Optical flow evaluation results Statistics:     Average   SD   R0.5   R1.0   R2.0   A90   A95   A99  
Error type:   endpoint   angle   interpolation   normalized interpolation  
Show images: below table   above table        
R1.0
normalized interpolation
error
avg. Mequon
(Hidden texture)
im0   GT   im1
Schefflera
(Hidden texture)
im0   GT   im1
Urban
(Synthetic)
im0   GT   im1
Teddy
(Stereo)
im0   GT   im1
Backyard
(High-speed camera)
im0   GT   im1
Basketball
(High-speed camera)
im0   GT   im1
Dumptruck
(High-speed camera)
im0   GT   im1
Evergreen
(High-speed camera)
im0   GT   im1
rank all disc untext all disc untext all disc untext all disc untext all disc untext all disc untext all disc untext all disc untext
MDP-Flow2 [68]14.6 6.71 3 9.78 4 8.39 10 6.36 10 11.5 13 6.23 12 7.12 6 9.73 8 5.42 2 21.2 6 18.8 10 41.2 12 30.1 9 26.5 6 45.6 56 27.7 21 19.5 31 36.0 23 6.48 3 13.8 7 10.3 7 8.19 31 16.4 23 7.11 37
PMMST [114]16.3 6.84 11 9.80 5 8.50 20 6.74 28 11.7 15 6.36 29 7.10 5 9.45 4 5.41 1 21.1 2 18.6 4 41.1 4 30.2 14 26.5 6 45.7 71 27.3 3 18.1 7 36.0 23 6.51 5 13.9 12 10.3 7 8.22 35 16.5 32 7.15 48
NNF-Local [87]18.0 6.74 5 10.1 8 8.30 3 5.97 2 10.3 3 6.14 3 7.09 4 9.63 7 5.44 3 21.7 34 20.5 69 41.2 12 30.3 24 26.6 12 45.4 29 27.9 40 20.6 70 36.0 23 6.51 5 13.8 7 10.3 7 8.04 16 16.1 11 7.10 35
PH-Flow [101]18.2 7.05 26 10.9 20 8.50 20 6.10 4 10.6 6 6.14 3 7.18 7 9.83 9 5.55 6 21.1 2 18.5 3 41.1 4 30.1 9 26.6 12 45.2 19 27.9 40 21.7 100 35.7 11 6.60 20 14.3 25 10.3 7 8.11 21 16.3 19 7.14 45
CombBMOF [113]18.9 6.93 15 9.87 7 8.43 13 6.33 9 11.4 10 6.22 11 7.60 48 10.3 16 6.25 86 21.5 21 19.4 26 41.2 12 30.2 14 26.6 12 45.3 25 27.7 21 19.2 23 36.1 36 6.57 13 14.1 15 10.3 7 7.67 4 15.4 6 6.82 3
NN-field [71]22.5 6.88 12 10.8 16 8.48 17 5.99 3 10.3 3 6.13 2 7.65 57 9.54 5 5.81 34 21.9 51 21.0 92 41.4 26 30.2 14 26.6 12 45.4 29 27.8 31 20.0 47 36.0 23 6.48 3 13.7 6 10.3 7 8.02 14 16.1 11 7.04 24
IROF++ [58]28.7 7.07 29 11.3 31 8.50 20 6.64 23 12.0 20 6.19 7 7.54 39 10.6 30 5.84 39 21.2 6 18.6 4 41.5 30 30.2 14 26.9 25 45.0 12 27.6 17 18.8 14 36.2 45 6.69 39 14.7 39 10.5 68 8.17 28 16.4 23 7.33 86
Layers++ [37]29.1 7.17 41 11.1 26 8.79 65 6.14 7 10.3 3 6.41 34 7.34 19 10.3 16 5.69 20 21.3 11 19.0 16 41.3 18 30.5 43 27.1 48 45.4 29 28.1 61 21.1 85 36.2 45 6.51 5 13.8 7 10.2 2 8.20 32 16.4 23 7.13 43
nLayers [57]30.5 7.15 38 10.4 12 8.81 70 6.44 13 11.4 10 6.42 35 7.23 10 9.30 2 5.65 14 21.4 16 19.1 19 41.5 30 30.7 80 27.4 91 45.6 56 28.0 51 20.8 74 36.2 45 6.54 11 13.6 4 10.3 7 8.07 17 16.3 19 6.91 8
Sparse-NonSparse [56]33.0 7.09 30 11.3 31 8.57 30 6.53 18 11.8 18 6.21 9 7.40 27 10.5 26 5.64 10 21.5 21 19.0 16 41.7 50 30.4 31 27.0 36 45.4 29 28.3 74 21.5 96 36.4 68 6.66 29 14.4 29 10.3 7 8.23 37 16.6 37 7.09 32
ProbFlowFields [128]33.2 7.03 22 11.8 57 8.66 44 6.41 11 11.7 15 6.31 22 7.18 7 10.3 16 5.58 7 21.7 34 19.6 33 41.8 59 30.7 80 27.1 48 46.0 120 27.9 40 20.5 61 36.2 45 6.51 5 13.8 7 10.3 7 7.80 6 15.6 7 7.14 45
2DHMM-SAS [92]33.7 7.27 55 12.0 65 8.59 33 7.82 70 14.2 60 6.36 29 7.25 11 10.5 26 5.74 26 21.4 16 18.7 7 41.4 26 30.3 24 26.9 25 45.2 19 27.9 40 20.3 54 36.0 23 6.64 27 14.4 29 10.3 7 8.41 55 17.0 53 7.08 29
AGIF+OF [85]34.6 7.11 33 11.3 31 8.46 14 6.68 27 12.2 24 6.27 16 7.38 24 10.1 12 5.71 24 21.2 6 18.6 4 41.1 4 30.8 92 27.6 115 45.4 29 28.3 74 22.2 118 36.0 23 6.67 31 14.0 14 10.3 7 8.34 47 17.0 53 6.91 8
NNF-EAC [103]34.6 7.35 66 11.1 26 8.79 65 6.92 39 12.5 32 6.29 20 7.52 38 10.1 12 5.76 29 21.8 41 19.5 31 42.8 114 30.2 14 26.6 12 45.4 29 27.5 9 18.9 16 36.0 23 6.58 15 14.2 20 10.4 37 8.32 44 16.8 41 7.19 58
FlowFields [110]34.8 7.02 21 11.5 41 8.54 28 6.66 26 12.5 32 6.44 39 7.45 31 11.5 58 5.64 10 21.9 51 20.5 69 41.8 59 30.6 60 27.0 36 45.5 45 27.7 21 20.2 49 36.0 23 6.59 17 14.3 25 10.4 37 8.00 13 16.2 16 7.08 29
CtxSyn [137]35.4 3.84 1 6.67 1 4.59 1 5.06 1 9.01 1 6.57 54 5.42 1 6.73 1 8.44 137 20.9 1 16.7 1 42.3 92 28.9 4 24.9 4 45.0 12 27.5 9 15.3 2 37.0 109 7.55 127 13.2 3 12.1 141 7.09 1 12.4 1 9.27 144
FlowFields+ [130]35.4 6.99 17 11.4 38 8.47 16 6.61 22 12.3 26 6.48 42 7.42 29 11.5 58 5.67 17 21.7 34 20.3 55 41.6 39 30.7 80 27.2 63 45.6 56 27.8 31 20.3 54 36.1 36 6.61 21 14.4 29 10.4 37 7.99 12 16.2 16 7.03 22
FMOF [94]35.7 7.36 68 12.0 65 8.73 54 6.42 12 11.3 9 6.30 21 7.63 53 10.4 21 6.02 70 21.8 41 19.9 40 41.2 12 30.5 43 27.0 36 45.4 29 28.0 51 20.5 61 36.1 36 6.51 5 13.8 7 10.2 2 8.30 42 16.7 40 7.12 39
S2F-IF [123]36.3 7.01 20 11.5 41 8.48 17 6.57 19 12.2 24 6.42 35 7.40 27 11.2 54 5.64 10 21.6 25 20.1 47 41.3 18 30.7 80 27.3 77 45.7 71 27.8 31 20.4 58 36.1 36 6.71 44 14.9 54 10.4 37 7.98 11 16.1 11 7.04 24
LSM [39]38.5 7.17 41 11.8 57 8.58 32 6.64 23 12.1 23 6.17 6 7.49 34 10.9 44 5.69 20 21.6 25 19.6 33 41.6 39 30.5 43 27.1 48 45.4 29 28.3 74 21.6 98 36.3 59 6.68 37 14.6 34 10.2 2 8.35 50 16.9 51 7.03 22
LME [70]38.7 6.72 4 9.86 6 8.36 6 6.97 41 12.4 29 7.40 95 7.51 36 11.8 67 5.70 23 21.3 11 19.2 22 41.3 18 31.0 121 27.6 115 46.6 134 27.8 31 20.5 61 36.0 23 6.45 2 13.6 4 10.2 2 8.08 19 16.3 19 7.12 39
FGIK [136]38.9 8.46 120 10.6 14 11.0 131 9.13 108 15.0 83 10.8 130 5.88 2 10.6 30 6.65 112 23.0 111 20.4 62 40.7 1 25.6 1 21.6 1 41.9 1 24.6 1 14.3 1 33.2 1 6.00 1 11.4 1 9.93 1 7.19 2 14.1 4 7.00 15
WLIF-Flow [93]40.0 6.99 17 11.0 23 8.48 17 6.76 29 12.4 29 6.39 32 7.38 24 10.3 16 5.68 19 21.4 16 18.8 10 41.9 72 30.4 31 26.9 25 45.9 109 28.8 112 21.9 107 36.9 105 6.56 12 13.9 12 10.3 7 8.34 47 16.8 41 7.15 48
TV-L1-MCT [64]40.2 7.50 87 12.5 96 8.79 65 7.19 48 13.4 47 6.37 31 7.28 14 10.6 30 5.80 33 21.4 16 18.8 10 41.3 18 30.5 43 27.1 48 45.1 16 27.9 40 18.6 10 36.6 85 6.72 48 15.0 63 10.4 37 7.92 9 15.9 9 7.20 62
ComponentFusion [96]40.4 6.91 13 10.8 16 8.55 29 6.49 17 12.0 20 6.10 1 7.49 34 11.2 54 5.72 25 21.3 11 19.4 26 41.2 12 30.6 60 27.2 63 45.8 88 27.8 31 19.6 34 36.2 45 6.92 84 16.4 106 10.4 37 8.43 59 17.0 53 7.16 52
COFM [59]41.3 7.04 23 10.7 15 8.70 48 6.60 20 11.9 19 6.35 26 7.26 12 9.93 10 5.63 9 21.2 6 18.8 10 41.0 2 30.4 31 27.3 77 44.9 11 27.7 21 22.6 122 35.1 3 6.86 76 14.7 39 11.2 125 8.67 90 17.2 72 7.78 124
MDP-Flow [26]42.3 6.83 9 10.8 16 8.50 20 6.65 25 12.4 29 6.51 48 7.46 32 10.6 30 5.88 48 22.1 71 20.6 74 41.7 50 30.4 31 26.8 20 45.6 56 28.2 70 21.9 107 36.3 59 6.69 39 14.8 48 10.4 37 8.15 24 16.6 37 7.10 35
HAST [109]43.6 6.97 16 10.2 9 8.69 46 6.46 14 11.6 14 6.26 15 7.72 67 11.1 50 5.97 62 21.1 2 18.7 7 41.1 4 30.5 43 27.5 100 44.8 9 28.2 70 22.8 128 35.5 6 6.76 58 15.2 72 10.4 37 8.81 97 18.0 107 6.96 13
LFNet_ROB [151]43.8 7.27 55 11.6 46 8.82 72 7.96 75 14.9 79 7.11 86 7.95 84 13.8 121 6.10 74 21.8 41 20.6 74 41.3 18 30.1 9 26.6 12 45.1 16 27.9 40 21.0 78 35.9 15 6.51 5 14.1 15 10.3 7 7.84 7 15.7 8 7.00 15
RNLOD-Flow [121]44.0 7.12 35 11.5 41 8.64 41 7.38 56 14.0 55 6.35 26 7.55 40 11.2 54 5.83 38 21.3 11 19.0 16 41.1 4 30.5 43 27.2 63 45.4 29 28.3 74 21.6 98 36.2 45 6.62 22 14.2 20 10.4 37 8.70 93 17.7 94 7.02 20
OFLAF [77]44.6 6.81 8 10.2 9 8.40 11 6.10 4 10.7 7 6.21 9 7.36 21 10.6 30 5.54 5 21.1 2 18.8 10 41.0 2 30.8 92 27.4 91 45.7 71 28.1 61 21.9 107 36.0 23 7.02 96 16.1 102 10.4 37 8.90 105 18.1 115 7.16 52
Ramp [62]47.0 7.31 62 12.1 69 8.78 62 6.60 20 12.0 20 6.27 16 7.36 21 10.4 21 5.65 14 21.3 11 18.9 15 41.4 26 30.5 43 27.1 48 45.4 29 28.7 106 22.3 120 36.6 85 6.73 54 14.9 54 10.3 7 8.55 75 17.3 77 7.26 73
Second-order prior [8]48.6 7.30 61 11.3 31 8.90 81 8.52 92 15.6 94 6.74 68 8.32 108 13.6 116 6.42 103 21.8 41 20.0 43 41.5 30 30.1 9 26.5 6 45.5 45 27.5 9 19.0 18 36.0 23 6.67 31 14.6 34 10.3 7 8.25 38 16.8 41 7.11 37
PGM-C [120]48.9 7.19 46 12.1 69 8.72 51 6.82 31 12.9 36 6.62 56 7.67 61 12.2 79 5.78 32 21.9 51 20.9 86 41.8 59 30.6 60 27.1 48 45.8 88 27.7 21 19.5 31 36.2 45 6.65 28 14.7 39 10.3 7 8.20 32 16.6 37 7.31 80
DeepFlow2 [108]49.9 7.28 57 11.3 31 8.88 78 7.68 65 14.4 68 6.94 81 7.58 46 12.3 83 5.88 48 21.9 51 20.2 51 41.7 50 30.5 43 26.8 20 45.9 109 27.5 9 18.0 6 36.4 68 6.67 31 14.6 34 10.4 37 8.18 29 16.4 23 7.31 80
Aniso. Huber-L1 [22]50.1 7.61 94 12.2 77 9.19 94 8.99 103 15.7 97 7.12 88 7.73 69 11.0 47 5.86 45 21.8 41 20.0 43 41.6 39 30.2 14 26.6 12 45.5 45 27.4 5 19.6 34 35.7 11 6.68 37 14.6 34 10.3 7 8.34 47 16.8 41 7.29 79
FC-2Layers-FF [74]50.2 7.22 49 11.9 63 8.70 48 6.10 4 10.2 2 6.47 41 7.31 17 10.5 26 5.64 10 21.4 16 19.1 19 41.6 39 30.7 80 27.5 100 45.6 56 28.6 100 22.7 125 36.4 68 6.77 60 15.0 63 10.3 7 8.57 79 17.2 72 7.20 62
Classic+NL [31]50.3 7.44 81 12.3 81 8.86 75 6.78 30 12.3 26 6.28 18 7.32 18 10.4 21 5.69 20 21.6 25 19.4 26 41.8 59 30.5 43 27.1 48 45.5 45 28.6 100 21.8 103 36.6 85 6.72 48 14.7 39 10.3 7 8.50 68 17.2 72 7.24 70
SRR-TVOF-NL [91]51.3 7.42 79 11.5 41 8.86 75 7.79 67 14.8 76 7.08 85 7.62 50 11.5 58 5.85 43 21.7 34 19.6 33 41.1 4 30.3 24 27.1 48 45.2 19 27.5 9 20.5 61 35.3 5 6.72 48 14.8 48 10.4 37 8.97 112 18.3 121 7.17 55
PWC-Net_ROB [148]51.5 7.22 49 12.8 106 8.51 26 7.26 52 14.0 55 6.49 45 7.78 74 12.7 96 5.95 58 21.6 25 20.3 55 41.6 39 30.8 92 27.5 100 45.6 56 28.2 70 20.2 49 36.5 77 6.58 15 14.2 20 10.4 37 8.02 14 16.3 19 6.87 6
FF++_ROB [146]51.8 7.00 19 11.4 38 8.50 20 7.08 43 13.3 45 6.62 56 7.67 61 11.9 70 5.93 55 21.9 51 20.6 74 41.5 30 30.8 92 27.4 91 45.7 71 28.4 86 20.5 61 36.9 105 6.67 31 14.6 34 10.4 37 8.08 19 16.4 23 7.09 32
DF-Auto [115]52.8 7.54 89 11.1 26 9.32 103 8.42 88 14.5 71 8.82 108 7.35 20 10.3 16 5.65 14 22.0 62 20.2 51 41.5 30 30.4 31 26.7 19 45.8 88 27.5 9 18.7 12 36.1 36 6.82 68 15.3 76 10.5 68 8.43 59 17.1 61 7.20 62
LiteFlowNet [143]52.8 7.24 53 12.4 91 8.60 36 7.17 45 13.7 49 6.52 49 7.70 65 13.1 105 5.84 39 22.5 97 22.5 126 41.9 72 30.4 31 27.0 36 45.2 19 27.8 31 21.3 91 35.5 6 6.90 83 15.6 87 10.4 37 7.86 8 16.0 10 6.80 2
FESL [72]53.0 7.36 68 11.7 51 8.65 43 6.82 31 12.6 34 6.33 24 7.51 36 10.7 39 5.89 51 21.6 25 19.6 33 41.3 18 30.9 115 27.5 100 45.7 71 28.4 86 22.1 116 36.2 45 6.70 42 14.8 48 10.2 2 8.59 80 17.4 84 7.08 29
CPM-Flow [116]53.1 7.21 47 12.2 77 8.71 50 6.83 34 12.9 36 6.65 60 7.61 49 11.7 64 5.88 48 22.2 79 21.4 107 41.8 59 30.6 60 27.1 48 45.8 88 27.9 40 19.1 19 36.6 85 6.67 31 14.7 39 10.3 7 8.16 26 16.5 32 7.34 89
Classic+CPF [83]53.5 7.22 49 11.6 46 8.52 27 6.90 38 12.6 34 6.28 18 7.37 23 10.6 30 5.76 29 21.2 6 18.7 7 41.1 4 31.1 126 27.9 124 45.5 45 28.7 106 23.1 133 36.3 59 6.92 84 15.3 76 10.3 7 8.75 95 17.9 103 6.99 14
IROF-TV [53]55.2 7.33 64 12.3 81 8.82 72 6.83 34 12.3 26 6.23 12 7.70 65 12.9 102 5.93 55 21.5 21 19.5 31 42.0 80 30.8 92 27.3 77 45.9 109 27.5 9 20.2 49 35.6 8 6.75 56 15.1 70 10.5 68 8.18 29 16.4 23 7.37 93
S2D-Matching [84]55.2 7.37 71 12.3 81 8.80 68 7.62 64 14.2 60 6.43 38 7.28 14 10.4 21 5.74 26 21.6 25 19.1 19 42.2 89 30.6 60 27.3 77 45.4 29 28.6 100 22.5 121 36.4 68 6.76 58 14.5 32 10.3 7 8.46 62 17.0 53 7.32 83
DeepFlow [86]55.5 7.21 47 11.0 23 8.88 78 7.79 67 14.3 63 7.33 94 7.64 55 12.6 92 5.95 58 22.1 71 20.1 47 42.0 80 30.6 60 26.8 20 46.1 125 28.0 51 17.9 5 37.2 115 6.57 13 14.1 15 10.4 37 8.07 17 16.2 16 7.32 83
EPPM w/o HM [88]56.0 6.77 7 10.4 12 8.32 4 7.00 42 13.4 47 6.16 5 8.19 101 13.6 116 6.26 87 21.7 34 20.3 55 41.5 30 30.5 43 27.2 63 45.5 45 28.5 94 21.8 103 36.5 77 6.84 70 15.6 87 10.6 94 8.41 55 17.1 61 6.95 12
Brox et al. [5]56.7 7.28 57 11.4 38 8.76 56 7.86 71 14.6 73 6.92 80 8.03 90 13.1 105 6.34 94 21.9 51 19.9 40 41.4 26 30.6 60 27.0 36 45.8 88 27.7 21 19.5 31 36.2 45 6.80 63 15.4 82 10.4 37 8.16 26 16.5 32 7.19 58
p-harmonic [29]56.8 7.04 23 11.3 31 8.62 39 8.81 97 15.8 100 6.98 82 7.76 72 13.1 105 6.18 83 22.4 91 20.7 79 41.9 72 30.5 43 27.0 36 45.5 45 27.8 31 19.2 23 36.4 68 6.71 44 15.1 70 10.3 7 8.29 41 16.8 41 7.12 39
Efficient-NL [60]57.2 7.28 57 11.6 46 8.61 38 7.24 51 13.3 45 6.35 26 8.21 103 10.8 41 6.39 100 21.7 34 19.6 33 41.2 12 30.4 31 27.0 36 45.3 25 28.3 74 22.8 128 35.6 8 6.86 76 15.6 87 10.4 37 9.10 119 18.3 121 7.14 45
ProFlow_ROB [147]57.5 7.15 38 11.3 31 8.77 59 7.31 53 14.3 63 6.72 66 7.57 45 11.5 58 5.76 29 22.0 62 21.2 103 42.2 89 30.8 92 27.4 91 45.6 56 27.6 17 18.7 12 36.1 36 6.81 66 15.3 76 10.3 7 8.55 75 17.3 77 7.31 80
SepConv-v1 [127]58.1 4.07 2 8.88 2 4.61 2 6.87 36 13.0 40 7.47 96 6.42 3 9.58 6 9.25 142 23.4 122 20.0 43 44.0 127 30.2 14 26.3 5 45.7 71 27.9 40 16.5 4 37.4 121 7.61 130 15.6 87 12.9 147 7.71 5 13.8 3 9.78 146
JOF [141]58.1 7.63 96 12.3 81 9.19 94 6.48 15 11.4 10 6.48 42 7.27 13 10.1 12 5.67 17 21.8 41 19.2 22 42.6 110 30.8 92 27.3 77 45.8 88 28.7 106 22.2 118 36.6 85 6.59 17 14.1 15 10.3 7 8.55 75 17.1 61 7.46 100
SuperSlomo [132]60.1 6.74 5 9.03 3 8.40 11 9.03 106 13.1 42 12.7 141 8.09 97 10.5 26 9.15 140 22.7 103 18.4 2 43.7 124 28.3 2 24.4 2 44.1 3 28.5 94 15.3 2 38.7 137 7.11 101 12.9 2 12.9 147 7.43 3 13.2 2 9.86 147
EpicFlow [102]60.3 7.18 43 12.0 65 8.72 51 7.42 58 14.4 68 6.72 66 7.68 63 12.1 76 5.92 54 22.1 71 21.1 96 42.0 80 30.7 80 27.1 48 45.8 88 27.5 9 19.9 45 35.9 15 6.79 62 15.2 72 10.4 37 8.40 54 17.1 61 7.33 86
DPOF [18]61.4 7.58 93 13.2 116 9.07 84 6.27 8 11.0 8 6.54 51 8.10 98 10.6 30 6.27 89 22.0 62 20.5 69 41.9 72 30.2 14 26.8 20 45.4 29 28.0 51 21.2 87 35.8 14 6.84 70 15.0 63 10.7 104 8.62 84 17.4 84 7.26 73
PMF [73]61.4 6.83 9 10.3 11 8.37 8 6.96 40 13.1 42 6.19 7 7.86 78 13.1 105 6.03 71 21.5 21 19.4 26 41.3 18 31.0 121 27.7 119 45.8 88 28.7 106 20.5 61 37.2 115 6.80 63 15.0 63 10.5 68 8.87 101 18.2 118 7.00 15
ContFlow_ROB [150]62.2 7.62 95 12.2 77 9.15 91 8.55 93 15.4 88 8.82 108 8.15 100 13.6 116 6.17 82 22.1 71 21.3 105 41.7 50 30.4 31 26.9 25 45.5 45 27.9 40 21.0 78 35.9 15 6.59 17 14.3 25 10.4 37 8.22 35 16.8 41 7.07 28
ComplOF-FED-GPU [35]62.5 7.23 52 11.8 57 8.72 51 7.20 49 13.9 52 6.62 56 8.43 113 12.6 92 6.45 104 21.9 51 20.8 85 42.3 92 30.4 31 26.9 25 45.4 29 27.7 21 20.1 48 36.1 36 6.86 76 15.4 82 10.5 68 8.55 75 17.3 77 7.28 78
DMF_ROB [140]63.0 7.31 62 11.8 57 8.81 70 7.86 71 14.9 79 6.69 64 8.52 116 13.8 121 6.40 102 22.1 71 20.7 79 41.5 30 30.5 43 26.9 25 46.0 120 27.4 5 18.9 16 36.1 36 6.95 91 14.3 25 11.2 125 8.13 22 16.4 23 7.19 58
Sparse Occlusion [54]63.6 7.37 71 12.3 81 8.87 77 8.04 78 15.3 87 6.48 42 7.58 46 10.8 41 5.87 46 22.0 62 20.4 62 41.5 30 30.6 60 27.2 63 45.5 45 28.3 74 21.8 103 36.4 68 6.80 63 15.3 76 10.3 7 8.74 94 17.7 94 7.18 57
TC/T-Flow [76]64.0 7.37 71 11.8 57 8.59 33 7.31 53 14.0 55 6.42 35 7.47 33 11.1 50 5.81 34 21.8 41 20.5 69 41.7 50 30.8 92 27.5 100 45.7 71 28.1 61 20.9 75 36.2 45 7.03 98 16.0 100 10.6 94 8.62 84 17.6 91 7.13 43
AggregFlow [97]66.0 7.71 102 12.6 99 9.11 86 7.50 61 13.9 52 7.06 83 7.19 9 9.98 11 5.53 4 21.9 51 20.4 62 41.6 39 30.8 92 27.3 77 46.1 125 29.0 116 19.7 41 37.9 129 6.75 56 14.7 39 10.5 68 8.32 44 16.8 41 7.40 97
CLG-TV [48]66.2 7.52 88 12.3 81 9.14 88 8.67 95 15.8 100 7.11 86 7.97 86 12.7 96 6.26 87 22.1 71 20.3 55 42.0 80 30.5 43 26.9 25 45.7 71 27.6 17 19.1 19 36.2 45 6.71 44 14.9 54 10.4 37 8.53 74 17.3 77 7.24 70
SuperFlow [81]66.3 7.43 80 11.5 41 9.30 100 8.55 93 14.8 76 9.15 113 7.91 82 12.0 73 6.31 91 22.1 71 19.9 40 42.0 80 30.7 80 27.2 63 45.9 109 27.3 3 18.4 9 35.9 15 6.86 76 15.8 94 10.6 94 8.15 24 16.5 32 7.16 52
RFlow [90]66.4 7.24 53 12.1 69 8.90 81 8.42 88 15.6 94 6.49 45 7.72 67 12.2 79 6.01 69 22.0 62 20.6 74 41.7 50 30.4 31 27.1 48 45.7 71 27.4 5 19.8 44 35.6 8 6.84 70 15.9 98 10.5 68 8.91 108 18.0 107 7.47 104
TCOF [69]67.4 7.36 68 12.1 69 8.68 45 9.41 113 16.6 114 7.17 90 7.38 24 10.7 39 5.61 8 21.8 41 20.4 62 41.8 59 30.4 31 27.0 36 45.6 56 28.1 61 21.8 103 35.9 15 6.85 73 15.7 93 10.4 37 9.30 129 19.0 134 7.61 118
SIOF [67]67.5 7.66 99 12.6 99 9.09 85 9.45 114 16.6 114 8.48 104 7.65 57 11.9 70 5.98 63 21.9 51 20.1 47 41.8 59 30.0 7 26.5 6 45.3 25 28.1 61 19.7 41 36.6 85 6.63 25 14.7 39 10.5 68 8.82 98 17.9 103 7.46 100
TC-Flow [46]67.6 7.18 43 11.8 57 8.78 62 7.46 60 14.6 73 6.77 72 7.86 78 12.6 92 5.89 51 21.8 41 20.3 55 41.9 72 30.7 80 27.4 91 45.7 71 28.3 74 21.0 78 36.6 85 6.73 54 14.8 48 10.5 68 8.51 70 17.3 77 7.24 70
3DFlow [135]68.2 7.09 30 11.7 51 8.46 14 6.89 37 13.0 40 6.39 32 8.03 90 10.6 30 5.98 63 21.6 25 19.3 24 41.9 72 30.8 92 27.1 48 47.6 141 29.0 116 23.9 140 36.4 68 7.05 99 16.2 103 10.5 68 8.83 100 18.0 107 7.15 48
IAOF [50]68.3 8.70 125 12.9 109 10.3 123 12.4 138 19.2 144 9.77 125 7.74 70 12.0 73 6.21 84 22.8 105 20.2 51 42.0 80 30.2 14 26.5 6 45.5 45 27.7 21 19.6 34 36.1 36 6.67 31 15.0 63 10.3 7 8.41 55 17.1 61 7.12 39
OAR-Flow [125]69.2 7.45 83 11.7 51 8.98 83 7.57 63 14.4 68 6.91 79 7.62 50 12.4 86 5.82 36 21.6 25 20.3 55 41.6 39 30.9 115 27.5 100 45.8 88 28.0 51 20.5 61 36.4 68 6.97 93 15.6 87 10.5 68 8.46 62 17.1 61 7.34 89
ALD-Flow [66]71.2 7.54 89 12.1 69 9.14 88 7.43 59 14.3 63 6.85 76 7.66 60 12.5 89 5.87 46 21.8 41 20.4 62 42.3 92 30.8 92 27.4 91 45.9 109 28.1 61 19.9 45 36.6 85 6.62 22 14.2 20 10.5 68 8.68 91 17.5 90 7.46 100
SVFilterOh [111]72.5 7.18 43 10.9 20 8.76 56 6.48 15 11.7 15 6.45 40 7.62 50 10.2 15 5.99 66 21.7 34 19.4 26 42.5 104 31.3 131 28.0 130 46.6 134 28.6 100 22.0 112 36.5 77 6.92 84 14.1 15 11.4 131 8.97 112 17.8 100 8.09 130
OFH [38]72.6 7.39 76 12.1 69 8.88 78 8.07 79 15.0 83 6.66 62 8.03 90 13.8 121 5.96 61 21.9 51 21.1 96 42.1 86 30.5 43 27.3 77 45.4 29 27.8 31 20.4 58 36.2 45 7.11 101 16.4 106 10.5 68 8.61 83 17.6 91 7.19 58
MLDP_OF [89]74.0 7.10 32 11.2 30 8.64 41 7.33 55 13.7 49 6.31 22 7.44 30 10.9 44 5.75 28 22.0 62 19.8 39 42.3 92 30.6 60 27.3 77 46.2 131 31.0 143 22.6 122 40.0 143 6.93 87 15.2 72 11.0 119 8.65 88 17.4 84 7.79 126
ResPWCR_ROB [145]74.1 7.34 65 12.4 91 8.76 56 7.92 74 15.1 85 7.28 92 8.37 110 13.4 113 6.22 85 22.7 103 22.2 121 43.1 120 29.7 5 26.5 6 44.6 6 32.9 147 21.5 96 43.1 147 6.66 29 14.9 54 10.3 7 8.50 68 17.4 84 7.00 15
CostFilter [40]74.7 6.91 13 11.1 26 8.37 8 6.82 31 12.9 36 6.25 14 7.99 87 13.9 124 6.10 74 21.9 51 20.6 74 41.7 50 31.1 126 27.9 124 45.9 109 29.8 130 20.3 54 39.1 139 6.94 89 15.8 94 10.6 94 8.82 98 18.1 115 7.09 32
Fusion [6]75.1 7.13 37 12.3 81 8.60 36 7.18 47 13.1 42 6.56 52 7.63 53 10.9 44 6.13 79 22.5 97 21.1 96 41.5 30 30.7 80 28.2 134 44.3 4 28.1 61 23.8 138 35.2 4 7.22 111 17.9 121 10.6 94 9.64 137 19.9 141 7.32 83
Modified CLG [34]75.9 7.63 96 11.6 46 9.65 106 10.7 126 17.2 123 10.7 129 8.25 105 14.3 129 6.60 110 22.4 91 21.1 96 41.8 59 30.6 60 26.9 25 45.8 88 27.7 21 19.2 23 36.3 59 6.69 39 14.9 54 10.4 37 8.41 55 17.0 53 7.35 92
F-TV-L1 [15]77.0 8.24 114 13.1 113 9.92 115 9.28 109 16.3 109 7.48 97 8.00 88 13.2 111 6.35 96 22.3 85 20.9 86 42.3 92 29.9 6 26.9 25 44.8 9 27.9 40 19.4 28 36.5 77 6.87 80 15.4 82 10.5 68 8.46 62 16.8 41 7.58 114
IIOF-NLDP [131]77.0 7.04 23 10.9 20 8.36 6 7.81 69 14.8 76 6.64 59 8.07 96 11.0 47 6.12 76 22.3 85 20.0 43 42.8 114 30.4 31 27.0 36 45.9 109 29.1 122 23.2 134 36.5 77 8.40 147 24.6 148 11.3 128 8.78 96 17.8 100 6.86 5
EPMNet [133]77.9 9.02 128 14.8 135 10.2 121 8.29 84 14.1 59 8.78 107 8.03 90 12.5 89 6.15 81 22.8 105 23.9 140 41.7 50 30.9 115 27.5 100 45.8 88 28.0 51 21.4 93 35.9 15 6.72 48 14.9 54 10.4 37 8.21 34 16.8 41 6.83 4
FlowNet2 [122]77.9 9.30 130 14.6 132 10.5 126 8.42 88 14.6 73 9.24 117 8.03 90 12.5 89 6.14 80 22.2 79 21.9 117 41.8 59 30.9 115 27.5 100 45.8 88 28.0 51 20.5 61 36.0 23 6.72 48 14.9 54 10.4 37 8.31 43 16.9 51 7.01 19
AugFNG_ROB [144]78.5 8.05 108 12.5 96 9.84 114 8.99 103 15.9 105 9.32 120 8.37 110 15.6 135 6.30 90 22.5 97 22.3 122 41.9 72 31.2 129 28.0 130 45.6 56 27.8 31 20.2 49 35.9 15 6.83 69 15.0 63 10.4 37 7.96 10 16.4 23 6.68 1
Complementary OF [21]78.7 7.11 33 12.1 69 8.50 20 7.17 45 14.0 55 6.58 55 8.76 124 12.0 73 6.55 107 22.3 85 21.4 107 42.6 110 30.6 60 27.5 100 45.2 19 28.1 61 20.9 75 36.4 68 7.15 105 16.7 110 10.5 68 9.09 117 18.7 128 7.38 94
SimpleFlow [49]79.2 7.37 71 12.4 91 8.74 55 7.88 73 14.3 63 6.50 47 8.59 119 11.5 58 6.51 105 21.6 25 19.3 24 41.8 59 30.6 60 27.3 77 45.5 45 28.5 94 22.9 131 36.2 45 7.66 133 20.5 141 10.8 114 8.89 104 18.2 118 7.15 48
LDOF [28]79.9 8.08 109 12.3 81 9.79 112 8.94 101 14.9 79 9.18 115 8.23 104 13.5 115 6.52 106 22.3 85 21.1 96 42.4 102 30.6 60 27.0 36 45.8 88 27.9 40 18.8 14 36.6 85 6.77 60 15.3 76 10.4 37 8.44 61 17.1 61 7.38 94
TF+OM [100]80.2 7.41 77 12.1 69 9.19 94 7.21 50 12.9 36 7.83 99 7.55 40 12.3 83 5.82 36 22.2 79 21.0 92 41.9 72 30.8 92 27.5 100 46.0 120 28.3 74 20.5 61 36.8 100 6.97 93 16.3 104 10.5 68 8.65 88 17.3 77 7.75 122
ROF-ND [107]80.8 7.46 84 11.0 23 8.77 59 7.96 75 15.4 88 6.76 71 7.55 40 11.0 47 5.85 43 23.3 119 23.5 138 41.6 39 30.6 60 27.1 48 45.8 88 28.3 74 22.7 125 35.9 15 7.52 125 17.5 116 11.4 131 9.25 127 18.7 128 7.27 75
Local-TV-L1 [65]81.7 8.46 120 12.6 99 10.4 124 9.68 116 16.0 107 8.93 111 7.56 44 11.2 54 5.84 39 23.1 114 20.4 62 46.0 140 30.6 60 27.1 48 45.9 109 30.1 135 19.1 19 39.9 142 6.72 48 14.9 54 10.5 68 8.13 22 16.1 11 7.58 114
TriFlow [95]81.7 7.77 104 13.7 125 9.28 98 8.98 102 15.7 97 9.30 119 7.65 57 12.4 86 5.84 39 22.0 62 20.9 86 41.1 4 30.9 115 27.7 119 45.7 71 28.4 86 21.3 91 36.3 59 6.85 73 15.5 86 10.4 37 8.69 92 17.4 84 7.23 69
Classic++ [32]82.5 7.49 86 12.5 96 9.11 86 8.07 79 15.2 86 6.67 63 7.89 81 12.6 92 6.04 72 22.3 85 20.7 79 42.2 89 30.6 60 27.2 63 45.7 71 29.0 116 21.0 78 37.6 124 6.81 66 15.2 72 10.5 68 8.62 84 17.4 84 7.46 100
Occlusion-TV-L1 [63]82.8 7.44 81 12.3 81 9.14 88 8.91 100 16.5 112 6.85 76 7.83 76 12.8 99 6.32 92 22.6 102 21.5 112 42.5 104 30.5 43 26.9 25 45.8 88 28.4 86 19.6 34 37.1 111 7.15 105 14.8 48 10.7 104 8.51 70 17.1 61 7.34 89
2D-CLG [1]84.1 8.44 118 12.3 81 10.6 128 11.9 134 18.0 133 12.3 139 8.94 127 13.9 124 7.33 128 23.1 114 21.2 103 41.3 18 30.5 43 26.9 25 45.8 88 27.6 17 19.2 23 36.2 45 7.14 103 17.2 114 10.5 68 8.37 52 16.5 32 7.20 62
Nguyen [33]84.2 9.74 135 12.6 99 12.4 139 12.3 136 18.6 139 11.1 131 8.27 107 14.8 131 6.69 113 23.4 122 21.7 114 41.8 59 30.3 24 26.8 20 45.3 25 27.4 5 19.6 34 35.7 11 7.24 113 18.3 124 10.5 68 8.37 52 17.0 53 7.22 68
FlowNetS+ft+v [112]84.5 7.81 106 11.7 51 9.63 105 9.77 118 16.8 116 9.16 114 8.06 95 13.4 113 6.36 97 22.1 71 20.7 79 42.1 86 30.8 92 27.4 91 45.8 88 27.7 21 19.4 28 36.3 59 7.01 95 16.4 106 10.5 68 8.51 70 17.2 72 7.33 86
Aniso-Texture [82]85.5 7.16 40 11.6 46 8.78 62 8.84 98 16.5 112 6.86 78 8.38 112 11.8 67 5.99 66 22.4 91 21.4 107 42.5 104 31.0 121 27.5 100 46.0 120 29.0 116 24.2 142 36.7 95 6.70 42 14.7 39 10.3 7 8.90 105 18.0 107 7.27 75
Adaptive [20]85.8 7.71 102 13.2 116 9.21 97 9.40 112 16.8 116 7.07 84 7.87 80 12.4 86 6.12 76 22.0 62 20.3 55 41.8 59 30.7 80 27.3 77 45.6 56 28.4 86 20.7 73 36.8 100 6.95 91 16.0 100 10.4 37 8.87 101 17.9 103 7.55 111
Shiralkar [42]86.1 7.48 85 12.8 106 8.80 68 9.00 105 15.8 100 6.65 60 8.52 116 16.1 136 6.84 117 23.4 122 22.3 122 41.6 39 30.0 7 27.0 36 44.5 5 28.7 106 21.1 85 37.1 111 7.49 123 18.7 132 10.6 94 8.64 87 17.7 94 6.93 10
CNN-flow-warp+ref [117]89.7 7.35 66 10.8 16 9.30 100 8.87 99 16.2 108 8.14 103 8.60 120 14.1 127 6.62 111 23.7 127 21.9 117 42.7 112 30.8 92 27.3 77 45.9 109 28.0 51 19.1 19 36.7 95 7.37 118 18.5 129 10.6 94 8.33 46 16.8 41 7.27 75
Black & Anandan [4]90.1 8.54 122 12.8 106 10.2 121 10.9 128 17.3 126 9.40 121 9.06 129 13.6 116 6.99 121 22.9 110 21.3 105 41.7 50 30.7 80 27.2 63 45.9 109 28.0 51 18.6 10 36.7 95 6.93 87 15.9 98 10.4 37 8.46 62 17.0 53 7.20 62
CRTflow [80]90.1 7.69 100 12.6 99 9.28 98 8.45 91 15.5 91 6.81 74 8.55 118 14.0 126 7.29 126 22.4 91 20.7 79 43.8 126 30.7 80 27.2 63 45.7 71 28.1 61 19.6 34 36.7 95 6.87 80 15.8 94 10.6 94 8.59 80 17.2 72 7.65 119
HBpMotionGpu [43]90.8 9.39 132 14.6 132 11.3 134 11.7 133 18.9 141 11.5 134 7.55 40 11.1 50 6.00 68 23.3 119 22.3 122 43.5 123 30.3 24 27.2 63 45.2 19 28.7 106 20.9 75 37.1 111 6.62 22 14.2 20 10.5 68 8.99 114 17.8 100 8.04 129
GraphCuts [14]91.0 8.65 124 14.1 130 9.83 113 8.28 83 14.2 60 9.28 118 9.89 138 10.6 30 7.38 129 23.0 111 21.1 96 42.5 104 30.3 24 27.3 77 44.7 8 27.2 2 21.4 93 34.7 2 7.42 121 17.8 119 11.0 119 9.32 130 18.9 132 7.66 120
StereoOF-V1MT [119]91.4 7.65 98 13.5 122 8.77 59 8.69 96 15.9 105 6.52 49 9.43 134 15.4 133 7.23 124 24.4 133 22.3 122 43.2 121 30.5 43 27.2 63 45.0 12 28.9 114 21.2 87 37.2 115 7.77 136 19.4 136 11.0 119 8.26 40 16.4 23 6.93 10
CBF [12]92.2 7.41 77 11.9 63 9.31 102 8.07 79 14.9 79 7.14 89 7.69 64 11.1 50 5.95 58 22.8 105 20.7 79 45.1 136 30.8 92 27.3 77 47.0 138 28.2 70 20.6 70 36.5 77 7.17 108 16.6 109 11.2 125 9.16 124 17.9 103 8.83 139
HBM-GC [105]92.5 7.91 107 12.6 99 9.75 110 7.51 62 13.9 52 6.80 73 7.29 16 9.43 3 5.94 57 22.0 62 19.7 38 42.3 92 32.1 141 28.6 138 48.0 143 30.0 132 24.6 145 37.8 126 7.14 103 14.8 48 11.6 135 8.95 111 17.7 94 8.28 132
Steered-L1 [118]93.0 7.06 28 12.2 77 8.59 33 7.40 57 14.3 63 6.83 75 8.48 115 11.7 64 6.69 113 22.8 105 20.9 86 42.7 112 31.2 129 28.1 132 45.8 88 28.3 74 21.2 87 36.7 95 7.25 114 17.8 119 10.9 115 9.00 115 18.3 121 7.58 114
TriangleFlow [30]93.0 7.79 105 13.0 112 9.16 93 8.36 87 15.5 91 6.69 64 8.20 102 11.9 70 6.59 108 22.5 97 21.0 92 42.5 104 30.1 9 27.0 36 45.0 12 28.9 114 22.6 122 36.5 77 7.42 121 18.3 124 11.0 119 9.49 134 19.3 136 7.47 104
Correlation Flow [75]93.3 7.05 26 11.7 51 8.32 4 8.29 84 15.6 94 6.56 52 7.64 55 10.8 41 5.89 51 22.2 79 20.1 47 42.9 118 31.7 135 27.7 119 49.9 148 29.6 126 23.8 138 37.2 115 7.62 131 19.0 134 11.3 128 9.22 126 18.6 127 7.51 110
IAOF2 [51]95.5 8.43 117 13.6 124 9.76 111 9.86 120 17.4 127 8.67 105 7.74 70 12.2 79 6.33 93 23.1 114 21.7 114 42.3 92 31.0 121 27.9 124 45.6 56 28.5 94 21.0 78 36.6 85 6.71 44 15.0 63 10.3 7 9.14 122 18.4 124 7.49 108
BriefMatch [124]98.0 7.38 75 12.0 65 8.85 74 7.71 66 14.5 71 7.86 100 8.77 125 11.7 64 7.25 125 24.2 131 22.0 119 46.2 141 30.8 92 27.4 91 46.1 125 31.8 146 21.7 100 41.3 146 6.85 73 15.3 76 10.7 104 8.51 70 17.1 61 7.56 113
SegOF [10]99.9 8.16 112 12.4 91 10.1 120 9.10 107 15.5 91 8.83 110 9.48 135 14.1 127 7.46 131 22.8 105 23.0 136 41.6 39 30.8 92 27.4 91 45.8 88 28.3 74 22.0 112 36.3 59 7.83 137 21.5 143 11.0 119 8.46 62 17.1 61 7.17 55
TV-L1-improved [17]100.8 7.55 91 12.9 109 9.15 91 9.36 110 16.9 118 7.19 91 8.63 122 12.2 79 6.92 119 22.2 79 21.0 92 42.3 92 30.8 92 27.5 100 45.6 56 28.5 94 21.4 93 36.8 100 7.38 119 18.6 131 10.7 104 8.94 109 18.0 107 7.75 122
BlockOverlap [61]101.1 8.81 126 12.4 91 11.1 133 10.0 123 15.8 100 10.6 128 7.84 77 10.4 21 6.59 108 23.3 119 20.4 62 46.3 142 31.9 138 27.9 124 48.8 146 30.3 139 19.7 41 39.8 141 7.08 100 14.5 32 11.7 138 8.35 50 16.1 11 8.60 137
OFRF [134]101.7 9.30 130 13.4 121 11.0 131 9.59 115 15.7 97 9.07 112 7.92 83 12.7 96 6.05 73 22.3 85 20.2 51 42.8 114 31.0 121 27.9 124 45.4 29 29.6 126 23.3 135 37.4 121 7.34 116 17.7 118 10.5 68 9.09 117 18.8 131 7.05 26
Dynamic MRF [7]102.5 7.29 60 13.1 113 8.69 46 8.20 82 16.3 109 6.74 68 9.18 131 16.4 139 7.22 123 24.5 135 23.1 137 44.4 130 30.3 24 27.2 63 45.1 16 29.2 123 23.4 136 37.2 115 7.64 132 19.8 139 10.7 104 9.14 122 18.0 107 7.48 107
LocallyOriented [52]102.9 8.08 109 13.1 113 9.72 108 9.73 117 17.0 121 7.88 101 8.34 109 12.8 99 6.34 94 23.0 111 22.1 120 43.0 119 30.6 60 27.2 63 45.6 56 30.0 132 21.9 107 38.5 136 7.02 96 15.8 94 10.5 68 9.05 116 18.4 124 7.39 96
AdaConv-v1 [126]103.1 9.81 137 13.9 128 11.6 137 12.1 135 17.6 129 16.0 146 11.4 145 16.1 136 13.1 148 26.5 144 24.4 145 45.3 137 28.4 3 24.4 2 44.6 6 28.4 86 18.1 7 37.7 125 7.74 135 16.3 104 13.1 149 8.25 38 15.1 5 10.1 148
SPSA-learn [13]103.7 8.28 115 12.9 109 9.95 117 9.92 121 16.3 109 9.49 122 9.15 130 12.8 99 7.30 127 23.1 114 20.5 69 41.6 39 30.8 92 27.5 100 45.7 71 28.0 51 20.4 58 36.3 59 8.81 149 27.1 150 11.8 139 10.0 142 21.0 145 7.20 62
Rannacher [23]104.0 7.69 100 13.2 116 9.32 103 9.37 111 16.9 118 7.28 92 8.67 123 13.0 103 6.91 118 22.2 79 21.1 96 42.4 102 30.8 92 27.5 100 45.7 71 28.5 94 21.2 87 36.9 105 7.35 117 18.5 129 10.7 104 8.90 105 18.0 107 7.78 124
Ad-TV-NDC [36]105.4 10.8 141 13.9 128 13.4 141 11.6 132 17.6 129 11.2 132 7.77 73 12.3 83 6.12 76 24.0 129 21.6 113 44.4 130 31.1 126 27.6 115 46.1 125 29.0 116 19.3 27 38.0 130 6.87 80 15.4 82 10.5 68 8.59 80 17.0 53 7.71 121
ACK-Prior [27]105.6 7.12 35 11.7 51 8.57 30 7.08 43 13.8 51 6.34 25 8.81 126 11.8 67 6.69 113 22.5 97 21.4 107 42.3 92 32.6 145 29.3 144 48.2 144 30.7 142 25.6 147 38.1 132 7.95 140 18.8 133 12.0 140 10.8 147 21.8 147 8.53 136
TVL1_ROB [139]107.0 10.2 140 13.8 126 12.5 140 12.6 142 19.1 142 11.9 136 8.01 89 13.7 120 6.37 98 23.6 125 21.4 107 42.3 92 30.8 92 27.3 77 46.0 120 28.6 100 20.2 49 37.1 111 7.29 115 17.6 117 10.7 104 8.49 67 17.1 61 7.40 97
Horn & Schunck [3]107.0 8.45 119 13.3 120 10.0 118 11.4 131 18.1 136 9.84 126 9.65 136 16.1 136 7.89 134 24.6 136 22.8 130 42.8 114 30.6 60 27.2 63 45.6 56 28.3 74 19.4 28 36.8 100 7.41 120 18.0 123 10.6 94 8.94 109 17.7 94 7.55 111
UnFlow [129]109.5 9.13 129 15.0 137 10.7 129 10.9 128 18.1 136 9.23 116 9.21 132 16.9 142 7.18 122 22.4 91 21.8 116 41.8 59 30.8 92 27.6 115 45.9 109 28.6 100 22.7 125 36.0 23 6.94 89 15.6 87 10.5 68 10.0 142 19.3 136 7.47 104
StereoFlow [44]110.0 13.8 146 20.2 149 14.0 142 14.1 146 21.3 149 12.0 138 7.79 75 13.3 112 5.98 63 22.4 91 20.9 86 42.1 86 33.7 149 32.3 149 46.1 125 30.5 140 31.8 150 36.3 59 6.63 25 14.7 39 10.4 37 9.98 141 21.0 145 7.42 99
TI-DOFE [24]110.4 11.8 143 14.7 134 14.8 144 13.9 145 20.3 146 13.5 144 9.26 133 16.5 141 7.69 133 25.1 138 22.8 130 43.3 122 30.2 14 27.1 48 45.4 29 28.4 86 19.6 34 36.8 100 7.22 111 17.2 114 10.7 104 9.21 125 18.1 115 7.59 117
Filter Flow [19]115.8 8.30 116 13.2 116 10.0 118 10.8 127 17.1 122 11.7 135 7.96 85 12.1 76 6.38 99 23.7 127 20.9 86 44.5 134 31.5 134 28.2 134 46.7 136 28.8 112 21.0 78 37.3 120 7.15 105 17.0 113 10.7 104 9.54 136 18.9 132 8.47 135
WOLF_ROB [149]115.9 8.87 127 16.2 142 9.71 107 9.97 122 16.9 118 7.77 98 8.60 120 13.0 103 6.39 100 23.2 118 23.5 138 43.7 124 30.9 115 27.9 124 45.8 88 30.1 135 22.8 128 38.2 134 7.54 126 19.0 134 10.7 104 9.12 121 18.7 128 7.06 27
NL-TV-NCC [25]118.6 7.56 92 12.7 105 8.62 39 8.00 77 15.4 88 6.74 68 8.46 114 13.1 105 6.70 116 24.2 131 24.0 143 45.0 135 32.8 146 28.3 137 52.0 151 29.4 124 24.1 141 36.9 105 7.89 139 17.9 121 12.4 145 10.1 144 19.9 141 8.92 140
Bartels [41]120.0 8.10 111 13.8 126 9.94 116 8.35 86 15.8 100 8.75 106 8.11 99 12.1 76 6.97 120 24.1 130 22.7 127 47.6 144 32.4 142 27.8 123 51.1 150 35.4 149 23.0 132 46.5 150 7.18 109 14.9 54 12.3 144 9.36 133 18.0 107 9.76 145
SILK [79]120.0 9.77 136 15.1 138 11.8 138 12.3 136 18.7 140 11.2 132 10.3 139 16.4 139 8.14 136 25.2 139 22.8 130 45.9 139 30.8 92 27.5 100 45.7 71 30.6 141 20.3 54 40.1 144 7.19 110 16.8 111 10.9 115 8.87 101 17.6 91 7.50 109
H+S_ROB [138]120.4 9.52 133 14.2 131 11.3 134 12.4 138 18.0 133 11.9 136 12.0 146 20.2 147 9.80 144 28.3 146 22.7 127 44.0 127 30.6 60 27.7 119 45.6 56 28.4 86 21.0 78 36.5 77 8.24 143 22.1 144 11.4 131 9.34 132 17.7 94 7.84 127
SLK [47]126.0 11.4 142 15.4 139 14.4 143 12.4 138 18.0 133 12.6 140 10.9 143 17.6 144 8.85 139 27.8 145 25.2 146 46.6 143 30.6 60 28.1 132 43.6 2 29.0 116 21.9 107 37.0 109 8.25 144 22.4 145 11.3 128 9.33 131 18.5 126 7.91 128
GroupFlow [9]126.8 10.1 139 16.9 144 11.3 134 10.4 125 17.8 132 10.0 127 10.8 142 17.5 143 9.21 141 23.6 125 23.9 140 42.5 104 31.9 138 29.3 144 46.2 131 30.1 135 24.5 144 37.8 126 7.55 127 18.4 127 10.6 94 9.52 135 19.8 140 6.89 7
Heeger++ [104]127.2 9.81 137 17.3 145 10.4 124 11.3 130 17.2 123 9.67 123 13.6 148 23.8 149 10.2 146 26.3 142 22.8 130 44.4 130 31.8 137 28.9 143 46.3 133 29.6 126 22.0 112 37.5 123 8.17 142 19.8 139 10.9 115 9.10 119 18.2 118 7.02 20
Learning Flow [11]128.8 8.21 113 14.8 135 9.74 109 9.78 119 17.6 129 8.11 102 9.68 137 15.5 134 7.56 132 25.0 137 24.3 144 45.4 138 31.9 138 28.7 141 47.3 139 29.4 124 22.0 112 37.8 126 7.49 123 18.3 124 10.9 115 10.2 145 20.2 143 8.31 133
2bit-BM-tele [98]129.9 8.61 123 13.5 122 10.5 126 10.0 123 17.5 128 9.73 124 8.26 106 11.5 58 7.40 130 24.4 133 22.7 127 48.1 145 32.5 144 28.6 138 50.2 149 34.7 148 24.3 143 44.9 148 9.35 150 26.2 149 13.8 150 9.25 127 17.3 77 10.2 149
FFV1MT [106]134.4 9.53 134 16.7 143 10.7 129 12.6 142 18.2 138 12.8 142 13.3 147 23.5 148 10.5 147 26.3 142 22.8 130 44.4 130 31.4 133 28.2 134 46.1 125 29.8 130 20.6 70 38.1 132 8.32 146 20.5 141 11.0 119 10.5 146 20.4 144 8.45 134
Adaptive flow [45]137.3 13.2 145 15.9 140 16.2 146 14.2 147 19.9 145 16.4 147 9.02 128 13.1 105 8.03 135 26.0 141 22.8 130 48.6 146 32.4 142 29.4 146 47.9 142 30.1 135 24.6 145 38.0 130 7.55 127 16.9 112 12.2 142 9.85 138 19.5 138 8.97 143
FOLKI [16]137.8 15.0 148 17.4 146 19.4 148 14.3 148 20.9 148 14.4 145 10.7 141 19.2 146 9.99 145 29.8 148 26.8 147 53.1 149 31.3 131 28.6 138 45.8 88 30.0 132 21.7 100 38.9 138 7.85 138 19.4 136 11.6 135 9.85 138 19.2 135 8.80 138
Pyramid LK [2]140.0 16.3 149 16.1 141 21.6 149 16.0 149 20.3 146 18.2 149 16.7 149 15.3 132 14.3 149 35.7 150 36.7 150 56.5 150 32.8 146 31.2 148 45.7 71 29.7 129 22.1 116 38.2 134 8.31 145 23.1 146 11.6 135 11.8 148 25.0 148 8.22 131
PGAM+LK [55]141.3 12.7 144 18.1 147 15.3 145 12.4 138 19.1 142 13.0 143 11.1 144 18.6 145 9.33 143 29.2 147 27.5 148 51.6 148 31.7 135 28.8 142 46.7 136 31.3 144 23.7 137 40.1 144 7.67 134 19.4 136 11.4 131 9.89 140 19.5 138 8.95 141
HCIC-L [99]142.1 18.0 150 18.7 148 23.1 150 12.7 144 17.2 123 17.0 148 10.5 140 14.4 130 8.49 138 25.7 140 23.9 140 44.3 129 33.2 148 29.8 147 49.0 147 31.7 145 26.4 149 39.2 140 7.98 141 18.4 127 12.4 145 12.4 149 25.3 150 8.96 142
Periodicity [78]148.7 14.9 147 20.8 150 18.2 147 20.1 150 22.0 150 21.5 150 17.7 150 26.4 150 16.1 150 29.8 148 34.8 149 49.7 147 35.4 150 34.2 150 48.7 145 37.1 150 25.8 148 47.4 151 8.68 148 23.6 147 12.2 142 13.3 150 25.1 149 11.6 150
AVG_FLOW_ROB [142]150.5 46.4 151 51.9 151 42.8 151 44.1 151 40.8 151 44.3 151 39.2 151 37.3 151 32.5 151 57.2 151 58.7 151 63.7 151 41.6 151 42.4 151 47.5 140 46.3 151 59.4 151 45.4 149 25.5 151 33.4 151 16.2 151 33.6 151 39.6 151 34.2 151
Move the mouse over the numbers in the table to see the corresponding images. Click to compare with the ground truth.

References

Methodtime*framescolor Reference and notes
[1] 2D-CLG 844 2 gray The 2D-CLG method by Bruhn et al. as implemented by Stefan Roth. [A. Bruhn, J. Weickert, and C. Schnörr. Lucas/Kanade meets Horn/Schunck: combining local and global optic flow methods. IJCV 63(3), 2005.] Parameters were set to match the published performance on Yosemite sequence, which may not be optimal for other sequences.
[2] Pyramid LK 12 2 color A modification of Bouguet's pyramidal implementation of Lucas-Kanade.
[3] Horn & Schunck 49 2 gray A modern Matlab implementation of the Horn & Schunck method by Deqing Sun. Parameters set to optimize AAE on all training data.
[4] Black & Anandan 328 2 gray A modern Matlab implementation of the Black & Anandan method by Deqing Sun.
[5] Brox et al. 18 2 color T. Brox, A. Bruhn, N. Papenberg, and J. Weickert. High accuracy optical flow estimation based on a theory for warping. ECCV 2004. (Improved using separate robust functions as proposed in A. Bruhn and J. Weickert, Towards ultimate motion estimation, ICCV 2005; improved by training on the training set.)
[6] Fusion 2,666 2 color V. Lempitsky, S. Roth, and C. Rother. Discrete-continuous optimization for optical flow estimation. CVPR 2008.
[7] Dynamic MRF 366 2 gray B. Glocker, N. Paragios, N. Komodakis, G. Tziritas, and N. Navab. Optical flow estimation with uncertainties through dynamic MRFs. CVPR 2008. (Method improved since publication.)
[8] Second-order prior 14 2 gray W. Trobin, T. Pock, D. Cremers, and H. Bischof. An unbiased second-order prior for high-accuracy motion estimation. DAGM 2008. (Method improved since publication; for details see W. Trobin, Ph.D. thesis, 2009.)
[9] GroupFlow 600 2 gray X. Ren. Local Grouping for Optical Flow. CVPR 2008.
[10] SegOF 60 2 color L. Xu, J. Chen, and J. Jia. Segmentation based variational model for accurate optical flow estimation. ECCV 2008. Code available.
[11] Learning Flow 825 2 gray D. Sun, S. Roth, J.P. Lewis, and M. Black. Learning optical flow (SRF-LFC). ECCV 2008.
[12] CBF 69 2 color W. Trobin, T. Pock, D. Cremers, and H. Bischof. Continuous energy minimization via repeated binary fusion. ECCV 2008. (Method improved since publication; for details see W. Trobin, Ph.D. thesis, 2009.)
[13] SPSA-learn 200 2 color Y. Li and D. Huttenlocher. Learning for optical flow using stochastic optimization. ECCV 2008.
[14] GraphCuts 1,200 2 color T. Cooke. Two applications of graph-cuts to image processing. DICTA 2008.
[15] F-TV-L1 8 2 gray A. Wedel, T. Pock, J. Braun, U. Franke, and D. Cremers. Duality TV-L1 flow with fundamental matrix prior. IVCNZ 2008.
[16] FOLKI 1.4 2 gray G. Le Besnerais and F. Champagnat. Dense optical flow by iterative local window registration. ICIP 2005.
[17] TV-L1-improved 2.9 2 gray A. Wedel, T. Pock, C. Zach, H. Bischof, and D. Cremers. An improved algorithm for TV-L1 optical flow computation. Proceedings of the Dagstuhl Visual Motion Analysis Workshop 2008. Code at GPU4Vision.
[18] DPOF 287 2 color C. Lei and Y.-H. Yang. Optical flow estimation on coarse-to-fine region-trees using discrete optimization. ICCV 2009. (Method improved since publication.)
[19] Filter Flow 34,000 2 color S. Seitz and S. Baker. Filter flow. ICCV 2009.
[20] Adaptive 9.2 2 gray A. Wedel, D. Cremers, T. Pock, and H. Bischof. Structure- and motion-adaptive regularization for high accuracy optic flow. ICCV 2009.
[21] Complementary OF 44 2 color H. Zimmer, A. Bruhn, J. Weickert, L. Valgaerts, A. Salgado, B. Rosenhahn, and H.-P. Seidel. Complementary optic flow. EMMCVPR 2009.
[22] Aniso. Huber-L1 2 2 gray M. Werlberger, W. Trobin, T. Pock, A. Wedel, D. Cremers, and H. Bischof. Anisotropic Huber-L1 optical flow. BMVC 2009. Code at GPU4Vision.
[23] Rannacher 0.12 2 gray J. Rannacher. Realtime 3D motion estimation on graphics hardware. Bachelor thesis, Heidelberg University, 2009.
[24] TI-DOFE 260 2 gray C. Cassisa, S. Simoens, and V. Prinet. Two-frame optical flow formulation in an unwarped multiresolution scheme. CIARP 2009.
[25] NL-TV-NCC 20 2 color M. Werlberger, T. Pock, and H. Bischof. Motion estimation with non-local total variation regularization. CVPR 2010.
[26] MDP-Flow 188 2 color L. Xu, J. Jia, and Y. Matsushita. Motion detail preserving optical flow estimation. CVPR 2010.
[27] ACK-Prior 5872 2 color K. Lee, D. Kwon, I. Yun, and S. Lee. Optical flow estimation with adaptive convolution kernel prior on discrete framework. CVPR 2010.
[28] LDOF 122 2 color T. Brox and J. Malik. Large displacement optical flow: descriptor matching in variational motion estimation. PAMI 33(3):500-513, 2011.
[29] p-harmonic 565 2 gray J. Gai and R. Stevenson. Optical flow estimation with p-harmonic regularization. ICIP 2010.
[30] TriangleFlow 4200 2 gray B. Glocker, H. Heibel, N. Navab, P. Kohli, and C. Rother. TriangleFlow: Optical flow with triangulation-based higher-order likelihoods. ECCV 2010.
[31] Classic+NL 972 2 color D. Sun, S. Roth, and M. Black. Secrets of optical flow estimation and their principles. CVPR 2010. Matlab code.
[32] Classic++ 486 2 gray A modern implementation of the classical formulation descended from Horn & Schunck and Black & Anandan; see D. Sun, S. Roth, and M. Black, Secrets of optical flow estimation and their principles, CVPR 2010.
[33] Nguyen 33 2 gray D. Nguyen. Tuning optical flow estimation with image-driven functions. ICRA 2011.
[34] Modified CLG 133 2 gray R. Fezzani, F. Champagnat, and G. Le Besnerais. Combined local global method for optic flow computation. EUSIPCO 2010.
[35] ComplOF-FED-GPU 0.97 2 color P. Gwosdek, H. Zimmer, S. Grewenig, A. Bruhn, and J. Weickert. A highly efficient GPU implementation for variational optic flow based on the Euler-Lagrange framework. CVGPU Workshop 2010.
[36] Ad-TV-NDC 35 2 gray M. Nawaz. Motion estimation with adaptive regularization and neighborhood dependent constraint. DICTA 2010.
[37] Layers++ 18206 2 color D. Sun, E. Sudderth, and M. Black. Layered image motion with explicit occlusions, temporal consistency, and depth ordering. NIPS 2010.
[38] OFH 620 3 color H. Zimmer, A. Bruhn, J. Weickert. Optic flow in harmony. IJCV 93(3) 2011.
[39] LSM 1615 2 color K. Jia, X. Wang, and X. Tang. Optical flow estimation using learned sparse model. ICCV 2011.
[40] CostFilter 55 2 color C. Rhemann, A. Hosni, M. Bleyer, C. Rother, and M. Gelautz. Fast cost-volume filtering for visual correspondence and beyond. CVPR 2011.
[41] Bartels 0.15 2 gray C. Bartels and G. de Haan. Smoothness constraints in recursive search motion estimation for picture rate conversion. IEEE TCSVT 2010. Version improved since publication: mapped on GPU.
[42] Shiralkar 600 2 gray M. Shiralkar and R. Schalkoff. A self organization-based optical flow estimator with GPU implementation. MVA 23(6):1229-1242.
[43] HBpMotionGpu 1000 5 gray S. Grauer-Gray and C. Kambhamettu. Hierarchical belief propagation to reduce search space using CUDA for stereo and motion estimation. WACV 2009. (Method improved since publication.)
[44] StereoFlow 7200 2 color G. Rosman, S. Shem-Tov, D. Bitton, T. Nir, G. Adiv, R. Kimmel, A. Feuer, and A. Bruckstein. Over-parameterized optical flow using a stereoscopic constraint. SSVM 2011:761-772.
[45] Adaptive flow 121 2 gray T. Arici. Energy minimization based motion estimation using adaptive smoothness priors. Submitted to IEEE TIP 2011.
[46] TC-Flow 2500 5 color S. Volz, A. Bruhn, L. Valgaerts, and H. Zimmer. Modeling temporal coherence for optical flow. ICCV 2011.
[47] SLK 300 2 gray T. Corpetti and E. Mémin. Stochastic uncertainty models for the luminance consistency assumption. IEEE TIP 2011.
[48] CLG-TV 29 2 gray M. Drulea. Total variation regularization of local-global optical flow. ITSC 2011. Matlab code.
[49] SimpleFlow 1.7 2 color M. Tao, J. Bai, P. Kohli, S. Paris. SimpleFlow: a non-iterative, sublinear optical flow algorithm. EUROGRAPHICS 2012.
[50] IAOF 57 2 gray D. Nguyen. Improving motion estimation using image-driven functions and hybrid scheme. PSIVT 2011.
[51] IAOF2 56 2 gray D. Nguyen. Enhancing the sharpness of flow field using image-driven functions with occlusion-aware filter. Submitted to TIP 2011.
[52] LocallyOriented 9541 2 gray Y.Niu, A. Dick, and M. Brooks. Locally oriented optical flow computation. To appear in TIP 2012.
[53] IROF-TV 261 2 color H. Rashwan, D. Puig, and M. Garcia. On improving the robustness of differential optical flow. ICCV 2011 Artemis workshop.
[54] Sparse Occlusion 2312 2 color A. Ayvaci, M. Raptis, and S. Soatto. Sparse occlusion detection with optical flow. Submitted to IJCV 2011.
[55] PGAM+LK 0.37 2 gray A. Alba, E. Arce-Santana, and M. Rivera. Optical flow estimation with prior models obtained from phase correlation. ISVC 2010.
[56] Sparse-NonSparse 713 2 color L. Chen, J. Wang, and Y. Wu. Decomposing and regularizing sparse/non-sparse components for motion field estimation. Submitted to PAMI 2013.
[57] nLayers 36150 4 color D. Sun, E. Sudderth, and M. Black. Layered segmentation and optical flow estimation over time. CVPR 2012.
[58] IROF++ 187 2 color H. Rashwan, D. Puig, and M. Garcia. Variational optical flow estimation based on stick tensor voting. IEEE TIP 2013.
[59] COFM 600 3 color M. Mozerov. Constrained optical flow estimation as a matching problem. IEEE TIP 2013.
[60] Efficient-NL 400 2 color P. Krähenbühl and V. Koltun. Efficient nonlocal regularization for optical flow. ECCV 2012.
[61] BlockOverlap 2 2 gray M. Santoro, G. AlRegib, and Y. Altunbasak. Motion estimation using block overlap minimization. Submitted to MMSP 2012.
[62] Ramp 1200 2 color A. Singh and N. Ahuja. Exploiting ramp structures for improving optical flow estimation. ICPR 2012.
[63] Occlusion-TV-L1 538 3 gray C. Ballester, L. Garrido, V. Lazcano, and V. Caselles. A TV-L1 optical flow method with occlusion detection. DAGM-OAGM 2012.
[64] TV-L1-MCT 90 2 color M. Mohamed and B. Mertsching. TV-L1 optical flow estimation with image details recovering based on modified census transform. ISVC 2012.
[65] Local-TV-L1 500 2 gray L. Raket. Local smoothness for global optical flow. ICIP 2012.
[66] ALD-Flow 61 2 color M. Stoll, A. Bruhn, and S. Volz. Adaptive integration of feature matches into variational optic flow methods. ACCV 2012.
[67] SIOF 234 2 color L. Xu, Z. Dai, and J. Jia. Scale invariant optical flow. ECCV 2012.
[68] MDP-Flow2 342 2 color L. Xu, J. Jia, and Y. Matsushita. Motion detail preserving optical flow estimation. PAMI 34(9):1744-1757, 2012. Code available.
[69] TCOF 1421 all gray J. Sanchez, A. Salgado, and N. Monzon. Optical flow estimation with consistent spatio-temporal coherence models. VISAPP 2013.
[70] LME 476 2 color W. Li, D. Cosker, M. Brown, and R. Tang. Optical flow estimation using Laplacian mesh energy. CVPR 2013.
[71] NN-field 362 2 color L. Chen, H. Jin, Z. Lin, S. Cohen, and Y. Wu. Large displacement optical flow from nearest neighbor fields. CVPR 2013.
[72] FESL 3310 2 color W. Dong, G. Shi, X. Hu, and Y. Ma. Nonlocal sparse and low-rank regularization for optical flow estimation. Submitted to IEEE TIP 2013.
[73] PMF 35 2 color J. Lu, H. Yang, D. Min, and M. Do. PatchMatch filter: efficient edge-aware filtering meets randomized search for fast correspondence field estimation. CVPR 2013.
[74] FC-2Layers-FF 2662 4 color D. Sun, J. Wulff, E. Sudderth, H. Pfister, and M. Black. A fully-connected layered model of foreground and background flow. CVPR 2013.
[75] Correlation Flow 290 2 color M. Drulea and S. Nedevschi. Motion estimation using the correlation transform. TIP 2013. Matlab code.
[76] TC/T-Flow 341 5 color M. Stoll, S. Volz, and A. Bruhn. Joint trilateral filtering for multiframe optical flow. ICIP 2013.
[77] OFLAF 1530 2 color T. Kim, H. Lee, and K. Lee. Optical flow via locally adaptive fusion of complementary data costs. ICCV 2013.
[78] Periodicity 8000 4 color G. Khachaturov, S. Gonzalez-Brambila, and J. Gonzalez-Trejo. Periodicity-based computation of optical flow. Submitted to Computacion y Sistemas (CyS) 2013.
[79] SILK 572 2 gray P. Zille, C. Xu, T. Corpetti, L. Shao. Observation models based on scale interactions for optical flow estimation. Submitted to IEEE TIP.
[80] CRTflow 13 3 color O. Demetz, D. Hafner, and J. Weickert. The complete rank transform: a tool for accurate and morphologically invariant matching of structures. BMVC 2013.
[81] SuperFlow 178 2 color Anonymous. Superpixel based optical flow estimation. ICCV 2013 submission 507.
[82] Aniso-Texture 300 2 color Anonymous. Texture information-based optical flow estimation using an incremental multi-resolution approach. ITC-CSCC 2013 submission 267.
[83] Classic+CPF 640 2 gray Z. Tu, R. Veltkamp, and N. van der Aa. A combined post-filtering method to improve accuracy of variational optical flow estimation. Submitted to Pattern Recognition 2013.
[84] S2D-Matching 1200 2 color Anonymous. Locally affine sparse-to-dense matching for motion and occlusion estimation. ICCV 2013 submission 1479.
[85] AGIF+OF 438 2 gray Z. Tu, R. Poppe, and R. Veltkamp. Adaptive guided image filter to warped interpolation image for variational optical flow computation. Submitted to Signal Processing 2015.
[86] DeepFlow 13 2 color P. Weinzaepfel, J. Revaud, Z. Harchaoui, and C. Schmid. DeepFlow: large displacement optical flow with deep matching. ICCV 2013.
[87] NNF-Local 673 2 color Z. Chen, H. Jin, Z. Lin, S. Cohen, and Y. Wu. Large displacement optical flow with nearest neighbor field. Submitted to PAMI 2014.
[88] EPPM w/o HM 2.5 2 color L. Bao, Q. Yang, and H. Jin. Fast edge-preserving PatchMatch for large displacement optical flow. CVPR 2014.
[89] MLDP_OF 165 2 gray M. Mohamed, H. Rashwan, B. Mertsching, M. Garcia, and D. Puig. Illumination-robust optical flow approach using local directional pattern. IEEE TCSVT 24(9):1499-1508, 2014.
[90] RFlow 20 2 gray S. Ali, C. Daul, and W. Blondel. Robust and accurate optical flow estimation for weak texture and varying illumination condition: Application to cystoscopy. IPTA 2014.
[91] SRR-TVOF-NL 32 all color P. Pohl, M. Sirotenko, E. Tolstaya, and V. Bucha. Edge preserving motion estimation with occlusions correction for assisted 2D to 3D conversion. IS&T/SPIE Electronic Imaging 2014.
[92] 2DHMM-SAS 157 2 color M.-C. Shih, R. Shenoy, and K. Rose. A two-dimensional hidden Markov model with spatially-adaptive states with application of optical flow. ICIP 2014 submission.
[93] WLIF-Flow 700 2 color Z. Tu, R. Veltkamp, N. van der Aa, and C. Van Gemeren. Weighted local intensity fusion method for variational optical flow estimation. Submitted to TIP 2014.
[94] FMOF 215 2 color N. Jith, A. Ramakanth, and V. Babu. Optical flow estimation using approximate nearest neighbor field fusion. ICASSP 2014.
[95] TriFlow 150 2 color TriFlow. Optical flow with geometric occlusion estimation and fusion of multiple frames. ECCV 2014 submission 914.
[96] ComponentFusion 6.5 2 color Anonymous. Fast optical flow by component fusion. ECCV 2014 submission 941.
[97] AggregFlow 1642 2 color D. Fortun, P. Bouthemy, and C. Kervrann. Aggregation of local parametric candidates and exemplar-based occlusion handling for optical flow. Preprint arXiv:1407.5759.
[98] 2bit-BM-tele 124 2 gray R. Xu and D. Taubman. Robust dense block-based motion estimation using a two-bit transform on a Laplacian pyramid. ICIP 2013.
[99] HCIC-L 330 2 color Anonymous. Globally-optimal image correspondence using a hierarchical graphical model. NIPS 2014 submission 114.
[100] TF+OM 600 2 color R. Kennedy and C. Taylor. Optical flow with geometric occlusion estimation and fusion of multiple frames. EMMCVPR 2015.
[101] PH-Flow 800 2 color J. Yang and H. Li. Dense, accurate optical flow estimation with piecewise parametric model. CVPR 2015.
[102] EpicFlow 16 2 color J. Revaud, P. Weinzaepfel, Z. Harchaoui, and C. Schmid. EpicFlow: edge-preserving interpolation of correspondences for optical flow. CVPR 2015.
[103] NNF-EAC 380 2 color Anonymous. Variational method for joint optical flow estimation and edge-aware image restoration. CVPR 2015 submission 2336.
[104] Heeger++ 6600 5 gray Anonymous. A context aware biologically inspired algorithm for optical flow (updated results). CVPR 2015 submission 2238.
[105] HBM-GC 330 2 color A. Zheng and Y. Yuan. Motion estimation via hierarchical block matching and graph cut. Submitted to ICIP 2015.
[106] FFV1MT 358 5 gray F. Solari, M. Chessa, N. Medathati, and P. Kornprobst. What can we expect from a V1-MT feedforward architecture for optical flow estimation? Submitted to Signal Processing: Image Communication 2015.
[107] ROF-ND 4 2 color S. Ali, C. Daul, E. Galbrun, and W. Blondel. Illumination invariant large displacement optical flow using robust neighbourhood descriptors. Submitted to CVIU 2015.
[108] DeepFlow2 16 2 color J. Revaud, P. Weinzaepfel, Z. Harchaoui, and C. Schmid. Deep convolutional matching. Submitted to IJCV, 2015.
[109] HAST 2667 2 color Anonymous. Highly accurate optical flow estimation on superpixel tree. ICCV 2015 submission 2221.
[110] FlowFields 15 2 color C. Bailer, B. Taetz, and D. Stricker. Flow Fields: Dense unregularized correspondence fields for highly accurate large displacement optical flow estimation. ICCV 2015.
[111] SVFilterOh 1.56 2 color Anonymous. Fast estimation of large displacement optical flow using PatchMatch and dominant motion patterns. CVPR 2016 submission 1788.
[112] FlowNetS+ft+v 0.5 2 color Anonymous. Learning optical flow with convolutional neural networks. ICCV 2015 submission 235.
[113] CombBMOF 51 2 color M. Brüggemann, R. Kays, P. Springer, and O. Erdler. Combined block-matching and adaptive differential motion estimation in a hierarchical multi-scale framework. ICGIP 2014. (Method improved since publication.)
[114] PMMST 182 2 color F. Zhang, S. Xu, and X. Zhang. High accuracy correspondence field estimation via MST based patch matching. Submitted to TIP 2015.
[115] DF-Auto 70 2 color N. Monzon, A. Salgado, and J. Sanchez. Regularization strategies for discontinuity-preserving optical flow methods. Submitted to TIP 2015.
[116] CPM-Flow 3 2 color Anonymous. Efficient coarse-to-fine PatchMatch for large displacement optical flow. CVPR 2016 submission 241.
[117] CNN-flow-warp+ref 1.4 3 color D. Teney and M. Hebert. Learning to extract motion from videos in convolutional neural networks. ArXiv 1601.07532, 2016.
[118] Steered-L1 804 2 color Anonymous. Optical flow estimation via steered-L1 norm. Submitted to WSCG 2016.
[119] StereoOF-V1MT 343 2 gray Anonymous. Visual features for action-oriented tasks: a cortical-like model for disparity and optic flow computation. BMVC 2016 submission 132.
[120] PGM-C 5 2 color Y. Li. Pyramidal gradient matching for optical flow estimation. Submitted to PAMI 2016.
[121] RNLOD-Flow 1040 2 gray C. Zhang, Z. Chen, M. Wang, M. Li, and S. Jiang. Robust non-local TV-L1 optical flow estimation with occlusion detection. IEEE TIP 26(8):4055-4067, 2017.
[122] FlowNet2 0.091 2 color Anonymous. FlowNet 2.0: Evolution of optical flow estimation with deep networks. CVPR 2017 submission 900.
[123] S2F-IF 20 2 color Anonymous. S2F-IF: Slow-to-fast interpolator flow. CVPR 2017 submission 765.
[124] BriefMatch 0.068 2 gray G. Eilertsen, P.-E. Forssen, and J. Unger. Dense binary feature matching for real-time optical flow estimation. SCIA 2017 submission 62.
[125] OAR-Flow 60 2 color Anonymous. Order-adaptive regularisation for variational optical flow: global, local and in between. SSVM 2017 submission 20.
[126] AdaConv-v1 2.8 2 color S. Niklaus, L. Mai, and F. Liu. (Interpolation results only.) Video frame interpolation via adaptive convolution. CVPR 2017.
[127] SepConv-v1 0.2 2 color S. Niklaus, L. Mai, and F. Liu. (Interpolation results only.) Video frame interpolation via adaptive separable convolution. ICCV 2017.
[128] ProbFlowFields 37 2 color A. Wannenwetsch, M. Keuper, and S. Roth. ProbFlow: joint optical flow and uncertainty estimation. ICCV 2017.
[129] UnFlow 0.12 2 color Anonymous. UnFlow: Unsupervised learning of optical flow with a bidirectional census loss. Submitted to AAAI 2018.
[130] FlowFields+ 10.5 2 color C. Bailer, B. Taetz, and D. Stricker. Flow fields: Dense correspondence fields for highly accurate large displacement optical flow estimation. Submitted to PAMI 2017.
[131] IIOF-NLDP 150 2 color D.-H. Trinh, W. Blondel, and C. Daul. A general form of illumination-invariant descriptors in variational optical flow estimation. ICIP 2017.
[132] SuperSlomo 0.5 2 color Anonymous. (Interpolation results only.) Super SloMo: High quality estimation of multiple intermediate frames for video interpolation. CVPR 2018 submission 325.
[133] EPMNet 0.061 2 color Anonymous. EPM-convolution multilayer-network for optical flow estimation. ICME 2018 submission 1119.
[134] OFRF 90 2 color T. Mai, M. Gouiffes, and S. Bouchafa. Optical Flow refinement using iterative propagation under color, proximity and flow reliability constraints. Submitted to Signal, Image and Video Processing 2017.
[135] 3DFlow 328 2 color J. Chen, Z. Cai, J. Lai, and X. Xie. A filtering based framework for optical flow estimation. IEEE TCSVT 2018.
[136] FGIK 0.18 2 color Anonymous. (Interpolation results only.) Learning flow-guided interpolation kernels for video frame synthesis. ECCV 2018 submission 433.
[137] CtxSyn 0.07 2 color S. Niklaus and F. Liu. (Interpolation results only.) Context-aware synthesis for video frame interpolation. CVPR 2018.
[138] H+S_ROB 5 2 color ROB 2018 baseline submission, based on: E. Meinhardt-Llopis, J. Sanchez, and D. Kondermann. Horn-Schunck optical flow with a multi-scale strategy. Image Processing On Line 3:151–172, 2013.
[139] TVL1_ROB 1 2 color ROB 2018 baseline submission, based on: J. Sanchez, E. Meinhardt-Llopis, and G. Facciolo. TV-L1 optical flow estimation. Image Processing On Line 3:137-150, 2013.
[140] DMF_ROB 10 2 color ROB 2018 baseline submission, based on: P. Weinzaepfel, J. Revaud, Z. Harchaoui, and C. Schmid. DeepFlow: large displacement optical flow with deep matching. ICCV 2013.
[141] JOF 657 2 gray C. Zhang, L. Ge, Z. Chen, M. Li, W. Liu, and H. Chen. Refined TV-L1 optical flow estimation using joint filtering. Submitted to IEEE TMM, 2018.
[142] AVG_FLOW_ROB N/A 2 N/A Average flow field of ROB 2018 training set.
[143] LiteFlowNet 0.06 2 color T.-W. Hui, X. Tang, and C. C. Loy. LiteFlowNet: A lightweight convolutional neural network for optical flow estimation. CVPR 2018.
[144] AugFNG_ROB 0.10 all color Anonymous. FusionNet and AugmentedFlowNet: Selective proxy ground truth for training on unlabeled images. ECCV 2018 submission 2834.
[145] ResPWCR_ROB 0.2 2 color Anonymous. Learning optical flow with residual connections. ROB 2018 submission.
[146] FF++_ROB 17.43 2 color R. Schuster, C. Bailer, O. Wasenmueller, D. Stricker. FlowFields++: Accurate optical flow correspondences meet robust interpolation. ICIP 2018. Submitted to ROB 2018.
[147] ProFlow_ROB 76 3 color Anonymous. ProFlow: Learning to predict optical flow. BMVC 2018 submission 277.
[148] PWC-Net_ROB 0.069 2 color D. Sun, X. Yang, M.-Y. Liu, and J. Kautz. PWC-Net: CNNs for optical flow using pyramid, warping, and cost volume. CVPR 2018.
[149] WOLF_ROB 0.02 2 color Anonymous. Reversed deep neural network for optical flow. ROB 2018 submission.
[150] ContFlow_ROB 0.45 all color Anonymous. Continual Flow. ROB 2018 submission.
[151] LFNet_ROB 0.068 2 color Anonymous. Learning a flow network. ROB 2018 submission.
* The "time" column lists the reported runtime in seconds on the "Urban" sequence. Note that these runtimes are not normalized by processor speed or type.