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 701–800 of 1,002
Mould, N. A., Regens, J. L., Jensen, C. J., & Edger, D. N. (2014) · Journal of Policing Intelligence and Counter Terrorism · doi:10.1080/18335330.2014.940819 · Cited by 13
“In Feng et al. ’ s (2010) study, optical flow (Zach, Pock, & Bischof, 2007) vectors are computed at each pixel location and two-dimensional Gaussians are used to characterise the distributions of flow vectors in non-overlapping video blocks.”
MethodApparent resolution enhancement for animations
Templin, K., Didyk, P., Ritschel, T., Eisemann, E., Myszkowski, K., & Seidel, H. (2011) · Spring Conference on Computer Graphics · doi:10.1145/2461217.2461230 · Cited by 13
“The function f can be obtained by a separate render pass in any advanced 3D animation authoring software (a so-called “motion vector pass”) or – in the case of pre-existing animations or video material – using optical flow estimation methods [Horn and Schunck 1981; Zach et al. 2007].”
MethodReal time motion estimation using a neural architecture implemented on GPUs
García‐Rodríguez, J., Orts‐Escolano, S., Angelopoulou, A., Ψαρρού, Α., Azorín-López, J., & García‐Chamizo, J. M. (2014) · Journal of Real-Time Image Processing · doi:10.1007/s11554-014-0417-y · Cited by 13
“Due to temporal restrictions in most applications, some of the methods proposed relaxed versions of the algorithm to accelerate its processing [14, 15, 16, 16].”
MethodSpatio-Temporal Fusion Networks for Action Recognition
Cho, S., & Foroosh, H. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-20887-5_22 · Cited by 12
“In: IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015 (2015) [51] Zach, C.”
BackgroundMethodTwo-Level Attention Model Based Video Action Recognition Network
Sang, H., Zhao, Z., & He, D. (2019) · IEEE Access · doi:10.1109/access.2019.2936628 · Cited by 14
“spatial information of the action, the temporal stream 2D CNN processes the stacked optical flow field composed of consecutive multiple frames of dense optical flow graph [9] to obtain motion information, then fuse the probability scores of the two networks.”
MethodVisual Quality Assessment for Interpolated Slow-Motion Videos Based on a Novel Database
Men, H., Hosu, V., Lin, H., Bruhn, A., & Saupe, D. (2020) · International Conference on Quality of Multimedia Experience (QoMEX) · doi:10.1109/qomex48832.2020.9123096 · Cited by 13
“0 KoSMo-1k [5] [8] Classic+NL (ClassicNL) [26] 45% 21% 1 OAR-Flow (OAR) [27] – – 2 Black & Anandan (BA) [28] [29] 48% 26% 3 BeyondPixles (Beyond) [30] – – 4 Dual-TVL1 (DualTVL1) [31] – – 5 FFV1MT [32] 6% 24% 6 LKpyramid (LK) [33] 2% 6% 7 2D-CLG (CLG) [34] [35] 38% 42% 8 Brox et al.”
Vision-based Individual Factors Acquisition for Thermal Comfort Assessment in a Built Environment
Liu, J., Foged, I. W., & Moeslund, T. B. (2020) · International Conference on Automatic Face and Gesture Recognition · doi:10.1109/fg47880.2020.00057 · Cited by 13
“It is to be noted that the word “input” in the previous paragraph is with a quotation mark since it has several modalities: (i) an original 24-bit thermal image; (ii) a concatenated image of an 8-bit thermal image (the first channel of the original 24-bit one) and its two optical flow fields (in horizontal and vertica…”
MethodTowards Practical Compressed Video Action Recognition: A Temporal Enhanced Multi-Stream Network
Li, B., Kong, L., Zhang, D., Bao, X., Huang, D., & Wang, Y. (2021) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr48806.2021.9412213 · Cited by 13
“For instance, the Two-Stream Network [1], designs two separate CNNs which keep RGB sequence as the appearance modality and the optical flow [3] as the motion modality.”
BackgroundIllumination Invariant Cost Functions in Semi-Global Matching
Hermann, S., Morales, S., Vaudrey, T., & Klette, R. (2011) · Lecture notes in computer science · doi:10.1007/978-3-642-22819-3_25 · Cited by 13
“At the moment, state-of-the-art optimisation strategies can be split into four major groups: belief propagation [7], graph-cuts [3], dynamic programming [17] which has been extended to a semi-global-matching technique (SGM) [10], and variational techniques [4, 23].”
BackgroundDPU: Dynamic Prototype Updating for Multimodal Out-of-Distribution Detection
Li, S., Gong, H., Dong, H., Yang, T., Tu, Z., & Zhao, Y. (2025) · 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/cvpr52734.2025.00953 · Cited by 12
Mitra, S., & Soundararajan, R. (2022) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3503161.3548064 · Cited by 12
“The optical flow map is estimated using TV-L1 algorithm [48].”
MethodModeling Spatio-Temporal Dynamical Systems With Neural Discrete Learning and Levels-of-Experts
Wang, K., Wu, H., Zhang, G., Fang, J., Liang, Y., Wu, Y., Zimmermann, R., & Wang, Y. (2024) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2024.3363711 · Cited by 12
“Substantial advancements in optical flow estimation have been achieved with the incorporation of deep learning [72], [73], [74].”
MethodFlow Dynamics Correction for Action Recognition
Wang, L., & Koniusz, P. (2024) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp48485.2024.10446223 · Cited by 12
Rohlfs, L., & Weiss, J. (2024) · Physics of Fluids · doi:10.1063/5.0208040 · Cited by 12
UnSAMFlow: Unsupervised Optical Flow Guided by Segment Anything Model
Yuan, S., Luo, L., Hui, Z., Pu, C., Xiang, X., Ranjan, R., & Demandolx, D. (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/cvpr52733.2024.01800 · Cited by 12
“Unsupervised optical flow Traditional methods [18, 36, 70] optimize optical flow based on brightness constancy and local smoothness.”
MethodIOF-Tracker: A Two-Stage Multiple Targets Tracking Method Using Spatial-Temporal Fusion Algorithm
Liu, H., Zhao, Y., Dong, P., Guo, X., & Wang, Y. (2024) · Applied Sciences · doi:10.3390/app15010107 · Cited by 13
Real-Time Estimation of Optical Flow Based on Optimized Haar Wavelet Features
Salmen, J., Caup, L., & Igel, C. (2011) · Lecture notes in computer science · doi:10.1007/978-3-642-19893-9_31 · Cited by 11
“However, different improvements and extensions have been proposed recently which address these issues [12–15].”
Practical Algorithms of Spectral Clustering: Toward Large-Scale Vision-Based Motion Analysis
Sakai, T., & Imiya, A. (2010) · Advances in computer vision and pattern recognition · doi:10.1007/978-0-85729-057-1_1 · Cited by 12
“For each pair of subsequent frames of this sequence, we computed the dense optical-flow field by a TV-L 1 approach presented in [ 34 ].”
MethodApproximate nearest neighbor fields in video
Ben-Zrihem, N., & Zelnik‐Manor, L. (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.2015.7299160 · Cited by 12
“Another group of works that find matches between patches in video are those that estimate the optical flow field [35, 30, 17, 6].”
BackgroundA Prediction-Correction Approach for Real-Time Optical Flow Computation Using Stereo
Derome, M., Plyer, A., Sanfourche, M., & Besnerais, G. L. (2016) · Lecture notes in computer science · doi:10.1007/978-3-319-45886-1_30 · Cited by 12
“In the following, we consider High Accuracy Optical Flow (HAOF) [1], TV-L1 optical flow [25] and Large Displacement Optical Flow (LDOF) [2] algorithms,”
BackgroundMethodThe Isowarp: The Template-Based Visual Geometry of Isometric Surfaces
Casillas-Pérez, D., Pizarro, D., Fuentes-Jiménez, D., Mazo, M., & Bartoli, A. (2021) · International Journal of Computer Vision · doi:10.1007/s11263-021-01472-w · Cited by 12
“The most common smoother is based on penalizing the norm of the second-order derivatives of the warp and is known as the bending energy [8,20,40].”
BackgroundZhong, C., Reibman, A. R., Mina, H. A., & Deering, A. J. (2021) · Journal of Imaging · doi:10.3390/jimaging7090170 · Cited by 12
“We compute optical flow using TV-L1 [72] and rescale the result to the range [0, 255].”
Khan, M., & Kumar, P. (2023) · Expert Systems with Applications · doi:10.1016/j.eswa.2023.119628 · Cited by 13
Dense 3D Reconstruction with a Hand-Held Camera
Ummenhofer, B., & Brox, T. (2012) · Lecture notes in computer science · doi:10.1007/978-3-642-32717-9_11 · Cited by 12
“This decoupling approach allows for an efficient GPU implementation and has been used in many related problems [22, 17, 14].”
MethodVariational optic flow on the Sony PlayStation 3
Gwosdek, P., Bruhn, A., & Weickert, J. (2009) · Journal of Real-Time Image Processing · doi:10.1007/s11554-009-0132-2 · Cited by 12
“Even with this approach, which gives much better results than the simpler linear variant, we obtain results in real-time and outperform CPU and GPU implementations of variational optical flow models significantly [11, 44].”
Wedel, A., & Cremers, D. (2011) · Springer eBooks · doi:10.1007/978-0-85729-965-9_2 · Cited by 4
Tu, Z., Hürst, W., Xie, W. C., Xiong, S., & Qin, Q. (2012) · Optical Engineering · doi:10.1117/1.oe.51.3.037202 · Cited by 12
Kitt, B., Ranft, B., & Lategahn, H. (2010) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc.2010.5625181 · Cited by 12
Multi-image interpolation based on graph-cuts and symmetric optical flow
Linz, C., Lipski, C., & Magnor, M. A. (2010) · ACM SIGGRAPH Posters · doi:10.1145/1836845.1836983 · Cited by 15
NONLOCAL VARIATIONAL MODELS FOR INPAINTING AND INTERPOLATION
Arias, P., Caselles, V., Facciolo, G., Lazcano, V. A., & Sadek, R. (2012) · Mathematical Models and Methods in Applied Sciences · doi:10.1142/s0218202512300037 · Cited by 12
Hulens, D., Aerts, B., Chakravarty, P., Diba, A., Goedemé, T., Roussel, T., Zegers, J., Tuytelaars, T., Eycken, L. V., Gool, L. V., et al. (2018) · Lecture notes in computer science · doi:10.1007/978-3-319-73603-7_42 · Cited by 12
“The two stream approaches utilize two modalities of data: RGB frames and motion Optical-Flow [22] frames.”
MethodV-BMS360: A Video Extention to the BMS360 Image Saliency Model
Lebreton, P., Fremerey, S., & Raake, A. (2018) · IEEE International Conference on Multimedia & Expo Workshops (ICMEW) · doi:10.1109/icmew.2018.8551523 · Cited by 12
Li, L., Qin, S., Lu, Z., Zhang, D., Xu, K., & Hu, Z. (2021) · Pattern Analysis and Applications · doi:10.1007/s10044-021-00965-1 · Cited by 12
“In order to further improve the performance on IsoGD, we generated additional optical flow data from RGB videos using the TV-L1 algorithm [52].”
MethodDecoupling photometry and geometry in dense variational camera calibration
Aubry, M., Kolev, K., Goldlüecke, B., & Cremers, D. (2011) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2011.6126396 · Cited by 12
“In the proposed formulation, the first part of the energy is comprised of the data term and the derivative part of the Sobolev norm It resembles a TV-L 1 optical flow model and can be minimized with the algorithm detailed in [24].”
MethodImproving acoustic event detection using generalizable visual features and multi-modality modeling
Huang, P.-S., Zhuang, X., & Hasegawa‐Johnson, M. (2011) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2011.5946412 · Cited by 12
“We utilize a highly efficient algorithm on variational methods utilizing a GPU [10] to calculate the optical flow, i.e. the horizontal and vertical movement for each pixel.”
MethodRobust Discontinuity Preserving Optical Flow Methods
Monzón, N., Salgado, A., & Sánchez, J. (2016) · Image Processing On Line · doi:10.5201/ipol.2016.172 · Cited by 12
Roy, A., Hammer, A., Heinemann, D., Lünsdorf, O., & Lezaca, J. (2022) · Environmental Research Letters · doi:10.1088/1748-9326/ac94e6 · Cited by 11
“The OpenCV (http://opencv.org/) implementation of Farnebäck (Pérez et al 2013) and TV-L1 (Zach et al 2007) are used in this analysis.”
MethodHCM: Online Action Detection With Hard Video Clip Mining
Liu, S., Cheng, J., Xia, Z., Xi, Z., Hou, Q., & Dong, Z. (2023) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2023.3313258 · Cited by 12
“On an Nvidia GTX 1080, we calculate the speed of computing optical flows using TV-L1 [60] in denseflow [50], extracting RGB and optical flow features by ResNet-50 [17], and detecting actions, which is measured in frames per second (FPS).”
MethodReparo: QoE-Aware Live Video Streaming in Low-Rate Networks by Intelligent Frame Recovery
Wang, F., Li, Q., Shi, W., Gareth Tyson, A., Jiang, Y., Ma, L., Zhang, P., Lan, Y., & Li, Z. (2023) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3581783.3613441 · Cited by 11
“Figure 8 displays the performance of VFI-DNN [27], the traditional video frame interpolation method [60], and the approach of directly replacing the missing intermediate frame with the preceding frame.”
MethodTowards efficient video-based action recognition: context-aware memory attention network
Koh, T. C., Yeo, C. K., Jing, X., & Sivadas, S. (2023) · SN Applied Sciences · doi:10.1007/s42452-023-05568-5 · Cited by 11
Video Anomaly Detection via Spatio-Temporal Pseudo-Anomaly Generation : A Unified Approach
Rai, A. K., Krishna, T., Hu, F., Drîmbărean, A., McGuinness, K., Smeaton, A. F., & O'Connor, N. E. (2024) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw63382.2024.00393 · Cited by 17
“More specifically, given a normal video v , its frame x t , and its corresponding segmentation mask m t and another consecutive frame x ( t + 1 ) , we compute the optical flow ϕ ( x t , x ( t + 1 ) ) using the TVL1 alogrithm [95].”
High-Dimension Multilabel Problems: Convex or Nonconvex Relaxation?
Papadakis, N., Yıldızoğlu, R., Aujol, J., & Caselles, V. (2013) · SIAM Journal on Imaging Sciences · doi:10.1137/120900307 · Cited by 11
Raudies, F., & Heiko Neumann, B. (2012) · PLoS ONE · doi:10.1371/journal.pone.0053456 · Cited by 11
“[51], [52], [53], [54]).”
BackgroundChițiboi, T., Muckley, M. J., Dane, B. R., Huang, C., Feng, L., & Chandarana, H. (2019) · Journal of Magnetic Resonance Imaging · doi:10.1002/jmri.26714 · Cited by 11
Srivastava, A., Dutta, O., Gupta, J., Agarwal, S., & Prathosh, A. P. (2021) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv48630.2021.00279 · Cited by 11
“Training with HMDB51 and UCF101 : For training our models with HMDB51 and UCF101, we use both RGB and TV-L1 optical flow [29].”
MethodHossain, M. I., Siddique, A., Hossain, M. A., Hossain, M. D., & Huh, E. (2020) · IEEE Access · doi:10.1109/access.2020.3037529 · Cited by 11
“We select popular TV-L1 algorithm [45] to extract flow frames from RGB frames using default parameters settings in OpenCV.”
MethodAdversarial Patch Defense for Optical Flow Networks in Video Action Recognition
Anand, A., Gokul, H., Srinivasan, H., Vijay, P., & Vijayaraghavan, V. (2020) · International Conference on Machine Learning and Applications · doi:10.1109/icmla51294.2020.00202 · Cited by 12
Weakly Supervised Temporal Action Detection With Temporal Dependency Learning
Li, B., Liu, R., Chen, T., & Zhu, Y. (2021) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2021.3125701 · Cited by 13
“The optical flows are extracted by TV-L1 algorithm [79], and the feature dimension is 1024 for both streams.”
MethodDing, L., Kang, T. D., Kuriyan, A. E., Ramchandran, R. S., Wykoff, C. C., & Gaurav Sharma, A. (2022) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2022.3196458 · Cited by 11
“For methods that take the intensity image as input, we consider the non-parametric deformable registration (diffeomorphic demons [47]), the optic flow using the TV-L1 solver algorithm [48], and the intensity-based registration that maxi-”
MethodTemporal Dropout for Weakly Supervised Action Localization
Xie, C., Zhuang, Z., Zhao, S., & Liang, S. (2022) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3567827 · Cited by 11
Yang, Z., Simon, R. M., & Linte, C. A. (2023) · Journal of medical imaging · doi:10.1117/1.jmi.10.4.045001 · Cited by 11
“Global methods can provide relatively accurate results; however, they can be time-consuming and may require good initialization, especially for large displacements.(13,14) Learning-based stereo matching methods could be classified into fully-supervised methods(15,16) and self-supervised methods.”
MethodGilles, J., Alvarez, F., Ferrante, N., Fortman, M., Tahir, L., Tarter, A., & Seeger, A. (2018) · Image and Vision Computing · doi:10.1016/j.imavis.2018.03.006 · Cited by 11
“Optical flow based techniques [4, 5, 6, 7, 8, 9] aim at estimating the velocity vector field corresponding to movement present in the sequence, i.e. each pixel contains a vector representing the displacement of that pixel from the current frame to the next one.”
BackgroundAction Recognition with Multiple Relative Descriptors of Trajectories
Liao, Z., Hu, H., & Liu, Y. (2019) · Neural Processing Letters · doi:10.1007/s11063-019-10091-z · Cited by 11
“TV-L1 [37] is applied to compute optical flow, then we discretize the values of optical flow fields into integers and set their range as 0–255 just like images.”
MethodSALAD: Self-Assessment Learning for Action Detection
Vaudaux-Ruth, G., Chan-Hon-Tong, A., & Achard, C. (2021) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv48630.2021.00131 · Cited by 11
“The videos are previously sampled at 25 frames per second, and TV-L1 optical flow algorithm [53] is applied.”
MethodRobust optical flow estimation via edge preserving filtering
Rao, S., & Wang, H. (2021) · Signal Processing Image Communication · doi:10.1016/j.image.2021.116309 · Cited by 11
Interaction-Aware Spatio-Temporal Pyramid Attention Networks for Action Classification
Hu, W., Liu, H., Du, Y., Yuan, C., Li, B., & Maybank, S. J. (2021) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2021.3100277 · Cited by 11
“RGB images and optical flow gray images [56] are input to train, respectively, two CNNs with the same structure.”
An evaluation of robust cost functions for RGB direct mapping
Concha, A., & Civera, J. (2015) · European Conference on Mobile Robots · doi:10.1109/ecmr.2015.7324174 · Cited by 11
“The algorithm we use in this paper, using the robust L(1) norm for the photometric term, was first proposed in [11] and had the key advantage of resulting in a parallel algorithm suitable for implementation in modern Graphics Processing Units.”
MethodNRSfM-Flow: Recovering Non-Rigid Scene Flow from Monocular Image Sequences
Golyanik, V., Mathur, A., & Stricker, D. (2016) · British Machine Vision Conference (BMVC) · doi:10.5244/c.30.41 · Cited by 11
“Figure 3: Experimental results on the barn owl sequence [9]: (a) input frame number 51; (b) result of the MSF recovery by NRSfM-Flow (between frames 51 and 52), the colour scheme is the same as in [29]; (c) geometry with an overlayed scene flow for the frame 51; (d) result of the optical flow between frames 51 and 52…”
MethodAction Recognition with 3D ConvNet-GRU Architecture
Yao, G., Liu, X., & Leí, T. (2018) · International Conference on Robotics, Control and Automation (ICRCA) · doi:10.1145/3265639.3265672 · Cited by 12
SF-Net: Single-Frame Supervision for Temporal Action LocalizationPreprint
Ma, F., Zhu, L., Yang, Y., Zha, S., Kundu, G., Feiszli, M., & Shou, Z. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2003.06845 · Cited by 11
PipeMamba: State Space Model for Efficient Video-Based Sewer Defect Classification
Zhao, C., Hu, C., Shao, H., & Liu, J. (2025) · IEEE Transactions on Artificial Intelligence · doi:10.1109/tai.2025.3597580 · Cited by 10
Tube ConvNets: Better exploiting motion for action recognition
Li, Z., Wang, W., Li, N., & Wang, J. (2016) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2016.7532921 · Cited by 10
“In our experiments, we extract optical flow fields with TV -L(1) algorithm [22] and use the code released by [1] to get improved dense trajectories.”
MethodRegional foremost matching for internet scene images
Shen, X., Tao, X., Zhou, C., Gao, H., & Jia, J. (2016) · ACM Transactions on Graphics · doi:10.1145/2980179.2980249 · Cited by 10
“Regularization terms are usually in the form of robust functions, such as L1 norm and Charbonnier function [Brox et al. 2004; Black and Anandan 1996; Wedel et al. 2008; Zach et al. 2007], to achieve piece-wise smooth flow fields.”
BackgroundHigh-Quality Correspondence and Segmentation Estimation for Dual-Lens Smart-Phone Portraits
Shen, X.-Y., Gao, H., Tao, X., Zhou, C., & Jia, J.-Y. (2017) · IEEE International Conference on Computer Vision · doi:10.1109/iccv.2017.353 · Cited by 13
“We use 500 uniformly sampled values from [−50, 50] × [50, 50] to get all ws.”
BackgroundMethodA Decoupled Learning Scheme for Real-World Burst Denoising from Raw Images
Liang, Z., Guo, S. Z., Gu, H., Zhang, H., & Lei Zhang, A. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58595-2_10 · Cited by 10
“M-UNet and M-RIDNet first denoise each frame in the noisy sequence, and then apply optical flow alignment [38] to fuse the denoised frames by average fusion, resulting in the finally denoised sequences.”
MethodDadashzadeh, A., Whone, A. L., Rolinski, M., & Mirmehdi, M. (2021) · International Conference on Pattern Recognition Applications and Methods · doi:10.5220/0010309200890097 · Cited by 10
Augmented two stream network for robust action recognition adaptive to various action videos
Leng, C., Ding, Q., Wu, C., & Chen, A. (2021) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2021.103344 · Cited by 11
Needle in a Haystack: Spotting and recognising micro-expressions “in the wild”
Gan, Y. S., See, J., Khor, H.-Q., Liu, K., & Liong, S. (2022) · Neurocomputing · doi:10.1016/j.neucom.2022.06.101 · Cited by 10
YogaTube: A Video Benchmark for Yoga Action Recognition
Yadav, S. K., Singh, G., Verma, M., Tiwari, K., Pandey, H. M., Akbar, S., & Corcoran, P. (2022) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn55064.2022.9892122 · Cited by 10
A vision based motion estimation in underwater images
Kumar, P., Kumar, S., & Balasubramanian, R. (2015) · International Conference on Advances in Computing, Communications and Informatics · doi:10.1109/icacci.2015.7275771 · Cited by 9
A general form of illumination-invariant descriptors in variational optical flow estimation
Trink, D.-H., Blondel, W., & Daul, C. (2017) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2017.8296739 · Cited by 10
“Among the numerous formulations proposed in the literature for Ereg, regularization terms using the total variation [1], the non-local total variation [2–4], or the non-local total generalized variation [5] of the flow are known as the most effective ones.”
Huber-$$L_1$$-based non-isometric surface registration
Jiang, T., Yang, X., Zhang, J., Tian, F., Liu, S., Xiang, N., & Qian, K. (2019) · The Visual Computer · doi:10.1007/s00371-019-01670-1 · Cited by 10
Activity Normalization for Activity Detection in Surveillance Videos
Hosono, T., Sawada, K., Sun, Y., Hayase, K., & Shimamura, J. (2020) · Proceedings - International Conference on Image Processing · doi:10.1109/icip40778.2020.9190884 · Cited by 10
“Since the RGB and optical flow calculated using TV-L1 [20] are input to I3D, our method is applied to both the RGB and optical flow proposals.”
MethodMotion Estimation for Spike Camera Data Sequence via Spike Interval Analysis
Zhao, J. H., Xiong, R., Zhao, R. Q., Wang, J., Ma, S., & Huang, T. (2020) · IEEE International Conference on Visual Communications and Image Processing (VCIP) · doi:10.1109/vcip49819.2020.9301840 · Cited by 10
Heemeyer, F., Choudhary, A., & Desai, J. P. (2020) · International Symposium on Medical Robotics · doi:10.1109/ismr48331.2020.9312944 · Cited by 10
“minimization [27], which is based on brightness constancy and spatial smoothness.”
BackgroundA deep learning method for video‐based action recognition
Zhang, G., Rao, Y., Wang, C., Zhou, W., & Ji, X. (2021) · IET Image Processing · doi:10.1049/ipr2.12303 · Cited by 11
Optical flow algorithms optimized for speed, energy and accuracy on embedded GPUs
Romera, T., Petreto, A., Lemaitre, F., Bouyer, M., Meunier, Q., Lacassagne, L., & Etiemble, D. (2023) · Journal of Real-Time Image Processing · doi:10.1007/s11554-023-01288-6 · Cited by 10
Transfer learning for the classification of video-recorded crowd movements
Bendali-Braham, M., Weber, J., Forestier, G., Idoumghar, L., & Müller, P.-A. (2019) · International symposium on image and signal processing and analysis/ISPA ... · doi:10.1109/ispa.2019.8868704 · Cited by 11
“We extracted the optical flow of each video clip using the TV-L1 algorithm [30].”
Fast instantaneous center of rotation estimation algorithm for a skied-steered robot
Kniaz, V. V. (2015) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2184834 · Cited by 10
“In recent years constantly increasing processor performance and usage of GPU parallel computing made it possible to realize optical flow algorithms for VGA video with real-time performance on consumer grade desktop computers [9].”
BackgroundA Methodology for Evaluating Illumination Artifact Removal for Corresponding Images
Vaudrey, T., Wedel, A., & Klette, R. (2009) · Lecture notes in computer science · doi:10.1007/978-3-642-03767-2_135 · Cited by 10
“To highlight the effect of using residual images, we used a high ranking (see [13]) optical flow technique called TV-L(1) optical flow [21].”
MethodCounting people from above: Airborne video based crowd analysisPreprint
Perko, R., Schnabel, T., Fritz, G., Almér, A., & Paletta, L. (2013) · arXiv (Cornell University) · doi:10.48550/arxiv.1304.6213 · Cited by 10
“The motion is estimated based on the optical flow in image geometry [15] where we used the implementation at4.”
MethodApplying information theory to efficient SLAM
Chli, M. (2009) · Spiral (Imperial College London) · doi:10.25560/5634 · Cited by 10
Hybrid Learning of Optical Flow and Next Frame Prediction to Boost Optical Flow in the WildPreprint
Sedaghat, N., Zolfaghari, M., & Brox, T. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1612.03777 · Cited by 10
“rm that this sequential learning is not sufficient. The multitasking scheme is necessary to make good use of the auxiliary task on the real data. We report also number of two variational methods, TVL1 [33] and EpicFlow [22]. They provide a higher accu6 FlowNet 1:9 1:5 1:3 1:1 4:1 EpicFlow Figure 6: Qualitative comparis…”
MethodFlow-Distilled IP Two-Stream Networks for Compressed Video Action RecognitionPreprint
Huang, S., Lin, X., Karaman, S., & Chang, S. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1912.04462 · Cited by 10
“Most state-of-the-art works keep RGB frame sequences as the appearance modality, and use optical flow (OF) [37] calculated from consecutive RGB frames as the motion modality.”
InfluentialVehicle Tracking and Motion Estimation Based on Stereo Vision Sequences
Barth, A. (2013) · bonndoc (University of Bonn) · Cited by 10
Deep-learning optical flow for measuring velocity fields from experimental data
Tran, P. N., Ray, S., Lemma, L., Li, Y., Sweeney, R., Baskaran, A., Dogic, Z., Hong, P., & Hagan, M. F. (2024) · Soft Matter · doi:10.1039/d4sm00483c · Cited by 9
Xu, G., Yang, S., & Feng, Z. (2025) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2025.3548654 · Cited by 9
“Therefore, we designed an SF module based on the TVL1 algorithm [41].”
Learning correlations for human action recognition in videos
Yi, Y., Wang, H., & Zhang, B. (2017) · Multimedia Tools and Applications · doi:10.1007/s11042-017-4416-4 · Cited by 9
“Meanwhile, the temporal ConvNet, which stacks optical flow [45] fields, captures the motion information in videos.”
BackgroundWan, P., Chen, F., Shao, W., Liu, C., Zhang, Y., Wen, B., Kong, W., & Zhang, D. (2020) · IEEE Transactions on Ultrasonics Ferroelectrics and Frequency Control · doi:10.1109/tuffc.2020.3033984 · Cited by 9
Eating detection with a head-mounted video camera
Bi, S., & Kotz, D. F. (2022) · IEEE International Conference on Healthcare Informatics (ICHI) · doi:10.1109/ichi54592.2022.00021 · Cited by 9
“Note that this is estimate is only a rough upper bound on GPU power consumption.”
InfluentialA Review of Automated Micro-expression Analysis
Min, K. S., Zulkifley, M. A., & Kamari, N. A. M. (2022) · Jurnal Kejuruteraan · doi:10.17576/jkukm-2022-34(5)-02 · Cited by 10
MPI-Flow: Learning Realistic Optical Flow with Multiplane Images
Liang, Y., Liu, J., Zhang, D., & Fu, Y. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01274 · Cited by 10
“In recent years, with the rapid development of neural networks, learning-based methods [44, 45] have demonstrated significant advances compared to traditional model-based algorithms [3, 49, 50].”
He, W., Zhang, H., Wang, R., Guo, J., Zhang, Y., & Wu, K. (2024) · Measurement · doi:10.1016/j.measurement.2024.115255 · Cited by 9
Time of flight motion compensation revisited
Gottfried, J.-M., Nair, R., Meister, S., Garbe, C. S., & Kondermann, D. (2014) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2014.7026184 · Cited by 9
“Recently, many stateof-the-art methods [11, 12, 13, 14] have been implemented using Cuda and OpenCL in the OpenCV library.”
MethodTrinocular optical flow estimation for intelligent vehicle applications
Kitt, B., & Lategahn, H. (2012) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc.2012.6338661 · Cited by 9
Multi-step flow fusion: towards accurate and dense correspondences in long video shots
Crivelli, T., Conze, P.-H., Robert, P., Fradet, M., & Pérez, P. (2012) · British Machine Vision Conference · doi:10.5244/c.26.107 · Cited by 13
“We draw the following remarks from the plot: a) the three methods based on optical-flow integration (TV-L1, LDOF, STEP1) are the worst performing, supporting our claim that high precision in instantaneous motion estimation does not guarantee long-term tracking accuracy; b) Multi-step methods start to perform better af…”
BackgroundMethodInfluentialActivities of daily living recognition using optimal trajectories from motion boundaries
Avgerinakis, K., Briassouli, A., & Kompatsiaris, I. Y. (2015) · Journal of Ambient Intelligence and Smart Environments · doi:10.3233/ais-150347 · Cited by 9
“The second method [53] is more precise, but has a higher computational cost (0.”
InfluentialA fractional order total variation model for the estimation of optical flow
Kumar, P., Kumar, S., & Balasubramanian, R. (2015) · National Conference on Computer Vision, Pattern Recognition, Image Processing and Graphics (NCVPRIPG) · doi:10.1109/ncvpripg.2015.7490034 · Cited by 9
“Variational models proposed in [7], [8], used total variation regularization (TV) and L1 terms.”
MethodSynthesizing the motion of the vocal folds using optical flow based techniques
Andrade-Miranda, G. X., Bernardoni, N. H., & Godino-Llorente, J. I. (2017) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2017.01.002 · Cited by 9
Minimum Delay Moving Object Detection
Lao, D., & Sundaramoorthi, G. (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.2017.511 · Cited by 9
“, affine) of flow [14] or dense optical flow [3, 4, 39, 10, 25].”
Background
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
