resolution
Zhang, H., Liu, D., & Xiong, Z. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00889 · Cited by 62
“Here the optical flow is calculated by the TVL(1) method [43] as most action recognition works do.”
MethodResultInfluentialOptical Flow with Geometric Occlusion Estimation and Fusion of Multiple Frames
Kennedy, R., & Taylor, C. J. (2015) · Energy Minimization Methods in Computer Vision and Pattern Recognition · doi:10.1007/978-3-319-14612-6_27 · Cited by 84
“93% DB-TV-L1 [34] 14.”
SpatialTracker: Tracking Any 2D Pixels in 3D Space
Xiao, Y.-X., Wang, Q.-Q., Zhang, S.-Z., Xue, N., Peng, S.-D., Shen, Y., & Zhou, X.-W. (2024) · Computer Vision and Pattern Recognition · doi:10.1109/cvpr52733.2024.01929 · Cited by 230
“While traditional methods [1, 6, 8, 19, 74, 78] formulate it as an energy minimization problem, recent approaches [13, 23, 24, 61, 76] have demonstrated the ability to predict optical flow directly using deep neural networks.”
MethodContour Motion Estimation for Asynchronous Event-Driven Cameras
Barranco, F., Fermuller, C. M., & Aloimonos, Y. (2014) · Proceedings of the IEEE · doi:10.1109/jproc.2014.2347207 · Cited by 59
“They use different penalty functions; for example, the L1-based minimizations have been shown to be very successful [24], [25].”
A Fourth-Order Variational Image Registration Model and Its Fast Multigrid Algorithm
Chumchob, N., Chen, K., & Loeza, C. F. B. (2011) · Multiscale Modeling and Simulation · doi:10.1137/100788239 · Cited by 58
Markerless Shape and Motion Capture From Multiview Video Sequences
Li, K., Dai, Q., & Xu, W. (2011) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2011.2106251 · Cited by 58
Dynamic Video Deblurring Using a Locally Adaptive Blur Model
Kim, T. H., Nah, S., & Lee, K. M. (2017) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2017.2761348 · Cited by 58
Emotion recognition from an ensemble of features
Tariq, U., Lin, K.-H., Li, Z., Zhou, X., Wang, Z., Le, V., Huang, T. S., Lv, X., & Han, T. X. (2011) · International Conference on Automatic Face and Gesture Recognition · doi:10.1109/fg.2011.5771365 · Cited by 57
“For this reason we use the algorithm implemented on GPU [18] which decreases computation time by orders of magnitude.”
MethodDepthwise Spatio-Temporal STFT Convolutional Neural Networks for Human Action Recognition
Kumawat, S., Verma, M., Nakashima, Y., & Raman, S. (2021) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2021.3076522 · Cited by 56
“A major drawback of the two-stream networks is that they require to pre-compute optical flow, which is expensive in terms of both time and space [39].”
BackgroundMultimodal information fusion for urban scene understanding
Xu, P., Davoine, F., Bordes, J.-B., Zhao, H., & Denœux, T. (2014) · Machine Vision and Applications · doi:10.1007/s00138-014-0649-7 · Cited by 56
“We used the OpenCV implementation of the TV-L1 formulation as proposed by Zach et al. [36].”
MethodAttention-Aware Sampling via Deep Reinforcement Learning for Action Recognition
Dong, W., Zhang, Z., & Tan, T. (2019) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v33i01.33018247 · Cited by 55
“For the extraction of optical flow, we choose the TVL1 (Zach, Pock, and Bischof 2007) optical flow algorithm implemented in OpenCV with CUDA.”
MethodEfficient Nonlocal Regularization for Optical Flow
Krähenbühl, P., & Koltun, V. (2012) · Lecture notes in computer science · doi:10.1007/978-3-642-33718-5_26 · Cited by 56
“Here ui is the flow vector at pixel i, ED is the data term, and EL is the traditional grid-connected regularization term [5, 8, 11, 25].”
BackgroundInfluentialLearning to detect Motion Boundaries
Weinzaepfel, P., Revaud, J., Harchaoui, Z., & Schmid, C. (2015) · 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.2015.7298873 · Cited by 55
“TV-L1 uses the dual space for minimizing this energy.”
MethodClassifier based graph construction for video segmentation
Khoreva, A., Galasso, F., Hein, M., & Schiele, B. (2015) · 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.2015.7298697 · Cited by 55
“We consider an optical flow estimate [58], which we smooth spatially (preserving the across-superpixels boundaries with bilateral filtering) and temporally (median filtered ±2 frames).”
A survey of micro-expression recognition
Zhou, L., Shao, X., & Mao, Q. (2020) · Image and Vision Computing · doi:10.1016/j.imavis.2020.104043 · Cited by 57
Tamou, A. B., Benzinou, A., & Nasreddine, K. (2021) · Applied Intelligence · doi:10.1007/s10489-020-02155-8 · Cited by 56
“In this work, we use total variation (TV) regularization and the robust L 1 norm (TV-L 1 ) algorithm [49].”
MethodAn Unbiased Second-Order Prior for High-Accuracy Motion Estimation
Trobin, W., Pock, T., Cremers, D., & Bischof, H. (2008) · Lecture notes in computer science · doi:10.1007/978-3-540-69321-5_40 · Cited by 54
“First, we demonstrate the effects of the new second-order prior by comparing the proposed model to an implementation of the TV-regularized model [8] on the Middlebury “Venus” sequence.”
BackgroundMethodInfluentialSelf-citationJoint motion estimation and segmentation of complex scenes with label costs and occlusion modeling
Unger, M., Werlberger, M., Pock, T., & Bischof, H. (2012) · 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.2012.6247887 · Cited by 54
“We use recent advances in solving the Potts model [9, 27, 14] to derive a novel primaldual energy minimization problem.”
MethodSelf-citationDense versus Sparse Approaches for Estimating the Fundamental Matrix
Valgaerts, L., Bruhn, A., Mainberger, M., & Weickert, J. (2011) · International Journal of Computer Vision · doi:10.1007/s11263-011-0466-7 · Cited by 54
Ye, W., Cheng, J., Yang, F., & Xu, Y. (2019) · IEEE Access · doi:10.1109/access.2019.2918808 · Cited by 54
“There are two common kinds of algorithm for optical flow extracting—Brox [21] and TV-L1(total variation-L1) [22].”
BackgroundComputational Visual Attention
Frintrop, S. (2011) · Springer eBooks · doi:10.1007/978-0-85729-994-9_4 · Cited by 54
“Here, we use a method based on total variation regularization that determines a dense optical flow field and is capable to operate in real-time [29].”
MethodMou, L., Zhou, C., Xie, P., Zhao, P., Jain, R., Gao, W., & Yin, B. (2021) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2021.3128738 · Cited by 55
“Finally, an improved total variation (TV) regularization and the robust L 1 norm (TV-L 1 ) method [54] are used to extract the optical flow of head movement.”
MethodGesture Recognition using Wearable Vision Sensors to Enhance Visitors' Museum Experiences
Baraldi, L., Paci, F., Serra, G., Luca Benini, C., & Cucchiara, R. (2015) · IEEE Sensors Journal · doi:10.1109/jsen.2015.2411994 · Cited by 53
“The choice of this particular optical flow algorithm is induced by our preliminary tests, in which Farneback’s optical flow showed the best performance when compared to other popular optical flow algorithms, such as TV-L1 [26] and SimpleFlow [27].”
MethodReal-Time Dense and Accurate Parallel Optical Flow using CUDA
Marzat, J., Dumortier, Y., & Ducrot, A. (2009) · Digital Library (University of West Bohemia) · Cited by 53
Spatial-temporal pyramid based Convolutional Neural Network for action recognition
Zheng, Z., An, G., Wu, D. O., & Ruan, Q. (2019) · Neurocomputing · doi:10.1016/j.neucom.2019.05.058 · Cited by 52
Raudies, F., Gilmore, R. O., Kretch, K. S., Franchak, J. M., & Adolph, K. E. (2012) · IEEE International Conference on Development and Learning (ICDL) · doi:10.1109/devlrn.2012.6400584 · Cited by 51
Burger, M., Dirks, H., Frerking, L., Hauptmann, A., Helin, T., & Siltanen, S. (2017) · Inverse Problems · doi:10.1088/1361-6420/aa99cf · Cited by 51
A fast human action recognition network based on spatio-temporal features
Xu, J., Song, R., Wei, H., Guo, J., Zhou, Y., & Huang, X. (2020) · Neurocomputing · doi:10.1016/j.neucom.2020.04.150 · Cited by 50
Spatio-Temporal Adaptive Network With Bidirectional Temporal Difference for Action Recognition
Li, Z., Li, J., Ma, Y., Wang, R., Shi, Z., Ding, Y., & Liu, X. (2023) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2023.3250646 · Cited by 52
“extract the optical flow [19] to represent the short-term motion changes.”
BackgroundInceptr: micro-expression recognition integrating inception-CBAM and vision transformer
Zhou, H., Huang, S., & Xu, Y. (2023) · Multimedia Systems · doi:10.1007/s00530-023-01164-0 · Cited by 53
“Therefore, we employ the TV-L1 [43] algorithm to extract optical flow as motion variation information.”
MethodTwo-Stream Region Convolutional 3D Network for Temporal Activity Detection
Xu, H., Das, A., & Saenko, K. (2019) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2019.2921539 · Cited by 49
“Specifically, we chose the following scales - [1, 2, 3, 4, 5, 6, 7, 8, 10, 12, 14, 16, 20, 24, 28, 32, 40, 48, 56, 64].”
BackgroundMethodMemFlow: Optical Flow Estimation and Prediction with Memory
Dong, Q., & Fu, Y. (2024) · 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/cvpr52733.2024.01804 · Cited by 63
“Traditionally, it is solved by optimizing energy function [2–5, 24, 50, 70] to maximize visual similarity between images.”
BackgroundRealtime phase-based optical flow on the GPU
Pauwels, K., & Hulle, M. M. V. (2008) · 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops · doi:10.1109/cvprw.2008.4563090 · Cited by 49
“These results are very similar to results obtained with state-of-the-art algorithms and realtime implementations [12, 16].”
ResultHigher-order total variation approaches and generalisations
Bredies, K., & Höller, M. (2020) · Inverse Problems · doi:10.1088/1361-6420/ab8f80 · Cited by 49
Crowd Event Detection on Optical Flow Manifolds
Rao, A. S., Gubbi, J., Marusic, S., & Palaniswami, M. S. (2015) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2015.2451136 · Cited by 49
“[60] applied the method of the total variations using the L1 norm of penalizing the flow variations as opposed to the quadratic approach”
MethodGüney, F., & Geiger, A. (2016) · Asian Conference on Computer Vision · doi:10.1007/978-3-319-54190-7_13 · Cited by 75
Rethinking Optical Flow from Geometric Matching Consistent Perspective
Dong, Q., Cao, C., & Fu, Y. (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/cvpr52729.2023.00135 · Cited by 48
Ordered Pooling of Optical Flow Sequences for Action Recognition
Wang, J., Cherian, A., & Porikli, F. (2017) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv.2017.26 · Cited by 48
“We use the TVL1 optical flow [32] algorithm to generate the flow images using its OpenCV implementation.”
MethodMaking Convolutional Networks Recurrent for Visual Sequence Learning
Yang, X., Molchanov, P. A., & Kautz, J. (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.2018.00677 · Cited by 49
“Our temporal network employs the pre-computed optical flow [50] stacking with 10 frames.”
MethodLuqman, H., & El-Alfy, E.-S. M. (2021) · Electronics · doi:10.3390/electronics10141739 · Cited by 48
A TV-L1 Optical Flow Method with Occlusion Detection
Ballester, C., Garrido, L., Lazcano, V. A., & Caselles, V. (2012) · Lecture notes in computer science · doi:10.1007/978-3-642-32717-9_4 · Cited by 47
“As in [19], in order to cope with the nonlinearities of both Ed(u, χ) and Er(u, χ) we introduce an auxiliary variable v representing the optical flow and we penalize its deviation form u.”
BackgroundMethodAction recognition in RGB-D egocentric videos
Tang, Y., Tian, Y., Lu, J., Feng, J., & Zhou, J. (2017) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2017.8296915 · Cited by 47
“In order to extract optical flow fields as we mention above, we adopt the TVL1 algorithm [33] due to its easy usage and promising performance.”
MethodFirst Person Action Recognition via Two-stream ConvNet with Long-term Fusion Pooling
Kwon, H., Kim, Y.-H., Lee, J. S., & Cho, M. (2018) · Pattern Recognition Letters · doi:10.1016/j.patrec.2018.07.011 · Cited by 46
Action recognition using spatial-optical data organization and sequential learning framework
Yuan, Y., Zhao, Y., & Wang, Q. (2018) · Neurocomputing · doi:10.1016/j.neucom.2018.06.071 · Cited by 48
ACM-Net: Action Context Modeling Network for Weakly-Supervised Temporal Action LocalizationPreprint
Qu, S., Chen, G., Li, Z., Zhang, L., Fan, L., & Knoll, A. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2104.02967 · Cited by 59
“3) Implementation Details: For the feature extraction, we first sample RGB frames at 25 fps for each video and apply the TV-L1 algorithm [52] to generate optical flow frames.”
An Approach to Streaming Video Segmentation With Sub-Optimal Low-Rank Decomposition
Li, C., Lin, L., Zuo, W., Wang, W., & Tang, J. (2016) · IEEE Transactions on Image Processing · doi:10.1109/tip.2016.2537211 · Cited by 46
“Herein, the baseline introduced by [8] is the extension of a state-of-the-art image segmentation method [47] by propagating the segmentation [47] of the central frame to the other frames with optical flow [48] and labelling the image segments”
Method
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
