Optical flow evaluation results Statistics:     Average   SD   R2.5   R5.0   R10.0   A50   A75   A95  
Error type:   endpoint   angle   interpolation   normalized interpolation  
Show images: below table   above table        
A95
angle
error
avg. Army
(Hidden texture)
GT   im0   im1
Mequon
(Hidden texture)
GT   im0   im1
Schefflera
(Hidden texture)
GT   im0   im1
Wooden
(Hidden texture)
GT   im0   im1
Grove
(Synthetic)
GT   im0   im1
Urban
(Synthetic)
GT   im0   im1
Yosemite
(Synthetic)
GT   im0   im1
Teddy
(Stereo)
GT   im0   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
NNF-Local [87]10.1 7.47 3 40.1 11 3.98 3 6.49 13 30.3 3 5.60 17 5.82 4 26.1 3 4.68 17 3.86 10 53.5 7 2.99 20 9.77 1 12.4 1 5.36 3 8.67 5 31.8 3 7.03 3 5.10 23 10.1 58 3.70 12 2.31 10 5.34 11 1.21 2
NN-field [71]13.8 8.38 14 43.1 27 4.19 4 7.34 24 28.7 2 6.26 28 5.82 4 28.9 6 4.68 17 2.94 2 54.1 8 2.16 3 10.4 4 13.2 4 5.24 2 6.12 1 17.5 1 4.46 1 6.24 57 10.6 74 4.10 13 2.35 14 6.44 21 1.14 1
MDP-Flow2 [68]16.2 8.02 8 38.6 6 5.75 21 5.17 2 31.1 4 4.55 3 5.48 3 30.8 9 4.22 9 4.49 20 99.9 58 3.27 28 11.3 13 13.4 6 8.04 21 10.8 14 54.4 37 10.5 21 4.84 12 9.33 38 4.31 18 2.69 28 4.85 6 2.20 4
OFLAF [77]17.6 7.70 5 39.8 8 4.74 8 6.40 12 32.5 7 5.82 23 4.73 2 25.3 2 3.96 5 4.47 19 99.9 58 3.55 46 10.2 3 13.0 2 6.29 7 13.3 35 42.1 19 9.90 18 5.10 23 8.01 10 4.66 28 2.75 29 5.59 15 6.33 39
PMMST [114]18.8 8.63 17 31.3 1 6.03 27 8.51 38 26.8 1 8.18 63 7.50 12 28.0 5 6.07 37 4.26 17 34.8 5 3.29 29 10.9 8 13.2 4 6.26 6 10.4 12 29.9 2 9.42 13 5.00 19 10.1 58 4.37 19 3.25 40 4.40 5 3.36 13
nLayers [57]26.4 8.19 11 45.3 53 4.62 6 9.65 62 31.7 6 8.88 83 8.87 20 33.6 11 8.22 65 3.62 6 99.9 58 2.93 17 10.5 5 13.6 8 6.52 8 11.3 20 33.4 4 9.45 15 6.02 52 8.56 16 4.99 36 2.31 10 6.80 26 5.53 35
NNF-EAC [103]27.1 8.80 19 40.8 14 6.14 29 6.13 7 39.3 28 5.36 11 6.97 9 35.1 12 4.73 19 5.83 42 87.9 39 3.49 42 12.1 29 14.6 20 8.87 32 12.5 31 41.2 15 11.8 38 5.35 32 10.1 58 4.61 26 3.19 37 7.62 37 3.99 25
ComponentFusion [96]28.6 8.30 13 49.1 75 5.87 24 5.69 5 35.4 15 5.40 12 7.24 10 35.3 13 4.99 23 3.69 7 99.9 58 2.32 6 11.7 19 14.1 13 8.75 30 15.8 65 66.6 61 15.0 78 5.71 41 8.88 24 5.09 41 2.64 25 5.25 10 3.68 19
FC-2Layers-FF [74]30.6 8.26 12 41.4 18 6.27 30 8.85 42 37.9 22 7.81 47 6.02 7 31.8 10 6.25 40 3.91 12 88.8 42 2.86 15 11.0 11 13.7 10 7.20 10 16.5 73 40.4 11 16.3 92 7.14 77 10.7 78 6.61 66 2.15 3 3.87 1 2.77 6
FESL [72]31.2 7.69 4 40.2 12 4.90 9 11.0 84 48.5 57 9.07 86 10.5 29 42.1 26 6.42 42 3.60 5 99.9 58 2.55 9 10.9 8 13.6 8 8.86 31 11.1 18 36.4 6 10.6 22 6.73 71 10.2 62 5.95 54 2.51 22 5.37 13 3.35 12
LME [70]31.9 7.85 7 42.7 25 6.01 26 5.33 3 34.6 10 4.91 7 14.6 51 54.5 53 40.7 114 4.66 22 73.0 15 3.25 25 11.5 16 13.8 11 9.65 57 11.6 22 70.4 71 12.1 43 5.14 25 9.97 56 4.51 25 2.86 30 6.45 22 4.29 30
TC/T-Flow [76]32.3 9.01 22 38.1 4 3.81 1 6.64 15 55.1 78 4.62 4 8.13 15 46.4 39 4.20 8 5.32 35 99.9 58 2.88 16 11.5 16 14.1 13 7.28 12 8.85 6 38.5 9 9.44 14 5.85 48 10.7 78 10.0 109 3.61 48 10.0 54 8.53 73
HAST [109]33.3 6.42 1 43.9 36 3.97 2 7.16 19 33.1 8 5.92 25 3.76 1 23.5 1 2.83 1 3.36 3 99.9 58 2.08 2 10.0 2 13.0 2 4.83 1 16.7 78 59.3 44 19.4 112 11.3 122 12.9 109 17.6 133 2.67 27 4.13 3 2.93 9
WLIF-Flow [93]34.9 8.04 9 40.7 13 5.53 18 7.98 33 34.6 10 7.20 40 8.75 19 40.5 21 5.74 34 4.27 18 96.8 55 2.94 18 13.4 87 16.1 95 9.77 61 13.5 39 41.3 17 11.8 38 5.65 39 9.12 30 5.96 55 2.29 9 7.30 30 6.96 49
PMF [73]35.2 9.38 30 48.9 73 4.67 7 7.10 17 37.5 19 5.58 16 7.91 13 30.0 8 4.06 6 4.89 31 99.9 58 3.29 29 10.6 7 14.0 12 5.76 5 12.4 30 54.8 38 11.5 33 11.2 121 17.9 141 11.1 114 2.06 1 4.95 9 4.00 26
ALD-Flow [66]35.5 8.18 10 41.6 23 4.51 5 6.32 10 54.8 76 5.03 9 10.7 33 61.8 61 4.24 10 4.24 16 99.9 58 2.61 10 11.7 19 14.3 15 7.21 11 10.8 14 61.4 47 9.93 19 5.96 51 9.55 43 9.77 108 3.87 53 15.7 64 9.38 86
Layers++ [37]35.9 9.07 23 44.4 39 8.41 66 8.47 37 31.5 5 8.03 56 5.85 6 37.9 15 6.02 36 3.76 9 62.6 9 2.79 12 10.5 5 13.5 7 8.22 26 17.6 86 55.0 39 14.5 75 7.44 82 10.9 82 5.70 53 2.27 8 4.86 7 9.14 78
3DFlow [135]36.3 9.37 28 41.4 18 5.02 11 7.10 17 38.0 23 5.51 15 7.41 11 53.2 51 3.81 4 4.74 23 14.8 2 4.00 58 12.0 27 15.2 46 8.89 34 19.9 104 60.3 45 18.1 109 8.06 90 9.24 35 10.1 110 2.26 6 4.12 2 2.04 3
Efficient-NL [60]37.0 8.40 15 43.6 33 5.38 16 9.12 46 37.4 18 7.83 49 11.1 37 56.6 55 6.16 38 5.71 39 99.9 58 3.32 32 11.3 13 14.7 24 7.66 13 16.6 77 37.6 7 12.7 51 6.93 75 10.8 80 5.98 58 2.89 32 5.47 14 2.90 8
RNLOD-Flow [121]38.0 7.26 2 38.7 7 5.25 15 7.65 29 43.0 37 6.20 27 12.7 44 75.2 68 4.59 14 3.52 4 99.9 58 2.46 7 11.4 15 14.9 33 7.74 15 16.2 67 40.2 10 16.2 89 8.20 92 12.2 100 7.63 86 2.49 21 5.98 18 7.10 53
ProFlow_ROB [147]38.1 10.5 45 51.4 84 5.14 14 7.28 21 50.8 64 5.66 19 17.6 54 65.7 64 5.43 29 3.88 11 98.0 56 2.18 4 12.3 37 15.2 46 7.81 17 12.1 28 52.1 35 12.9 53 4.24 5 9.88 53 3.68 10 3.79 51 19.4 70 6.62 44
SVFilterOh [111]38.2 8.96 21 55.2 106 5.63 19 7.20 20 37.8 21 6.39 31 6.18 8 43.0 29 5.02 24 3.71 8 99.9 58 2.49 8 11.1 12 14.3 15 5.54 4 13.5 39 50.6 34 14.1 68 10.1 113 16.8 137 12.4 125 2.20 4 4.91 8 2.50 5
AGIF+OF [85]38.4 8.89 20 44.1 37 6.77 35 10.2 69 44.4 42 8.51 78 10.2 26 43.5 31 6.63 46 4.80 27 99.9 58 3.25 25 11.7 19 14.7 24 9.52 52 13.7 42 40.9 12 12.6 48 5.73 44 8.99 25 5.96 55 2.36 16 7.50 33 7.55 57
TC-Flow [46]39.0 8.59 16 41.0 16 5.12 12 5.47 4 45.6 46 4.31 2 10.2 26 94.7 84 3.49 2 6.08 48 99.9 58 3.50 43 11.9 25 14.5 18 7.90 19 11.9 27 61.7 49 11.5 33 5.72 42 9.85 51 11.5 119 4.02 57 15.0 62 9.14 78
IROF++ [58]40.9 9.29 27 45.0 51 6.30 31 9.55 58 43.0 37 8.20 64 10.8 34 43.1 30 7.57 57 6.28 53 99.9 58 3.90 53 12.2 33 14.8 29 9.41 50 15.2 61 44.1 28 14.5 75 5.27 27 9.48 41 3.69 11 2.62 24 6.48 23 4.14 27
PH-Flow [101]41.0 10.2 39 44.6 43 7.86 48 9.41 53 42.0 34 8.11 59 8.41 18 38.9 17 7.40 56 6.39 55 99.9 58 3.93 55 11.8 23 14.5 18 7.72 14 13.8 43 42.9 23 13.1 55 7.21 78 10.3 65 7.61 85 2.34 13 4.20 4 4.20 28
Correlation Flow [75]42.1 9.27 26 38.3 5 5.40 17 6.33 11 36.7 16 4.85 6 18.4 56 99.9 94 3.58 3 4.87 30 35.8 6 3.47 39 12.9 64 15.8 77 9.17 44 16.0 66 68.6 64 16.5 98 6.59 67 9.85 51 7.93 93 2.88 31 7.57 36 3.06 10
Classic+CPF [83]42.1 9.64 33 43.4 29 7.93 51 9.43 54 46.1 49 7.82 48 10.6 32 51.0 46 6.68 47 5.09 33 99.9 58 3.22 24 12.0 27 15.2 46 9.15 42 14.7 53 34.4 5 13.4 60 6.42 61 10.1 58 6.89 71 2.26 6 6.77 25 7.08 52
PWC-Net_ROB [148]43.2 17.5 98 49.3 78 9.65 89 9.55 58 42.8 35 8.41 69 19.6 58 46.1 38 10.6 72 6.18 52 74.3 16 3.39 35 12.4 40 15.4 58 8.92 35 11.8 25 45.5 29 11.5 33 4.32 7 9.79 47 2.49 3 3.08 34 6.38 20 2.86 7
JOF [141]44.2 8.67 18 45.9 55 6.43 32 10.3 72 48.5 57 8.94 85 8.37 17 39.4 19 7.19 51 4.81 28 99.9 58 2.79 12 11.9 25 14.6 20 8.03 20 13.9 46 43.0 24 10.8 23 9.32 106 11.4 90 11.3 118 2.21 5 7.25 28 7.07 51
CostFilter [40]44.9 10.5 45 46.8 64 6.98 39 7.51 26 38.1 24 6.31 30 9.22 21 29.7 7 4.74 20 5.86 44 99.9 58 3.97 57 10.9 8 14.3 15 6.77 9 13.4 36 56.1 40 12.3 45 11.6 125 20.5 145 14.2 127 2.12 2 8.52 44 6.71 46
OAR-Flow [125]46.6 11.1 53 48.8 71 5.85 22 9.88 64 82.9 119 6.47 32 27.3 77 99.9 94 8.06 61 6.75 61 99.9 58 2.80 14 12.4 40 15.1 44 8.20 25 10.3 10 58.1 42 8.37 9 4.07 2 8.06 12 5.45 50 4.81 66 9.74 53 6.36 40
ProbFlowFields [128]46.6 16.3 89 53.8 98 10.9 104 7.72 30 40.6 31 7.12 36 14.1 49 45.2 37 10.5 70 6.67 60 62.6 9 4.39 70 12.9 64 15.4 58 9.64 56 10.1 9 63.2 55 10.8 23 4.86 14 8.16 14 4.69 30 3.36 42 9.23 50 3.72 21
IIOF-NLDP [131]46.9 12.0 60 43.5 31 5.86 23 9.35 49 35.0 12 6.81 34 11.2 38 88.4 80 4.51 12 6.09 49 28.8 4 4.09 62 14.6 116 18.4 129 9.71 58 14.6 52 66.2 60 14.7 77 4.90 17 9.54 42 4.82 34 3.09 35 7.73 40 3.18 11
Sparse-NonSparse [56]47.6 9.96 34 44.2 38 8.85 74 9.39 52 50.6 63 8.08 58 10.1 25 43.7 33 7.21 52 6.10 50 88.2 40 3.41 37 12.5 46 15.5 68 8.96 36 16.3 69 41.9 18 16.2 89 6.48 63 9.05 27 6.27 62 2.33 12 7.33 31 7.76 65
MLDP_OF [89]48.1 11.8 59 41.7 24 8.40 65 6.97 16 35.2 14 5.88 24 11.3 41 65.3 63 5.23 27 4.76 26 99.9 58 3.09 21 12.2 33 14.8 29 8.87 32 13.4 36 48.1 31 17.2 105 10.0 112 10.9 82 18.1 134 3.58 47 8.07 43 4.53 33
LSM [39]48.4 10.0 37 42.9 26 8.48 67 9.36 51 49.6 61 7.99 53 10.5 29 43.6 32 6.80 48 5.80 40 88.6 41 3.38 34 12.5 46 15.4 58 9.03 38 16.5 73 42.3 21 16.3 92 6.94 76 9.84 48 6.71 67 2.42 19 7.96 42 7.72 62
Ramp [62]48.5 10.2 39 44.4 39 8.09 60 9.47 55 46.1 49 8.17 62 9.51 23 42.4 28 6.88 50 5.40 36 99.9 58 3.53 44 12.5 46 15.2 46 9.71 58 16.7 78 42.1 19 16.5 98 6.76 72 10.0 57 7.07 75 2.46 20 5.84 17 5.24 34
WRT [151]48.6 10.9 50 48.8 71 5.12 12 10.2 69 39.2 26 8.48 75 39.5 99 99.9 94 4.62 15 4.75 25 14.6 1 3.39 35 11.7 19 15.0 37 9.93 70 21.2 108 64.7 59 17.5 107 5.35 32 8.80 22 5.37 48 3.22 39 7.63 38 3.38 15
MDP-Flow [26]51.0 11.2 54 43.1 27 9.86 93 8.14 36 35.1 13 8.21 65 11.2 38 42.1 26 9.44 68 6.41 56 99.9 58 4.20 66 12.2 33 14.6 20 10.0 73 11.7 24 63.6 56 9.60 16 5.56 37 10.9 82 4.39 20 5.78 76 99.9 113 8.99 75
FMOF [94]51.9 9.17 24 43.6 33 8.04 57 10.0 67 48.1 54 8.48 75 8.35 16 38.3 16 6.49 44 5.08 32 99.9 58 3.45 38 12.6 52 15.4 58 9.19 45 18.1 92 41.2 15 15.5 83 6.67 70 10.6 74 7.47 82 3.00 33 16.1 65 7.74 63
Classic+NL [31]52.0 10.1 38 44.9 49 8.90 75 9.49 56 51.6 67 7.87 50 9.93 24 43.9 35 7.31 55 6.07 47 99.9 58 3.78 51 12.5 46 15.3 54 9.06 39 17.1 82 41.0 13 15.8 86 7.32 81 10.8 80 6.80 70 2.35 14 5.62 16 7.69 61
IROF-TV [53]52.4 10.4 43 44.5 42 8.16 61 9.69 63 51.1 66 8.44 70 12.6 43 46.8 40 7.27 54 6.80 62 87.5 38 3.93 55 13.0 70 15.7 73 10.4 78 18.3 94 86.9 106 13.7 64 4.44 8 7.40 5 3.05 7 2.60 23 7.55 35 7.56 58
TV-L1-MCT [64]52.8 9.57 31 44.7 45 8.66 70 10.9 83 48.1 54 9.11 87 11.8 42 58.1 57 6.61 45 4.74 23 99.9 58 3.34 33 12.9 64 15.2 46 9.89 66 17.8 90 47.8 30 16.0 88 5.28 28 8.09 13 7.71 87 3.33 41 7.26 29 7.53 56
CombBMOF [113]53.0 12.5 65 41.3 17 6.50 34 8.58 40 36.9 17 6.88 35 10.9 35 35.9 14 5.21 26 10.4 81 85.0 36 5.90 91 11.6 18 15.2 46 8.15 23 27.3 119 60.9 46 35.6 133 9.20 103 14.7 125 6.75 69 3.17 36 7.53 34 4.20 28
OFH [38]53.2 12.6 66 43.4 29 9.45 83 7.30 23 64.4 88 5.27 10 27.6 78 99.9 94 4.87 22 6.60 59 99.9 58 3.74 49 12.4 40 14.7 24 9.62 54 15.5 63 74.1 79 15.6 85 4.60 9 9.39 39 4.64 27 5.39 72 26.0 80 6.68 45
NL-TV-NCC [25]53.6 10.7 48 40.8 14 6.45 33 8.52 39 41.1 32 6.30 29 11.2 38 93.6 83 4.18 7 5.99 46 75.9 20 4.02 59 13.2 77 16.2 101 10.1 75 16.7 78 70.9 73 16.3 92 6.56 66 9.91 55 7.05 74 4.76 64 16.9 66 3.56 17
COFM [59]53.8 9.37 28 55.5 108 6.86 37 7.28 21 44.2 41 6.17 26 14.3 50 47.6 43 8.22 65 4.15 15 99.9 58 2.23 5 13.2 77 16.2 101 12.2 106 17.6 86 75.4 82 15.5 83 6.20 56 8.77 21 7.35 79 3.62 49 5.35 12 6.43 42
Adaptive [20]54.5 10.2 39 46.1 57 4.95 10 9.63 60 55.4 79 7.80 46 36.7 96 99.9 94 7.64 58 6.15 51 78.7 22 2.96 19 12.1 29 14.8 29 9.09 40 12.3 29 85.8 104 6.06 2 8.72 97 12.5 105 4.97 35 3.55 46 34.8 84 9.13 77
AggregFlow [97]55.9 13.2 68 62.1 130 6.79 36 14.9 96 73.1 98 10.6 95 26.8 76 55.1 54 20.5 97 5.48 37 99.9 58 3.67 47 12.5 46 15.0 37 7.76 16 8.55 4 38.2 8 8.90 11 5.78 45 10.6 74 4.74 32 5.43 73 8.53 45 7.58 59
S2F-IF [123]56.0 20.0 102 51.4 84 9.91 94 9.64 61 48.3 56 7.93 51 19.7 59 41.7 24 13.6 85 9.98 78 84.3 32 5.40 84 12.8 60 15.3 54 9.92 69 10.9 17 62.4 53 10.9 27 5.00 19 10.4 69 5.30 47 3.71 50 8.55 46 3.87 23
FlowFields+ [130]56.7 20.3 103 52.0 88 10.3 100 10.3 72 44.8 44 8.44 70 19.5 57 40.2 20 14.1 86 10.2 80 66.6 12 6.28 96 12.8 60 15.4 58 9.97 71 10.3 10 61.8 50 10.2 20 4.85 13 10.9 82 4.79 33 3.97 55 12.8 58 3.85 22
Sparse Occlusion [54]57.5 9.98 36 41.5 22 7.82 47 9.00 45 40.5 30 8.28 67 13.5 47 85.5 77 5.96 35 5.82 41 99.9 58 3.90 53 13.0 70 15.9 79 9.77 61 13.8 43 49.9 33 12.3 45 13.6 137 15.7 133 7.81 90 3.51 43 9.05 48 6.42 41
S2D-Matching [84]57.7 9.96 34 53.0 94 8.51 68 9.53 57 53.0 69 7.94 52 20.5 61 99.9 94 6.80 48 5.30 34 83.0 24 3.53 44 12.4 40 15.2 46 9.16 43 17.3 83 41.1 14 16.8 103 7.75 86 10.5 72 7.90 92 2.36 16 6.34 19 9.58 91
Complementary OF [21]57.8 13.6 72 46.2 58 9.35 80 6.20 8 50.4 62 4.92 8 12.8 45 58.8 59 5.45 30 7.89 69 99.9 58 5.59 86 12.3 37 14.6 20 9.99 72 18.9 99 69.9 68 14.3 72 5.44 34 7.80 7 7.78 89 6.13 80 26.9 81 9.66 94
FlowFields [110]57.9 20.3 103 52.3 90 10.2 96 10.2 69 49.0 59 8.46 73 20.3 60 40.5 21 14.7 87 10.8 85 76.6 21 6.16 94 12.9 64 15.4 58 10.0 73 11.1 18 69.9 68 11.0 29 4.97 18 8.63 18 5.12 43 4.04 59 14.0 59 3.98 24
RFlow [90]58.0 11.5 57 45.3 53 8.80 73 6.22 9 49.2 60 5.41 13 26.2 74 99.9 94 5.04 25 4.14 14 99.9 58 3.11 22 12.6 52 15.0 37 9.87 64 16.5 73 83.2 99 13.8 67 6.62 68 8.58 17 6.16 59 6.33 83 99.9 113 12.0 107
HBM-GC [105]58.8 10.9 50 57.5 118 7.03 41 9.35 49 40.2 29 8.80 81 8.09 14 52.3 49 6.42 42 6.91 64 84.3 32 6.17 95 11.8 23 14.7 24 8.40 27 14.7 53 43.2 25 12.6 48 9.81 111 17.8 140 8.50 99 3.54 45 10.1 55 10.1 98
2DHMM-SAS [92]59.1 10.3 42 44.8 47 8.03 56 10.5 78 52.4 68 8.21 65 21.6 63 97.4 89 8.20 64 6.88 63 99.9 58 3.86 52 12.4 40 15.0 37 9.87 64 17.7 88 43.3 26 15.9 87 6.81 73 10.2 62 7.15 78 2.65 26 7.68 39 7.29 54
Occlusion-TV-L1 [63]59.8 10.4 43 44.9 49 6.90 38 8.77 41 53.3 71 7.54 42 33.8 91 99.9 94 7.96 59 5.88 45 99.9 58 3.48 41 13.6 96 16.3 104 10.6 82 9.50 7 80.1 89 8.60 10 6.12 54 8.69 19 4.39 20 6.52 86 99.9 113 9.37 83
SimpleFlow [49]60.3 11.3 56 46.6 60 9.79 92 10.7 81 45.0 45 9.15 88 23.1 66 99.9 94 8.38 67 8.00 71 99.9 58 3.72 48 12.7 57 15.5 68 9.36 49 16.3 69 42.6 22 15.3 82 5.91 50 9.61 44 5.39 49 2.39 18 7.08 27 9.41 87
Aniso-Texture [82]60.7 7.75 6 40.0 10 5.87 24 7.36 25 41.5 33 7.19 39 41.3 106 99.9 94 6.22 39 2.73 1 65.4 11 1.92 1 12.9 64 15.3 54 10.1 75 29.2 121 99.9 120 16.5 98 11.8 128 14.9 129 8.22 96 3.80 52 12.6 57 8.58 74
ACK-Prior [27]61.5 10.7 48 37.9 3 7.90 50 6.01 6 38.5 25 4.80 5 10.2 26 41.5 23 4.35 11 4.56 21 99.9 58 3.75 50 13.2 77 15.9 79 11.3 93 27.3 119 82.2 92 23.1 120 11.6 125 14.9 129 16.2 131 6.43 84 15.5 63 6.11 37
PGM-C [120]61.5 20.6 106 54.2 101 9.59 87 10.1 68 60.8 83 8.47 74 22.3 65 44.3 36 14.8 89 10.7 84 99.9 58 4.15 64 13.1 75 15.4 58 9.90 67 11.8 25 64.6 58 12.0 42 4.86 14 7.96 8 5.01 38 4.78 65 14.4 60 7.01 50
EPPM w/o HM [88]61.5 15.3 84 41.4 18 8.08 59 7.60 27 33.9 9 5.66 19 13.0 46 47.0 41 5.57 31 8.73 75 99.9 58 4.81 79 12.6 52 15.7 73 10.8 86 18.6 96 62.9 54 16.4 96 11.9 131 12.5 105 17.2 132 3.20 38 7.49 32 6.00 36
ROF-ND [107]64.0 12.0 60 39.9 9 8.22 64 6.49 13 45.6 46 5.49 14 13.5 47 92.7 81 4.83 21 8.05 72 23.7 3 5.54 85 14.2 110 17.5 118 11.2 91 20.1 106 72.0 75 15.0 78 13.0 135 13.3 115 10.5 113 3.52 44 6.67 24 3.36 13
ComplOF-FED-GPU [35]64.9 13.2 68 44.7 45 7.95 52 9.18 47 82.6 118 5.63 18 15.3 52 58.5 58 5.67 32 7.59 67 99.9 58 4.68 75 12.3 37 14.8 29 9.20 46 18.2 93 83.8 100 16.4 96 7.54 84 9.84 48 11.1 114 5.44 74 31.6 83 7.74 63
CPM-Flow [116]66.2 20.6 106 54.3 102 9.58 85 10.3 72 62.5 86 8.50 77 21.9 64 43.8 34 14.7 87 10.6 83 99.9 58 4.06 60 13.2 77 15.4 58 9.85 63 12.6 32 68.9 65 13.3 57 5.02 21 9.16 33 5.04 40 5.27 69 19.2 68 9.63 93
EpicFlow [102]67.0 20.6 106 54.1 100 9.59 87 10.3 72 62.9 87 8.55 79 26.3 75 99.4 93 15.1 91 10.4 81 99.9 58 4.07 61 13.1 75 15.4 58 9.90 67 11.6 22 67.3 62 11.9 41 4.86 14 7.97 9 4.99 36 5.34 71 19.2 68 9.74 96
DMF_ROB [140]67.4 16.2 88 49.6 80 9.96 95 9.93 66 75.9 102 7.56 43 30.5 84 99.9 94 11.8 77 16.5 101 99.9 58 4.36 69 12.6 52 14.9 33 9.62 54 12.9 34 76.3 84 11.5 33 4.68 10 8.80 22 5.10 42 7.02 94 91.0 110 9.62 92
SRR-TVOF-NL [91]67.8 14.4 80 46.7 61 8.18 63 13.1 90 74.0 100 8.44 70 24.1 70 63.2 62 11.9 79 6.51 58 85.1 37 3.25 25 12.1 29 15.0 37 10.3 77 17.5 84 61.6 48 13.4 60 10.4 117 12.3 102 8.92 102 5.52 75 7.83 41 7.58 59
DeepFlow2 [108]68.2 14.3 77 47.4 67 7.03 41 10.4 76 77.1 105 8.01 54 23.3 67 99.9 94 11.8 77 16.1 97 99.9 58 4.41 71 12.4 40 15.0 37 8.16 24 13.4 36 68.4 63 14.2 69 5.65 39 9.07 29 8.50 99 8.52 102 92.9 112 10.6 102
Steered-L1 [118]68.2 9.19 25 36.3 2 6.06 28 4.59 1 39.2 26 4.30 1 9.30 22 52.3 49 4.57 13 4.86 29 99.9 58 3.30 31 13.6 96 15.9 79 12.3 107 24.7 118 77.0 85 20.2 113 15.1 140 13.7 120 40.0 142 14.7 119 91.5 111 20.9 121
TCOF [69]69.3 13.6 72 44.8 47 8.02 55 9.90 65 54.6 75 8.02 55 31.3 87 99.9 94 15.4 93 6.49 57 82.4 23 4.76 77 14.9 120 18.0 126 9.50 51 9.71 8 48.3 32 12.6 48 10.1 113 12.7 107 8.96 103 4.29 60 9.21 49 6.80 47
DPOF [18]70.5 17.4 97 49.1 75 7.77 46 12.3 86 45.6 46 8.91 84 10.9 35 26.6 4 8.09 62 7.81 68 99.3 57 5.31 81 13.5 92 16.0 90 11.0 89 17.5 84 61.8 50 12.2 44 13.1 136 10.9 82 18.1 134 5.04 67 9.50 52 4.49 31
F-TV-L1 [15]71.1 15.6 85 47.4 67 13.4 110 18.8 106 99.1 128 11.6 96 43.1 110 99.9 94 11.3 76 14.7 94 99.9 58 7.03 100 12.2 33 14.9 33 9.00 37 13.5 39 99.9 120 7.56 6 6.41 60 10.5 72 4.23 16 3.91 54 80.3 98 3.38 15
TF+OM [100]71.8 12.1 62 51.3 83 7.13 43 8.92 44 44.4 42 8.13 61 33.8 91 54.4 52 45.8 117 6.28 53 90.1 44 4.63 74 12.7 57 15.2 46 10.6 82 18.9 99 99.9 120 11.3 31 7.81 87 14.2 122 6.33 63 7.09 95 43.5 86 8.22 69
Aniso. Huber-L1 [22]71.9 11.7 58 43.8 35 8.16 61 13.6 92 66.3 90 12.0 98 35.9 94 99.9 94 10.5 70 10.0 79 72.9 14 5.00 80 13.4 87 16.3 104 9.61 53 15.1 59 63.7 57 7.96 7 8.96 100 11.6 94 7.95 94 4.02 57 26.9 81 7.97 67
LiteFlowNet [143]72.9 24.0 120 51.5 86 13.3 109 11.7 85 43.1 39 9.98 91 21.5 62 47.6 43 13.0 81 10.9 87 71.9 13 6.87 99 13.3 85 16.2 101 11.5 96 18.7 97 52.2 36 16.7 101 6.18 55 9.32 37 4.21 15 5.83 78 24.4 78 7.46 55
FF++_ROB [146]73.5 21.4 112 55.5 108 10.2 96 10.6 80 54.1 73 8.70 80 24.1 70 59.5 60 16.2 95 11.5 89 74.8 18 7.42 103 13.0 70 15.7 73 10.6 82 14.7 53 70.0 70 13.7 64 5.31 31 8.76 20 7.53 84 4.57 62 14.8 61 13.9 111
TV-L1-improved [17]75.2 10.9 50 45.2 52 7.42 44 8.12 34 54.0 72 6.79 33 36.5 95 99.9 94 7.26 53 5.84 43 99.9 58 3.15 23 13.2 77 15.9 79 9.11 41 22.1 112 99.9 120 20.8 114 9.59 109 13.3 115 9.04 104 6.19 81 88.8 106 9.71 95
SIOF [67]76.2 10.6 47 49.7 81 7.01 40 14.8 95 85.9 121 8.40 68 49.7 119 98.3 90 49.2 120 12.0 90 99.9 58 5.88 89 13.5 92 15.9 79 10.8 86 16.3 69 74.2 80 13.6 63 5.51 36 9.02 26 4.42 23 6.52 86 19.5 73 9.85 97
CRTflow [80]77.5 15.0 82 46.3 59 7.89 49 8.87 43 54.9 77 7.15 37 30.1 81 99.9 94 8.03 60 9.30 77 99.9 58 4.50 73 13.0 70 15.7 73 8.05 22 32.5 125 99.9 120 34.3 132 6.62 68 9.72 46 7.52 83 9.30 106 99.9 113 14.7 113
DeepFlow [86]77.8 14.7 81 49.0 74 9.78 91 12.9 88 79.3 109 9.80 90 30.1 81 96.1 87 24.4 103 21.4 109 99.9 58 5.36 83 12.5 46 15.1 44 8.59 28 14.0 47 71.9 74 15.1 81 5.46 35 8.01 10 8.73 101 14.2 117 99.9 113 15.7 117
Brox et al. [5]78.1 16.0 86 49.2 77 12.0 106 12.3 86 80.4 110 10.3 93 23.7 69 73.1 67 13.2 82 24.2 111 99.9 58 4.23 67 14.7 117 16.8 111 15.4 130 10.7 13 96.7 113 9.71 17 5.88 49 9.05 27 3.01 6 8.78 103 67.7 94 9.37 83
LocallyOriented [52]79.1 17.0 96 55.7 110 8.00 54 17.0 103 82.3 117 12.1 99 42.4 109 99.9 94 14.8 89 9.13 76 89.2 43 4.79 78 13.4 87 16.1 95 9.20 46 10.8 14 58.1 42 11.8 38 6.89 74 10.6 74 7.14 77 7.78 100 74.9 95 9.42 88
BriefMatch [124]80.3 9.63 32 44.4 39 5.74 20 7.64 28 51.0 65 5.70 21 10.5 29 39.2 18 4.63 16 3.95 13 99.9 58 2.72 11 16.2 131 17.6 122 33.0 143 41.4 137 99.4 119 43.4 139 12.7 134 13.2 113 67.6 146 79.5 137 99.9 113 99.9 144
Dynamic MRF [7]81.2 14.0 76 50.7 82 9.58 85 7.75 31 85.7 120 5.76 22 31.5 88 99.9 94 5.23 27 7.97 70 99.9 58 4.10 63 13.0 70 15.6 71 10.7 85 30.4 124 99.9 120 29.5 129 5.64 38 7.52 6 9.61 107 67.3 135 99.9 113 66.7 135
Classic++ [32]81.9 11.2 54 49.4 79 9.13 76 9.34 48 68.4 93 8.11 59 30.7 85 95.1 86 10.2 69 5.59 38 99.9 58 3.47 39 13.5 92 16.1 95 10.4 78 19.7 103 99.9 120 17.6 108 8.38 93 11.5 93 8.30 98 7.20 98 99.9 113 9.54 90
ContinualFlow_ROB [153]83.2 30.9 128 63.0 132 13.9 112 22.3 111 47.9 53 20.9 114 34.2 93 99.9 94 31.9 108 13.3 91 99.9 58 7.83 106 13.7 101 16.5 108 11.4 94 21.5 109 70.6 72 21.3 117 4.20 4 10.3 65 2.80 4 4.01 56 8.59 47 3.70 20
ResPWCR_ROB [145]84.0 23.4 118 46.7 61 16.7 121 14.7 94 42.9 36 13.2 101 23.4 68 48.7 45 22.0 100 16.4 100 84.5 34 11.3 116 13.2 77 15.8 77 12.6 112 15.2 61 72.1 76 16.3 92 9.13 102 12.3 102 7.37 80 7.09 95 17.6 67 9.31 82
Rannacher [23]84.2 13.8 75 47.5 69 10.8 103 10.5 78 62.0 85 8.84 82 41.1 105 99.9 94 11.0 74 8.49 73 99.9 58 4.28 68 13.5 92 16.1 95 9.72 60 22.5 114 99.9 120 17.0 104 7.66 85 9.88 53 7.82 91 4.72 63 75.1 96 9.37 83
SuperFlow [81]84.5 14.3 77 47.0 66 9.26 78 19.5 109 58.7 82 17.9 110 45.9 113 99.9 94 56.1 123 19.0 107 99.9 58 5.88 89 13.3 85 16.1 95 12.6 112 11.5 21 74.0 78 8.27 8 9.24 104 12.9 109 4.70 31 8.27 101 89.2 108 8.28 70
CBF [12]85.0 12.2 64 41.4 18 8.65 69 16.5 100 47.1 52 16.6 108 24.7 72 88.1 79 12.9 80 11.0 88 99.9 58 4.18 65 14.9 120 17.5 118 14.0 125 15.0 58 79.8 88 8.97 12 14.9 139 15.1 131 15.9 130 5.78 76 63.0 93 10.1 98
AugFNG_ROB [144]85.5 34.3 134 65.6 136 20.9 127 25.8 115 66.4 91 24.3 118 40.9 103 99.9 94 40.6 113 15.2 96 96.0 52 9.61 109 13.4 87 15.6 71 12.3 107 15.1 59 73.5 77 14.2 69 5.29 29 11.2 88 2.87 5 5.30 70 19.4 70 4.51 32
p-harmonic [29]85.8 15.1 83 48.5 70 14.1 113 10.4 76 53.1 70 9.31 89 41.9 107 99.9 94 15.1 91 19.4 108 99.9 58 10.6 113 12.7 57 14.9 33 11.8 98 18.0 91 85.6 103 18.4 111 7.85 88 10.4 69 5.46 51 6.98 93 99.9 113 9.28 81
Local-TV-L1 [65]86.4 16.5 91 52.3 90 11.7 105 27.7 117 96.9 125 22.5 116 68.8 127 99.9 94 47.5 118 34.6 121 99.9 58 7.04 101 12.6 52 15.0 37 9.25 48 17.7 88 84.7 101 13.7 64 5.09 22 7.36 4 5.03 39 20.7 122 88.8 106 29.4 127
CLG-TV [48]86.8 12.1 62 43.5 31 9.42 82 14.1 93 60.9 84 13.2 101 33.2 90 99.9 94 11.2 75 10.8 85 84.7 35 5.82 88 14.8 119 17.9 124 12.1 104 13.8 43 99.9 120 11.3 31 10.9 119 14.2 122 9.22 105 6.69 90 99.9 113 8.52 72
TriFlow [95]87.5 16.1 87 57.1 116 7.97 53 13.3 91 65.2 89 12.3 100 48.8 116 99.9 94 61.8 127 7.03 65 91.4 46 5.33 82 13.4 87 15.3 54 11.4 94 14.3 51 76.1 83 13.3 57 22.7 145 14.3 124 26.7 138 5.93 79 11.8 56 7.86 66
FlowNet2 [122]87.8 32.0 131 68.0 142 14.9 115 35.2 122 77.9 108 29.8 127 32.9 89 51.2 47 35.4 111 16.3 98 99.9 58 10.5 112 13.2 77 15.9 79 12.0 100 14.1 48 99.9 120 10.8 23 8.84 99 20.1 142 4.66 28 4.49 61 9.46 51 3.65 18
OFRF [134]88.2 13.1 67 58.6 122 9.77 90 49.6 135 99.9 129 43.7 136 49.2 117 99.9 94 32.0 109 16.3 98 96.4 53 10.0 110 12.1 29 14.7 24 7.87 18 14.9 57 43.6 27 13.3 57 8.66 96 12.0 99 11.8 122 21.9 125 19.7 74 36.6 129
EAI-Flow [152]88.5 25.2 121 57.0 115 16.6 120 19.3 108 76.7 103 14.1 105 29.6 80 51.5 48 22.8 101 23.6 110 93.8 49 10.3 111 12.8 60 16.0 90 10.4 78 15.6 64 74.4 81 14.4 74 12.6 133 11.3 89 6.90 72 5.08 68 20.5 75 8.20 68
DF-Auto [115]88.8 20.5 105 55.9 112 9.21 77 22.7 112 74.8 101 19.3 113 44.7 112 93.5 82 57.8 125 27.1 115 99.9 58 5.70 87 15.0 123 18.9 136 11.8 98 7.13 2 57.9 41 7.09 4 10.2 116 13.9 121 4.23 16 8.94 104 54.0 91 9.21 80
SegOF [10]89.1 22.8 117 54.8 103 15.4 117 27.9 118 56.0 80 27.4 122 39.3 98 87.9 78 33.2 110 37.5 122 75.4 19 22.3 122 14.4 114 16.3 104 14.7 128 21.7 110 99.9 120 24.5 122 4.09 3 7.28 2 2.18 2 6.79 91 48.3 89 6.93 48
EPMNet [133]89.3 29.4 127 62.1 130 16.4 119 36.2 123 95.3 124 29.0 126 30.2 83 47.4 42 30.9 107 18.3 104 99.9 58 11.1 114 13.2 77 15.9 79 12.0 100 14.1 48 99.9 120 10.8 23 8.19 91 16.9 138 4.18 14 6.63 89 19.4 70 6.18 38
TriangleFlow [30]90.0 13.2 68 46.8 64 9.41 81 10.8 82 73.2 99 7.30 41 26.1 73 99.9 94 5.70 33 7.23 66 99.9 58 4.46 72 17.0 136 21.3 141 15.2 129 23.0 115 69.8 67 22.9 119 9.71 110 16.1 134 9.40 106 6.89 92 23.8 76 11.5 105
Bartels [41]90.3 13.3 71 55.0 104 10.2 96 8.13 35 43.2 40 7.67 44 18.1 55 69.0 66 6.30 41 8.49 73 99.9 58 6.05 93 13.9 104 16.1 95 13.9 123 21.8 111 99.9 120 21.5 118 10.6 118 13.5 117 20.3 136 12.3 112 99.9 113 26.9 125
WOLF_ROB [149]91.2 21.5 113 55.2 106 10.2 96 37.6 125 99.9 129 17.3 109 54.7 122 99.9 94 23.9 102 25.1 112 99.9 58 12.8 117 12.8 60 15.4 58 11.2 91 18.5 95 61.8 50 17.2 105 5.80 47 9.26 36 6.20 61 11.4 110 24.3 77 15.7 117
Fusion [6]91.4 16.3 89 53.8 98 12.5 107 7.93 32 37.7 20 7.75 45 15.6 53 41.8 25 13.2 82 13.5 92 83.1 25 7.77 105 15.4 125 18.5 130 14.2 127 33.1 126 89.0 107 24.8 124 11.8 128 14.7 125 8.27 97 11.4 110 99.9 113 13.3 109
LFNet_ROB [150]94.0 28.9 126 53.4 96 17.2 122 15.3 98 46.9 51 13.7 104 31.0 86 96.3 88 21.7 99 18.4 105 92.1 47 14.4 119 13.7 101 16.6 109 12.0 100 18.8 98 90.4 109 16.7 101 6.48 63 10.9 82 5.97 57 6.22 82 99.9 113 10.3 100
CNN-flow-warp+ref [117]94.2 21.3 111 57.3 117 15.1 116 16.8 102 54.4 74 15.9 107 41.0 104 99.9 94 28.8 104 28.6 117 99.9 58 7.45 104 14.0 107 15.9 79 14.0 125 16.5 73 84.8 102 10.9 27 5.30 30 8.23 15 8.13 95 99.9 144 99.9 113 99.9 144
StereoFlow [44]95.5 48.0 143 74.6 152 41.1 144 61.0 140 99.9 129 51.6 139 71.4 128 99.9 94 63.9 129 65.6 137 99.9 58 61.2 137 16.2 131 15.9 79 22.6 137 7.22 3 77.8 87 7.39 5 3.38 1 7.35 3 1.99 1 7.18 97 99.9 113 11.4 104
LDOF [28]96.0 17.9 100 53.5 97 8.72 71 18.7 105 92.6 123 11.8 97 29.5 79 67.1 65 20.9 98 29.0 118 99.9 58 8.92 108 14.2 110 16.4 107 13.7 120 18.9 99 97.5 117 15.0 78 6.26 59 10.3 65 10.1 110 10.4 108 99.9 113 10.3 100
StereoOF-V1MT [119]96.8 16.5 91 46.0 56 9.27 79 18.9 107 99.9 129 7.16 38 40.5 101 99.9 94 8.18 63 14.7 94 96.4 53 6.30 97 14.1 109 16.8 111 12.9 115 29.9 122 91.4 110 27.0 127 5.78 45 10.4 69 10.4 112 99.9 144 99.9 113 99.9 144
Shiralkar [42]97.4 16.8 94 44.6 43 9.46 84 16.5 100 98.8 127 8.05 57 42.0 108 99.9 94 10.8 73 18.4 105 99.9 58 8.02 107 12.9 64 15.5 68 10.4 78 30.2 123 99.9 120 25.1 125 11.4 124 11.8 97 15.8 129 22.2 126 99.9 113 17.5 120
FlowNetS+ft+v [112]97.4 16.5 91 52.1 89 8.06 58 17.4 104 76.9 104 13.4 103 46.3 114 99.9 94 30.4 106 29.8 119 99.9 58 13.8 118 15.5 126 18.5 130 13.8 122 12.7 33 89.5 108 11.7 37 9.24 104 13.5 117 12.1 123 7.39 99 57.9 92 9.52 89
Learning Flow [11]98.1 13.7 74 52.8 92 7.67 45 12.9 88 87.1 122 10.0 92 40.5 101 95.0 85 13.4 84 38.1 123 99.9 58 4.74 76 17.1 138 21.7 142 12.5 111 24.2 117 99.9 120 13.5 62 7.95 89 12.7 107 6.98 73 23.9 128 99.9 113 14.9 114
Ad-TV-NDC [36]99.5 31.0 129 53.3 95 33.1 140 70.2 141 99.9 129 49.0 138 93.2 145 99.9 94 54.0 122 38.9 124 95.0 50 29.4 125 13.8 103 17.3 117 8.71 29 14.7 53 77.1 86 13.0 54 6.24 57 9.84 48 5.19 46 46.4 133 76.4 97 54.0 133
Second-order prior [8]100.0 14.3 77 46.7 61 8.79 72 15.2 97 72.5 97 10.5 94 39.2 97 99.9 94 16.6 96 17.5 102 99.9 58 6.01 92 14.4 114 17.5 118 10.8 86 38.6 135 99.9 120 24.7 123 11.3 122 12.2 100 11.2 116 9.13 105 89.5 109 15.6 116
Filter Flow [19]100.4 21.6 115 57.7 119 14.4 114 24.6 114 77.5 107 18.1 111 54.3 121 80.8 69 66.3 133 52.8 130 91.0 45 46.5 131 13.6 96 16.0 90 12.3 107 17.0 81 69.6 66 14.2 69 12.0 132 16.1 134 7.39 81 6.58 88 37.5 85 8.36 71
TVL1_ROB [139]101.7 31.2 130 59.9 127 29.1 136 48.6 134 99.9 129 41.9 134 91.6 144 99.9 94 71.5 136 52.0 129 99.9 58 44.1 130 14.2 110 16.9 113 12.0 100 14.2 50 99.9 120 13.2 56 4.83 11 9.14 32 3.56 9 21.0 123 99.9 113 22.5 123
GraphCuts [14]103.2 21.7 116 52.8 92 10.4 101 39.2 127 99.9 129 23.1 117 39.7 100 58.0 56 49.6 121 25.6 114 74.6 17 7.31 102 13.6 96 15.9 79 13.2 118 37.8 132 97.6 118 16.2 89 9.36 107 11.4 90 11.7 120 10.3 107 99.9 113 15.0 115
2D-CLG [1]104.2 46.1 141 67.5 141 28.2 135 39.5 128 77.3 106 38.9 130 93.9 147 99.9 94 74.9 140 53.6 131 99.9 58 51.0 133 13.9 104 15.9 79 13.5 119 24.0 116 99.9 120 21.2 115 4.28 6 7.24 1 4.50 24 12.7 113 99.9 113 11.6 106
HBpMotionGpu [43]104.5 19.5 101 63.6 133 13.5 111 31.0 119 99.9 129 27.4 122 99.9 150 99.9 94 59.3 126 18.2 103 99.9 58 6.69 98 13.6 96 16.0 90 12.4 110 16.2 67 91.5 111 11.1 30 11.1 120 13.0 112 6.73 68 21.3 124 99.9 113 21.6 122
IAOF2 [51]104.9 16.8 94 59.5 125 10.5 102 20.1 110 69.0 95 18.1 111 53.3 120 99.9 94 56.8 124 55.3 134 95.0 50 54.7 135 14.2 110 17.2 116 11.0 89 19.2 102 81.1 90 14.3 72 11.8 128 13.2 113 13.0 126 13.5 114 45.0 87 9.00 76
SPSA-learn [13]105.5 23.6 119 55.7 110 20.1 126 32.9 121 99.9 129 25.2 119 91.2 143 99.9 94 64.5 131 49.6 127 99.9 58 31.2 126 14.0 107 16.0 90 13.0 117 19.9 104 99.9 120 23.3 121 6.53 65 9.13 31 4.40 22 15.8 120 99.9 113 16.5 119
Modified CLG [34]107.3 27.9 124 58.6 122 23.4 130 26.5 116 71.1 96 26.2 121 93.4 146 99.9 94 73.1 138 49.2 126 99.9 58 22.6 123 15.2 124 18.1 127 13.7 120 16.3 69 99.9 120 12.8 52 6.43 62 10.2 62 11.7 120 11.3 109 99.9 113 11.3 103
UnFlow [129]108.8 49.8 144 69.9 149 23.3 129 32.8 120 57.8 81 33.0 128 49.4 118 98.6 91 39.8 112 53.6 131 99.9 58 52.1 134 16.2 131 18.5 130 20.6 134 35.3 129 99.9 120 38.4 135 8.81 98 11.4 90 3.08 8 6.46 85 99.9 113 6.47 43
IAOF [50]108.9 20.6 106 55.0 104 17.4 123 36.6 124 99.9 129 27.6 124 99.9 150 99.9 94 75.5 141 32.7 120 93.3 48 25.5 124 13.9 104 16.6 109 12.1 104 36.5 130 92.3 112 12.3 45 9.40 108 11.7 95 7.13 76 26.2 129 47.4 88 28.8 126
GroupFlow [9]111.6 27.3 123 66.8 139 21.6 128 41.1 129 99.9 129 35.1 129 71.4 128 99.9 94 61.8 127 25.1 112 99.9 58 14.4 119 14.7 117 17.9 124 11.7 97 40.6 136 97.2 116 40.6 137 5.72 42 11.7 95 6.48 64 16.4 121 53.8 90 23.0 124
Black & Anandan [4]111.9 21.5 113 51.7 87 19.8 125 38.6 126 99.9 129 25.8 120 81.3 132 99.9 94 65.5 132 50.4 128 99.9 58 31.6 127 15.5 126 19.1 137 12.8 114 22.4 113 96.8 115 18.2 110 10.1 113 12.4 104 5.16 45 13.9 116 99.9 113 12.9 108
Nguyen [33]112.5 27.2 122 56.2 113 18.7 124 46.7 133 99.9 129 44.1 137 97.7 149 99.9 94 74.1 139 45.7 125 99.9 58 38.0 128 16.4 134 17.7 123 21.8 136 20.8 107 99.9 120 21.2 115 7.21 78 9.42 40 5.61 52 14.6 118 99.9 113 14.2 112
2bit-BM-tele [98]115.5 20.8 110 59.0 124 16.3 118 16.4 99 68.5 94 15.3 106 43.9 111 99.9 94 15.4 93 14.6 93 99.9 58 11.2 115 15.6 128 17.5 118 16.8 132 34.9 128 99.9 120 31.1 130 18.8 143 20.4 144 30.8 139 23.7 127 99.9 113 61.9 134
BlockOverlap [61]117.4 17.6 99 56.5 114 13.1 108 24.0 113 66.9 92 21.5 115 67.6 126 99.9 94 49.0 119 28.2 116 99.9 58 15.2 121 17.6 140 18.3 128 32.7 142 38.1 133 81.6 91 26.6 126 14.5 138 15.5 132 67.6 146 39.8 131 82.5 99 84.7 136
Heeger++ [104]119.4 33.0 132 58.3 121 23.6 131 60.7 138 99.9 129 39.4 131 59.1 123 99.9 94 30.1 105 86.5 150 99.9 58 70.5 139 14.9 120 17.1 114 12.9 115 76.5 144 99.9 120 79.8 144 7.49 83 12.9 109 6.59 65 99.9 144 99.9 113 99.9 144
H+S_ROB [138]120.9 50.3 145 65.4 135 33.8 141 46.6 132 99.9 129 43.6 135 78.1 130 99.9 94 71.7 137 95.5 151 99.9 58 91.4 151 24.6 145 28.4 145 32.4 141 80.6 145 99.9 120 83.7 152 5.16 26 9.70 45 5.15 44 99.9 144 99.9 113 95.7 143
SILK [79]121.2 33.3 133 64.6 134 29.2 138 46.3 131 99.9 129 39.4 131 95.0 148 99.9 94 66.5 134 56.7 135 99.9 58 50.2 132 16.1 130 18.7 134 17.2 133 48.7 139 99.9 120 37.3 134 7.26 80 9.22 34 14.5 128 70.9 136 99.9 113 51.7 132
Horn & Schunck [3]121.5 28.8 125 57.7 119 25.3 132 41.1 129 99.9 129 28.2 125 80.0 131 99.9 94 75.9 142 75.5 138 99.9 58 66.3 138 15.7 129 18.5 130 13.9 123 41.9 138 99.9 120 41.4 138 11.6 125 13.6 119 6.16 59 45.4 132 99.9 113 39.5 130
TI-DOFE [24]123.8 40.2 137 61.2 129 38.6 143 60.2 137 99.9 129 53.7 140 90.4 142 99.9 94 78.2 144 83.9 148 99.9 58 82.8 149 16.6 135 19.4 139 16.4 131 38.1 133 99.9 120 38.7 136 8.51 95 10.3 65 7.75 88 56.3 134 99.9 113 49.7 131
HCIC-L [99]125.9 44.4 140 66.7 138 29.1 136 99.9 151 99.9 129 99.9 151 47.8 115 99.9 94 44.6 116 58.5 136 99.9 58 55.7 136 20.3 142 20.6 140 24.0 139 33.9 127 86.5 105 33.6 131 38.0 147 49.5 147 36.0 140 13.7 115 25.8 79 13.7 110
FFV1MT [106]128.5 36.4 136 71.0 150 25.8 133 50.3 136 99.9 129 39.7 133 65.6 125 99.1 92 41.6 115 99.9 152 99.9 58 97.4 153 20.2 141 19.3 138 33.1 144 62.5 141 99.9 120 71.1 143 9.04 101 14.8 127 11.2 116 99.9 144 99.9 113 99.9 144
SLK [47]129.1 53.1 146 66.3 137 59.5 153 60.9 139 98.1 126 58.6 141 89.7 140 99.9 94 67.7 135 99.9 152 99.9 58 95.1 152 17.0 136 18.8 135 23.5 138 56.6 140 99.9 120 51.2 140 8.43 94 11.9 98 12.2 124 99.9 144 99.9 113 99.9 144
PGAM+LK [55]130.1 43.6 138 67.1 140 43.7 145 73.8 142 99.9 129 77.5 142 61.9 124 82.9 76 63.9 129 76.6 139 99.9 58 72.5 140 17.1 138 17.1 114 31.6 140 66.2 142 99.9 120 64.8 142 18.9 144 20.3 143 22.1 137 99.9 144 99.9 113 99.9 144
Adaptive flow [45]130.3 34.4 135 59.7 126 29.7 139 84.5 149 99.9 129 77.7 143 87.6 139 99.9 94 92.8 152 54.9 133 99.9 58 39.3 129 20.6 143 23.8 144 20.9 135 37.5 131 96.7 113 29.2 128 35.2 146 30.1 146 58.6 145 38.1 130 99.9 113 33.0 128
AdaConv-v1 [126]130.8 66.7 148 69.5 143 54.6 146 80.0 143 82.0 111 79.5 144 81.7 133 82.0 70 80.9 145 82.3 141 83.1 25 82.6 143 85.8 146 85.9 146 85.2 146 88.2 147 83.0 93 82.9 146 75.7 149 67.1 149 77.8 149 85.1 138 86.4 100 84.8 137
SepConv-v1 [127]130.8 66.7 148 69.5 143 54.6 146 80.0 143 82.0 111 79.5 144 81.7 133 82.0 70 80.9 145 82.3 141 83.1 25 82.6 143 85.8 146 85.9 146 85.2 146 88.2 147 83.0 93 82.9 146 75.7 149 67.1 149 77.8 149 85.1 138 86.4 100 84.8 137
SuperSlomo [132]130.8 66.7 148 69.5 143 54.6 146 80.0 143 82.0 111 79.5 144 81.7 133 82.0 70 80.9 145 82.3 141 83.1 25 82.6 143 85.8 146 85.9 146 85.2 146 88.2 147 83.0 93 82.9 146 75.7 149 67.1 149 77.8 149 85.1 138 86.4 100 84.8 137
FGIK [136]130.8 66.7 148 69.5 143 54.6 146 80.0 143 82.0 111 79.5 144 81.7 133 82.0 70 80.9 145 82.3 141 83.1 25 82.6 143 85.8 146 85.9 146 85.2 146 88.2 147 83.0 93 82.9 146 75.7 149 67.1 149 77.8 149 85.1 138 86.4 100 84.8 137
CtxSyn [137]130.8 66.7 148 69.5 143 54.6 146 80.0 143 82.0 111 79.5 144 81.7 133 82.0 70 80.9 145 82.3 141 83.1 25 82.6 143 85.8 146 85.9 146 85.2 146 88.2 147 83.0 93 82.9 146 75.7 149 67.1 149 77.8 149 85.1 138 86.4 100 84.8 137
CyclicGen [154]130.8 66.7 148 69.5 143 54.6 146 80.0 143 82.0 111 79.5 144 81.7 133 82.0 70 80.9 145 82.3 141 83.1 25 82.6 143 85.8 146 85.9 146 85.2 146 88.2 147 83.0 93 82.9 146 75.7 149 67.1 149 77.8 149 85.1 138 86.4 100 84.8 137
Periodicity [78]133.9 54.4 147 84.3 153 26.5 134 99.9 151 99.9 129 99.9 151 99.9 150 99.9 94 99.9 153 81.4 140 99.9 58 76.3 141 99.9 153 99.9 152 99.9 153 99.9 153 99.9 120 99.9 153 6.06 53 14.8 127 70.6 148 99.9 144 99.9 113 99.9 144
FOLKI [16]135.4 43.8 139 74.4 151 38.0 142 99.9 151 99.9 129 99.9 151 89.7 140 99.9 94 76.2 143 85.0 149 99.9 58 81.0 142 23.2 144 22.5 143 38.8 145 66.2 142 99.9 120 62.7 141 17.0 142 17.6 139 42.6 143 99.9 144 99.9 113 99.9 144
Pyramid LK [2]137.1 47.7 142 60.7 128 56.1 152 88.6 150 99.9 129 91.9 150 99.9 150 99.9 94 89.4 151 83.0 147 99.9 58 83.2 150 98.4 152 99.9 152 87.9 152 87.4 146 99.9 120 81.4 145 16.9 141 16.6 136 57.0 144 99.9 144 99.9 113 99.9 144
AVG_FLOW_ROB [142]140.7 99.9 154 99.9 154 99.9 154 99.9 151 99.9 129 99.9 151 99.9 150 99.9 94 99.9 153 99.9 152 99.9 58 99.9 154 99.9 153 99.9 152 99.9 153 99.9 153 99.9 120 99.9 153 48.6 148 52.0 148 38.2 141 99.9 144 99.9 113 99.9 144
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] LFNet_ROB 0.068 2 color Anonymous. Learning a flow network. ROB 2018 submission.
[151] WRT 9 2 color L. Mei, J. Lai, X. Xie, J. Zhu, and J. Chen. Illumination-invariance optical flow estimation using weighted regularization transform. Submitted to IEEE TCSVT 2018.
[152] EAI-Flow 2.1 2 color Anonymous. Hierarchical coherency sensitive hashing and interpolation with RANSAC for large displacement optical flow. CVIU 2018 submission 17-678.
[153] ContinualFlow_ROB 0.5 all color Anonymous. Continual Flow. ACCV 2018 submission 273.
[154] CyclicGen 0.088 2 color Anonymous. (Interpolation results only.) Deep video frame interpolation using cyclic frame generation. AAAI 2019 submission 323.
* 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.