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        
R5.0
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]11.8 7.69 2 26.2 5 3.54 2 7.19 14 30.7 27 6.11 22 5.88 5 19.3 8 4.53 17 4.01 10 23.9 18 2.00 13 11.7 1 16.4 3 5.52 3 10.4 16 29.7 9 9.30 11 5.25 22 20.0 26 2.61 12 1.88 8 6.47 28 0.19 1
MDP-Flow2 [68]13.3 10.3 21 30.4 20 6.57 23 5.28 2 23.9 3 4.13 3 5.46 3 17.6 3 3.58 8 4.49 19 25.3 24 2.11 16 15.8 22 21.4 22 10.2 29 10.6 17 29.9 10 9.87 17 4.44 11 19.3 19 2.81 13 1.39 3 4.82 6 1.11 5
OFLAF [77]14.0 8.96 9 27.9 9 4.57 8 7.36 18 26.4 8 6.68 24 4.80 2 14.9 2 3.37 7 4.43 18 21.2 9 2.81 29 13.4 7 19.0 10 7.08 7 11.4 26 28.0 3 9.28 9 5.35 24 19.1 17 3.15 15 2.47 18 5.73 19 5.50 38
NN-field [71]15.5 8.65 5 28.3 10 4.00 4 8.38 28 33.1 43 7.44 30 5.86 4 19.0 7 4.53 17 3.15 2 21.4 11 1.25 3 12.0 4 16.9 5 5.41 2 6.58 2 20.2 1 3.45 1 8.64 57 23.6 60 2.88 14 2.47 18 8.48 43 0.20 2
PMMST [114]21.3 12.9 49 32.6 32 8.45 47 10.2 44 30.5 24 10.7 61 7.39 17 22.2 14 6.31 35 3.87 9 13.5 1 2.73 25 14.0 10 18.7 9 7.52 8 10.9 23 28.9 6 9.95 18 4.99 19 20.8 31 3.18 17 1.52 4 3.87 3 1.12 6
ComponentFusion [96]24.3 8.86 8 28.7 13 5.91 17 6.30 5 24.2 4 5.98 17 6.79 11 21.6 11 4.99 23 4.11 11 24.4 19 2.04 15 16.2 30 22.0 27 11.3 41 13.4 45 40.4 60 12.4 59 7.66 49 21.3 38 5.22 41 2.05 9 5.21 10 3.61 21
ALD-Flow [66]24.9 8.44 4 27.6 8 4.09 5 6.49 7 27.2 12 5.04 9 7.66 22 24.1 24 3.72 10 4.58 22 27.1 37 2.01 14 16.0 25 22.7 35 8.55 12 9.39 9 33.3 21 8.46 8 7.21 45 18.5 11 17.3 100 4.06 53 11.1 58 5.94 47
HAST [109]25.2 7.13 1 21.8 1 3.37 1 7.28 16 26.1 7 6.10 21 3.86 1 12.2 1 0.97 1 3.85 8 21.3 10 1.50 4 11.7 1 16.5 4 4.64 1 15.1 68 35.5 34 14.0 72 19.4 102 36.3 109 39.0 133 1.24 2 3.55 1 1.32 7
nLayers [57]25.8 8.66 6 25.4 3 4.54 6 13.0 85 32.1 36 13.7 93 7.74 24 21.8 12 8.85 65 3.29 4 18.2 2 1.89 10 11.8 3 16.2 2 6.65 6 12.2 32 28.4 4 10.3 21 8.59 56 21.5 43 4.98 37 2.36 16 5.74 20 5.08 34
LME [70]26.0 9.71 15 29.0 15 6.46 22 5.49 4 22.8 2 4.79 7 8.62 38 22.4 15 11.2 76 4.73 24 28.3 52 2.35 20 16.5 33 21.9 25 12.6 54 10.7 18 34.0 27 9.81 16 5.57 25 21.3 38 3.87 27 2.40 17 6.32 25 4.52 29
NNF-EAC [103]26.1 10.9 30 31.8 27 6.97 25 6.64 9 26.4 8 5.65 11 6.61 9 20.7 9 4.44 16 5.48 43 27.1 37 2.97 34 16.5 33 22.3 31 11.1 38 12.1 31 30.7 11 9.97 19 6.42 35 21.4 40 3.94 29 2.95 33 7.51 38 4.63 31
TC/T-Flow [76]27.4 9.15 10 32.1 29 3.69 3 6.89 12 31.2 31 4.32 4 7.32 16 23.1 19 4.08 12 5.14 34 27.2 40 2.80 28 15.6 21 21.9 25 9.71 22 8.63 6 29.2 7 8.40 6 6.72 38 20.2 28 19.7 114 3.86 51 9.43 47 6.28 54
ProFlow_ROB [146]27.7 10.7 25 32.8 38 5.26 13 7.26 15 30.8 28 5.92 14 9.43 46 29.1 50 5.30 26 4.35 17 25.3 24 1.73 9 17.4 43 24.7 48 9.39 17 9.20 7 33.5 22 9.28 9 3.74 7 21.4 40 0.91 3 4.20 55 11.6 60 6.06 49
WLIF-Flow [93]28.2 9.40 14 27.1 6 6.06 19 10.0 41 33.0 41 9.71 45 7.26 15 22.4 15 5.90 34 4.53 20 23.7 16 2.56 24 16.1 26 22.3 31 11.3 41 12.8 35 33.2 20 10.4 24 6.85 40 18.8 14 7.36 57 2.80 29 6.48 29 5.62 41
FC-2Layers-FF [74]29.7 9.97 17 28.7 13 7.96 36 10.4 46 35.3 52 9.95 47 6.11 7 18.1 6 6.82 39 4.12 12 20.9 7 2.48 23 13.3 6 17.7 6 9.47 19 13.5 46 32.6 15 11.3 37 14.0 83 26.0 74 12.5 81 1.84 6 4.18 4 4.53 30
RNLOD-Flow [121]30.0 7.93 3 24.8 2 5.41 15 8.33 27 31.7 33 6.84 26 7.47 18 23.3 20 4.60 19 3.62 6 21.1 8 1.66 7 14.4 12 21.0 16 7.80 9 12.7 34 32.9 16 12.2 58 17.4 96 34.4 104 20.7 116 2.55 21 5.57 17 5.30 36
Layers++ [37]30.6 10.2 19 29.1 16 8.58 50 10.8 52 30.6 25 11.0 67 5.90 6 17.7 4 6.34 36 3.40 5 18.2 2 1.66 7 12.2 5 16.0 1 9.69 21 13.9 52 33.6 23 11.9 52 13.9 82 27.9 80 8.74 63 2.33 15 4.94 7 5.70 44
FESL [72]31.7 8.67 7 25.6 4 4.73 9 13.6 91 39.4 90 12.9 88 8.22 32 24.3 28 7.10 43 3.76 7 20.3 6 2.12 17 14.1 11 20.1 12 9.08 15 11.2 25 31.0 12 9.80 15 11.3 66 30.4 90 7.66 59 2.10 10 5.40 13 2.40 11
SVFilterOh [111]32.4 12.7 47 28.4 11 8.58 50 9.41 36 29.1 19 8.31 36 6.39 8 17.7 4 5.05 25 3.23 3 18.8 5 0.95 2 14.9 15 20.9 15 6.51 5 13.2 40 32.1 14 11.6 44 22.0 108 46.5 136 30.5 127 1.61 5 4.97 9 2.45 13
TC-Flow [46]33.7 9.24 11 30.9 23 5.24 12 5.48 3 25.7 6 4.01 2 7.25 13 23.3 20 2.66 4 5.52 44 28.4 53 3.26 41 16.7 37 23.9 44 9.52 20 11.7 27 36.8 39 11.5 43 6.69 37 21.4 40 19.0 111 4.21 56 10.6 55 6.91 67
OAR-Flow [125]35.4 10.8 28 34.0 42 6.23 20 8.94 31 31.9 34 7.53 31 10.4 55 30.3 57 7.58 52 5.60 48 25.9 32 3.04 37 17.0 38 24.0 45 9.35 16 8.62 5 33.0 17 7.87 4 3.37 5 14.6 3 5.54 45 4.80 64 10.4 54 9.44 87
AGIF+OF [85]36.4 10.4 22 29.4 18 7.42 29 12.4 77 37.5 65 12.4 82 7.91 29 24.3 28 7.24 45 4.84 28 23.8 17 2.91 31 14.8 14 20.6 14 9.72 23 13.2 40 36.5 37 10.5 26 7.06 42 20.8 31 7.54 58 3.02 38 6.27 23 6.37 56
Efficient-NL [60]36.4 9.31 12 27.5 7 5.65 16 12.1 75 38.1 75 11.2 69 8.07 31 24.1 24 6.69 38 5.39 40 25.9 32 3.51 52 14.9 15 21.1 18 9.39 17 14.0 55 35.2 32 11.1 34 11.5 68 25.7 72 6.90 54 2.19 13 5.45 15 1.94 10
IROF++ [58]36.5 10.2 19 30.9 23 7.02 26 11.1 59 38.1 75 10.7 61 8.32 36 25.1 33 7.61 53 5.83 51 28.0 48 4.08 62 15.4 20 21.3 20 9.83 24 13.6 47 38.0 48 11.3 37 5.83 28 20.8 31 1.97 10 2.32 14 5.71 18 4.87 33
PMF [73]37.9 11.6 39 29.9 19 4.55 7 7.81 21 30.2 21 6.00 18 7.17 12 23.3 20 3.21 6 4.88 30 23.1 14 2.42 21 13.6 8 18.5 7 6.49 4 16.1 76 42.7 73 15.4 81 27.2 124 43.5 132 28.9 124 2.15 12 4.96 8 4.81 32
PH-Flow [101]38.2 10.9 30 32.6 32 7.94 35 10.9 54 37.4 63 10.5 55 7.56 21 22.8 18 7.74 57 5.75 50 27.2 40 4.04 60 14.4 12 19.8 11 8.81 13 13.0 37 33.6 23 11.1 34 12.7 76 23.1 54 17.6 102 1.84 6 4.19 5 4.50 28
Classic+CPF [83]39.3 10.9 30 31.7 26 7.88 34 11.5 65 37.9 71 10.9 65 8.27 34 25.1 33 7.51 51 5.05 32 25.8 30 3.16 39 15.1 18 21.0 16 10.6 30 13.1 38 34.6 30 10.3 21 9.87 61 22.0 48 13.1 87 2.68 24 5.85 21 5.53 39
COFM [59]40.0 10.1 18 32.0 28 7.63 30 8.06 23 30.4 23 7.17 27 8.93 43 25.9 39 8.04 61 4.17 13 24.9 22 1.63 6 18.8 53 24.0 45 18.6 99 14.4 59 33.0 17 11.7 46 8.15 52 20.4 29 14.7 94 3.16 40 5.36 12 8.09 80
3DFlow [135]40.4 12.6 46 34.9 43 5.05 11 8.12 24 31.5 32 5.95 15 6.67 10 21.9 13 2.51 3 4.59 23 18.4 4 3.30 44 16.6 35 23.0 37 10.9 33 19.7 96 46.8 95 19.8 105 18.3 99 24.2 66 33.3 129 1.20 1 3.86 2 0.68 3
Ramp [62]40.7 10.9 30 32.7 36 7.96 36 10.9 54 37.1 60 10.6 57 7.85 25 24.2 26 7.41 49 5.29 37 27.0 36 3.44 47 16.1 26 22.3 31 10.8 31 13.8 50 35.4 33 11.0 33 11.6 69 21.1 36 18.2 106 2.52 20 5.44 14 5.23 35
Sparse-NonSparse [56]41.3 10.7 25 32.5 31 8.38 45 10.9 54 36.8 58 10.7 61 7.95 30 24.5 32 7.30 47 5.42 42 27.6 43 3.49 51 16.1 26 22.1 28 11.0 36 13.3 42 36.0 36 10.6 28 10.6 64 21.1 36 10.9 72 2.91 30 5.93 22 6.18 52
Correlation Flow [75]41.6 11.9 43 35.3 44 6.03 18 6.85 11 28.0 14 4.77 6 8.29 35 25.8 37 2.17 2 4.84 28 27.2 40 2.77 27 18.5 51 25.9 59 11.7 43 16.9 84 39.5 54 16.7 88 12.1 72 24.6 68 17.8 103 2.59 22 7.33 36 3.08 14
LSM [39]42.1 10.4 22 32.6 32 8.24 40 10.8 52 37.4 63 10.4 54 7.85 25 24.3 28 7.05 41 5.32 38 27.6 43 3.41 46 15.8 22 21.5 23 11.1 38 13.7 48 35.6 35 10.9 32 13.0 77 23.2 55 12.5 81 2.99 37 6.43 27 6.14 51
JOF [140]42.2 9.77 16 29.1 16 7.11 27 12.1 75 37.9 71 12.0 78 7.25 13 21.2 10 7.73 55 4.82 26 25.8 30 3.02 36 14.9 15 20.2 13 10.0 28 13.1 38 33.7 25 10.6 28 17.8 97 29.5 85 28.8 123 2.94 32 6.80 32 5.71 45
ProbFlowFields [128]42.6 16.2 67 47.8 81 11.7 78 8.96 32 31.0 29 8.86 39 9.73 50 28.4 48 10.1 70 6.09 59 25.5 27 4.53 72 18.2 49 25.5 52 10.9 33 9.76 13 34.2 28 11.7 46 4.63 12 18.8 14 3.79 26 2.95 33 8.94 46 3.52 19
Classic+NL [31]44.7 10.5 24 31.4 25 8.38 45 11.1 59 37.9 71 10.6 57 7.87 27 24.0 23 7.48 50 5.57 46 27.6 43 3.62 54 15.8 22 21.5 23 10.8 31 14.1 56 37.4 43 11.4 40 14.8 86 25.9 73 13.4 90 2.61 23 5.29 11 6.10 50
S2D-Matching [84]46.1 10.7 25 32.2 30 8.71 52 10.7 50 36.6 57 10.2 50 8.94 44 27.2 44 6.96 40 5.17 35 26.0 34 3.36 45 16.3 31 22.1 28 10.9 33 14.4 59 37.0 40 11.7 46 16.4 91 26.0 74 16.5 98 2.79 28 5.49 16 6.49 57
FMOF [94]46.4 11.0 37 30.4 20 8.33 44 13.0 85 38.5 81 12.6 84 7.51 19 22.6 17 7.34 48 5.06 33 25.2 23 3.44 47 15.3 19 21.3 20 9.87 26 14.9 67 33.1 19 11.4 40 11.7 71 24.3 67 15.0 96 3.92 52 8.59 44 6.28 54
IIOF-NLDP [131]48.1 14.6 54 41.6 64 6.71 24 11.0 58 37.5 65 8.25 35 8.77 41 26.9 42 4.19 13 6.07 58 28.0 48 3.76 56 19.7 66 27.1 73 12.6 54 16.7 81 40.7 63 15.4 81 4.68 16 23.0 52 4.41 33 2.78 27 7.26 35 3.16 15
IROF-TV [53]48.3 11.6 39 35.3 44 9.03 55 11.2 62 38.2 78 10.9 65 8.85 42 26.5 40 7.73 55 6.04 57 33.0 76 3.62 54 17.1 41 23.1 39 13.5 68 16.3 77 44.8 81 13.5 70 3.41 6 16.9 5 1.13 6 2.71 25 6.80 32 5.67 43
TV-L1-MCT [64]49.7 10.9 30 30.5 22 8.56 49 13.8 94 40.9 102 13.2 90 8.68 39 25.8 37 7.98 60 4.83 27 25.7 28 3.26 41 17.4 43 23.5 42 13.7 72 14.8 66 36.7 38 12.7 62 5.84 29 19.4 20 10.1 69 3.53 46 6.42 26 6.63 60
2DHMM-SAS [92]50.7 10.9 30 32.6 32 8.06 38 11.5 65 39.5 91 10.6 57 10.0 53 28.3 47 7.91 59 5.93 53 28.2 51 4.07 61 16.1 26 22.3 31 11.0 36 13.7 48 38.3 50 11.1 34 12.3 74 23.2 55 18.0 105 3.08 39 6.48 29 6.24 53
SimpleFlow [49]50.9 11.6 39 33.7 39 8.98 54 12.5 81 38.9 84 12.6 84 10.4 55 29.3 53 9.20 66 5.99 54 27.6 43 4.08 62 16.3 31 22.2 30 11.1 38 16.7 81 37.4 43 12.7 62 8.29 53 19.9 24 6.11 50 2.74 26 6.28 24 5.86 46
AggregFlow [97]51.2 13.9 52 33.8 40 11.2 71 13.7 92 39.6 93 12.6 84 12.0 68 31.3 58 13.7 86 5.40 41 23.5 15 3.44 47 17.5 45 25.4 50 7.98 11 8.57 4 25.9 2 8.42 7 7.00 41 24.1 65 4.53 34 5.53 75 9.80 49 11.7 98
Aniso-Texture [82]51.5 9.33 13 28.5 12 7.26 28 9.17 34 26.8 11 10.2 50 10.1 54 29.2 52 7.28 46 2.77 1 22.6 12 0.94 1 19.9 69 27.1 73 13.3 66 14.6 63 38.5 51 12.4 59 31.5 131 46.3 135 18.2 106 4.45 59 10.1 51 6.60 59
CostFilter [40]51.5 14.1 53 36.2 50 8.48 48 8.61 30 30.6 25 7.43 29 8.26 33 26.9 42 4.40 15 5.72 49 28.1 50 3.24 40 13.7 9 18.5 7 7.81 10 16.6 80 45.0 86 16.0 85 26.8 122 48.6 139 32.7 128 2.93 31 7.59 39 5.38 37
Adaptive [20]52.1 10.9 30 33.8 40 4.92 10 10.5 47 35.0 51 9.53 44 12.2 69 33.7 62 7.68 54 5.57 46 30.3 62 2.95 33 21.7 94 26.7 68 20.6 106 10.8 20 34.9 31 7.26 3 14.0 83 28.8 84 4.88 35 4.50 61 10.2 53 6.84 65
MDP-Flow [26]52.5 12.2 45 40.6 58 8.88 53 9.32 35 28.3 16 10.5 55 9.09 45 28.1 46 9.37 68 6.03 56 30.6 64 3.99 58 17.2 42 23.1 39 12.4 49 13.9 52 42.7 73 12.5 61 7.10 44 23.6 60 4.09 31 5.35 71 13.2 70 7.09 70
RFlow [90]53.5 14.8 55 43.9 69 11.2 71 6.64 9 26.6 10 5.76 12 11.7 64 35.9 72 5.04 24 4.31 15 27.1 37 1.94 11 19.4 58 26.8 70 13.0 64 14.7 64 42.2 70 11.8 51 13.1 78 22.2 49 13.1 87 5.87 81 14.1 79 8.71 84
Occlusion-TV-L1 [63]54.0 12.9 49 36.1 49 8.26 43 9.51 39 32.7 38 8.99 40 12.3 70 34.4 67 8.27 62 5.53 45 29.8 60 3.04 37 20.5 82 28.5 94 13.8 74 9.95 14 37.9 45 11.6 44 7.64 48 21.8 47 3.47 20 5.69 78 13.9 77 7.59 75
WRT [150]54.3 15.4 62 39.7 54 5.30 14 15.2 98 40.5 99 13.7 93 12.4 71 33.7 62 4.05 11 4.55 21 22.8 13 2.34 19 17.0 38 22.7 35 13.0 64 24.7 120 45.1 87 18.9 101 6.51 36 23.5 58 6.55 52 3.24 42 7.37 37 3.24 16
OFH [38]55.3 15.0 57 40.9 60 14.4 91 7.06 13 29.9 20 5.37 10 10.8 58 33.1 61 4.86 22 5.84 52 30.6 64 3.46 50 19.5 61 26.1 61 15.3 82 15.6 72 46.5 94 16.6 87 4.19 9 21.7 45 3.74 25 5.39 73 15.4 90 7.23 71
PWC-Net_ROB [147]57.5 23.5 104 52.0 101 13.4 87 13.0 85 37.8 69 12.4 82 14.0 80 39.3 80 14.4 89 7.08 70 24.7 20 2.76 26 19.9 69 25.7 58 12.7 57 13.8 50 43.4 76 13.4 69 3.79 8 22.4 50 1.03 5 2.14 11 6.78 31 1.08 4
MLDP_OF [89]58.3 18.8 89 51.3 96 16.0 94 8.16 25 32.0 35 6.76 25 10.7 57 31.9 60 5.45 29 4.81 25 26.1 35 2.44 22 18.7 52 24.3 47 13.7 72 15.6 72 37.9 45 18.6 99 19.2 101 28.5 82 38.7 132 3.53 46 7.25 34 4.27 26
DeepFlow2 [108]60.2 15.0 57 43.6 68 11.0 67 10.1 42 34.2 48 9.29 42 12.9 73 36.8 73 11.1 75 7.47 76 32.1 71 4.75 76 17.8 46 25.4 50 9.97 27 10.7 18 40.2 59 10.3 21 6.78 39 18.7 13 13.3 89 9.05 105 17.3 100 15.3 109
DMF_ROB [139]61.3 16.8 73 47.5 80 11.2 71 10.6 49 34.1 47 9.86 46 14.7 83 41.2 92 11.9 79 7.41 75 33.9 82 4.25 65 18.8 53 25.9 59 12.7 57 11.9 29 41.3 68 11.7 46 4.39 10 18.8 14 5.30 43 6.00 83 14.9 86 8.11 81
S2F-IF [123]61.5 18.0 82 51.9 99 10.9 64 11.1 59 38.6 82 10.6 57 13.9 78 40.6 89 13.4 85 7.68 83 32.6 73 5.18 82 19.7 66 27.2 76 13.3 66 10.8 20 39.5 54 11.9 52 4.99 19 19.9 24 6.26 51 3.26 43 10.1 51 3.57 20
PGM-C [120]61.8 17.7 77 50.5 90 11.0 67 11.9 69 39.1 86 11.6 74 13.9 78 40.4 88 13.3 84 7.52 79 35.8 93 4.62 75 19.6 64 27.5 79 12.4 49 9.48 11 37.9 45 9.36 12 4.63 12 16.9 5 5.02 38 4.83 65 14.2 82 6.69 61
Steered-L1 [118]62.1 11.4 38 37.9 52 7.71 33 4.42 1 21.7 1 3.76 1 7.71 23 25.7 35 4.29 14 4.91 31 29.8 60 2.26 18 20.2 76 26.7 68 16.6 89 18.1 90 46.1 93 14.6 74 32.4 132 37.9 117 51.5 142 8.58 102 15.5 92 15.2 108
Sparse Occlusion [54]62.4 12.7 47 35.8 48 8.24 40 12.4 77 33.4 44 13.4 91 9.67 48 29.1 50 6.55 37 5.99 54 28.5 54 3.56 53 19.4 58 26.4 66 12.4 49 14.7 64 39.4 53 11.7 46 37.7 139 48.6 139 17.8 103 3.66 48 9.43 47 5.64 42
Classic++ [32]62.8 10.8 28 32.7 36 8.25 42 10.5 47 32.9 39 10.7 61 10.8 58 31.6 59 8.46 63 5.25 36 29.7 59 2.99 35 20.0 72 28.0 87 13.9 76 15.2 70 44.1 79 11.9 52 17.3 95 26.2 76 18.3 108 5.82 80 12.7 66 8.14 82
CPM-Flow [116]62.8 17.7 77 50.5 90 11.0 67 11.9 69 39.0 85 11.7 75 13.7 77 39.8 86 13.2 82 7.49 78 35.5 90 4.58 74 19.5 61 27.2 76 12.3 48 9.44 10 37.3 42 9.46 13 5.05 21 19.5 21 5.17 40 5.21 69 14.8 84 7.36 73
FlowFields+ [130]63.0 18.4 84 52.2 103 11.4 75 11.9 69 39.9 95 11.5 71 14.9 86 43.4 98 14.4 89 7.97 86 33.1 77 5.58 86 19.4 58 26.9 72 12.7 57 10.2 15 39.9 57 10.5 26 4.74 17 20.1 27 4.29 32 3.80 49 12.4 65 3.48 18
EpicFlow [102]64.1 17.7 77 50.6 92 10.9 64 12.0 73 39.3 89 11.7 75 14.5 82 42.2 94 13.2 82 7.47 76 35.5 90 4.57 73 19.8 68 27.6 80 12.8 62 9.73 12 38.1 49 10.1 20 4.63 12 17.2 7 4.88 35 5.31 70 14.3 83 7.47 74
NL-TV-NCC [25]64.8 16.5 70 40.4 56 9.10 56 10.7 50 37.0 59 8.07 33 8.59 37 26.8 41 3.17 5 6.24 61 33.4 78 3.26 41 21.4 91 29.7 108 12.7 57 21.2 102 48.2 97 17.3 93 13.4 81 35.6 107 13.0 86 4.73 63 12.8 67 3.24 16
ACK-Prior [27]65.0 19.5 94 41.5 62 14.3 90 6.57 8 27.6 13 4.53 5 7.87 27 25.7 35 3.70 9 4.33 16 25.7 28 1.53 5 20.5 82 25.6 55 18.3 96 23.1 115 44.0 78 18.5 98 29.9 127 33.1 98 45.6 140 7.91 98 14.8 84 11.7 98
BriefMatch [124]65.1 11.8 42 35.7 47 6.41 21 7.52 19 30.3 22 5.97 16 7.54 20 24.2 26 4.62 20 4.28 14 25.4 26 1.98 12 20.6 84 26.2 64 20.9 108 26.8 126 49.2 98 28.2 128 22.8 113 35.9 108 39.6 136 9.81 108 15.1 88 18.3 117
CombBMOF [113]65.6 15.2 59 48.2 83 7.67 31 11.3 63 34.5 49 9.95 47 8.75 40 27.2 44 5.37 28 7.60 82 32.1 71 5.65 89 18.0 48 23.0 37 13.9 76 21.7 107 44.9 83 24.3 119 22.6 110 37.2 112 14.5 93 2.97 36 7.73 41 4.35 27
FlowFields [110]66.0 18.3 83 51.9 99 11.1 70 11.9 69 39.5 91 11.5 71 14.8 84 43.3 97 14.2 87 7.96 85 33.5 81 5.52 85 19.9 69 27.6 80 13.6 70 11.0 24 40.5 61 12.1 56 4.93 18 19.7 22 5.34 44 3.85 50 12.3 64 3.89 23
Complementary OF [21]67.1 20.9 97 51.7 97 21.5 105 6.41 6 28.3 16 4.86 8 9.56 47 30.2 56 5.62 30 8.21 88 31.4 67 6.20 92 19.2 56 25.6 55 15.5 83 21.5 105 49.3 100 17.4 94 6.34 33 19.8 23 11.5 75 6.44 90 16.1 96 10.2 91
TF+OM [100]68.1 14.8 55 35.4 46 7.68 32 9.06 33 28.4 18 9.32 43 11.6 63 28.4 48 16.0 95 6.43 62 29.0 57 4.29 66 20.2 76 25.6 55 18.4 97 17.9 88 38.5 51 16.9 91 16.6 92 33.8 100 14.7 94 6.87 93 15.5 92 9.68 88
ROF-ND [107]68.6 18.4 84 45.8 76 11.5 76 7.31 17 25.4 5 6.02 19 9.70 49 29.4 54 4.66 21 9.09 94 28.7 55 5.98 91 21.6 93 29.5 105 14.5 79 19.9 97 44.8 81 15.3 80 33.3 136 41.0 123 30.1 126 2.95 33 7.63 40 2.41 12
DeepFlow [86]70.1 17.5 76 46.9 78 16.5 95 11.8 67 35.8 53 11.2 69 15.1 87 39.6 84 15.2 93 7.81 84 32.6 73 5.12 81 17.8 46 25.5 52 9.86 25 12.0 30 44.9 83 11.4 40 6.11 32 18.0 9 12.8 84 10.8 114 18.7 108 18.8 119
ComplOF-FED-GPU [35]70.3 17.9 80 52.0 101 15.4 93 7.90 22 33.9 46 5.82 13 10.8 58 34.2 65 5.67 31 6.99 69 31.5 68 4.51 71 19.2 56 26.3 65 12.9 63 18.2 91 50.5 107 18.6 99 15.1 88 23.6 60 22.3 118 5.37 72 15.4 90 6.76 62
TCOF [69]70.5 17.2 75 45.4 75 15.3 92 12.6 82 37.6 67 12.3 80 15.7 89 39.5 82 16.6 97 6.72 65 27.7 47 4.48 70 22.5 102 30.9 119 11.9 44 9.21 8 28.4 4 10.8 31 22.9 114 35.0 106 9.29 65 4.22 57 11.3 59 6.79 63
TV-L1-improved [17]70.8 11.9 43 36.8 51 8.23 39 8.49 29 31.0 29 7.83 32 11.9 66 33.7 62 7.19 44 5.35 39 28.9 56 2.91 31 20.3 80 28.0 87 12.0 46 27.2 128 55.4 120 30.4 130 23.1 116 38.0 118 22.9 119 5.61 77 14.0 78 7.74 78
EPPM w/o HM [88]71.5 19.4 92 53.2 106 11.2 71 8.23 26 34.8 50 6.07 20 11.1 61 35.1 70 5.89 33 7.31 73 33.4 78 4.76 77 18.9 55 23.2 41 17.1 91 21.3 103 50.3 106 20.1 106 20.7 106 30.3 89 40.9 138 3.20 41 8.13 42 5.59 40
HBM-GC [105]72.0 31.9 110 41.2 61 25.6 112 13.2 88 32.9 39 14.2 96 9.93 52 24.4 31 8.75 64 10.1 103 24.7 20 6.95 100 16.6 35 21.1 18 13.6 70 18.5 92 33.7 25 15.5 83 33.9 138 47.5 138 20.1 115 3.38 44 8.62 45 5.97 48
Aniso. Huber-L1 [22]74.3 13.6 51 40.4 56 9.77 57 19.4 102 40.1 97 22.0 104 16.4 93 38.4 75 18.3 100 7.56 80 33.4 78 5.00 80 20.1 75 27.7 85 12.5 52 14.5 61 39.7 56 10.4 24 20.8 107 32.0 93 12.9 85 4.35 58 10.8 56 6.56 58
Rannacher [23]74.9 15.5 63 43.5 67 10.7 61 11.4 64 35.8 53 11.5 71 14.2 81 39.0 79 10.8 72 6.59 64 30.8 66 4.20 64 21.0 87 29.6 107 12.6 54 19.1 94 50.8 108 15.2 78 14.7 85 26.8 77 16.7 99 4.86 66 12.9 68 7.03 69
SIOF [67]75.0 16.5 70 40.1 55 10.8 63 10.3 45 37.1 60 9.10 41 16.4 93 38.3 74 18.4 101 8.56 89 35.1 88 5.87 90 21.3 90 28.5 94 16.5 88 17.6 86 43.6 77 19.7 104 7.08 43 21.6 44 3.65 23 6.65 91 16.1 96 10.9 95
F-TV-L1 [15]75.3 31.8 109 60.6 111 43.6 128 13.7 92 38.4 80 13.1 89 15.6 88 39.4 81 10.1 70 10.9 106 37.3 99 8.78 107 20.0 72 26.5 67 16.0 87 12.9 36 40.7 63 10.7 30 9.68 60 23.7 63 3.52 22 4.49 60 12.0 62 4.19 25
FF++_ROB [145]75.5 19.0 90 52.4 105 12.2 82 12.4 77 40.0 96 11.9 77 16.2 91 45.2 101 16.3 96 9.16 96 35.6 92 7.36 103 20.2 76 28.0 87 13.8 74 13.3 42 40.5 61 12.8 64 5.68 26 19.2 18 8.50 61 4.50 61 11.8 61 7.66 76
Brox et al. [5]78.3 18.5 86 51.2 94 20.8 104 14.0 95 37.8 69 15.1 98 13.6 76 38.8 77 11.7 77 7.20 72 36.8 96 4.02 59 23.0 106 28.5 94 24.3 119 10.8 20 45.3 89 9.57 14 7.81 50 22.7 51 1.58 8 9.61 107 19.2 111 15.0 107
LocallyOriented [52]78.5 15.8 66 41.5 62 10.9 64 15.0 97 44.5 109 13.7 93 17.6 97 43.4 98 14.2 87 7.16 71 31.5 68 4.82 78 21.0 87 29.0 100 12.5 52 11.7 27 34.5 29 12.9 65 11.6 69 29.6 86 12.0 78 7.94 99 18.4 105 11.1 96
SRR-TVOF-NL [91]78.5 22.3 103 44.7 71 12.5 83 12.0 73 38.1 75 10.2 50 14.8 84 40.6 89 10.9 74 6.13 60 34.1 83 2.81 29 19.6 64 25.5 52 13.5 68 16.4 78 42.4 71 13.0 66 30.5 128 42.5 128 18.3 108 6.41 89 11.0 57 12.0 100
DPOF [18]79.3 20.5 96 50.2 89 10.5 58 12.6 82 41.8 104 11.0 67 11.8 65 34.3 66 10.8 72 8.61 91 38.9 107 5.43 83 19.5 61 26.1 61 15.1 80 16.8 83 41.5 69 15.2 78 23.3 117 23.9 64 50.1 141 5.05 67 14.1 79 4.13 24
CRTflow [80]80.2 16.5 70 49.5 86 10.6 60 9.63 40 33.8 45 8.65 38 13.1 74 38.8 77 7.80 58 6.86 67 34.3 85 4.44 69 20.0 72 27.8 86 12.2 47 31.4 134 59.0 130 36.7 134 10.3 62 30.4 90 12.0 78 8.56 101 20.4 118 12.9 105
Bartels [41]82.1 19.3 91 39.6 53 22.4 108 9.47 38 28.2 15 10.0 49 9.91 51 29.7 55 7.09 42 9.18 97 29.3 58 7.40 104 21.7 94 27.6 80 21.1 110 19.1 94 44.4 80 24.2 118 23.0 115 36.3 109 36.2 130 7.46 97 14.9 86 11.5 97
Dynamic MRF [7]83.3 22.0 101 52.3 104 25.2 110 7.67 20 33.0 41 6.18 23 12.4 71 39.8 86 5.34 27 6.49 63 35.4 89 3.86 57 22.9 104 29.2 103 20.7 107 22.2 109 57.8 127 22.9 115 7.42 46 18.1 10 25.1 120 13.2 121 21.3 123 20.5 123
CBF [12]85.3 15.2 59 44.8 72 12.1 80 23.7 110 37.7 68 30.9 115 13.2 75 34.6 68 14.5 91 6.86 67 32.8 75 4.32 67 22.6 103 28.4 93 20.2 104 15.6 72 41.0 66 12.1 56 32.9 134 39.7 121 29.8 125 5.49 74 13.2 70 8.30 83
Local-TV-L1 [65]85.8 24.6 105 51.2 94 30.0 113 22.5 109 40.6 100 25.2 108 23.5 109 46.1 103 28.3 111 9.73 100 37.4 100 6.92 99 18.3 50 25.2 49 12.7 57 13.9 52 43.2 75 12.0 55 5.25 22 20.6 30 5.15 39 15.8 126 21.0 121 32.1 131
DF-Auto [115]85.9 19.4 92 46.0 77 10.5 58 26.6 116 46.1 113 31.1 116 23.7 110 46.1 103 37.0 116 9.05 93 36.8 96 5.59 87 21.7 94 29.1 102 17.4 93 7.80 3 31.8 13 7.93 5 19.5 103 37.4 114 3.25 18 10.9 115 19.6 114 16.4 111
TriangleFlow [30]86.2 18.7 88 43.9 69 18.0 96 10.1 42 37.2 62 8.18 34 11.9 66 35.5 71 5.81 32 6.72 65 34.6 87 4.37 68 26.7 128 34.7 130 23.4 114 23.1 115 49.6 101 23.5 116 16.7 93 37.2 112 16.3 97 6.85 92 17.3 100 10.3 92
LDOF [28]87.2 17.1 74 48.0 82 12.9 84 13.3 89 40.6 100 12.2 79 15.8 90 42.4 95 12.7 81 9.70 99 44.0 115 6.27 94 20.7 85 28.0 87 16.8 90 14.3 58 45.9 92 13.8 71 8.36 54 23.3 57 7.98 60 11.2 118 21.2 122 18.3 117
SuperFlow [81]87.2 16.2 67 42.7 66 13.0 85 20.9 103 39.6 93 25.0 107 19.7 102 40.6 89 31.8 113 9.89 102 41.2 110 7.16 101 20.9 86 27.1 73 20.3 105 12.2 32 41.1 67 11.3 37 19.0 100 32.1 95 3.87 27 10.1 110 19.3 112 16.4 111
CNN-flow-warp+ref [117]88.5 18.5 86 50.0 87 13.9 89 17.8 100 37.9 71 21.1 102 21.3 106 47.3 107 29.7 112 9.13 95 38.8 104 6.72 97 21.8 97 28.2 91 19.6 101 14.2 57 45.7 91 13.1 67 5.94 30 18.5 11 10.9 72 12.4 120 20.6 120 16.4 111
CLG-TV [48]88.6 15.7 65 42.2 65 11.7 78 20.9 103 39.2 88 24.8 106 16.4 93 39.5 82 18.0 99 9.23 98 37.9 101 6.54 95 22.9 104 30.0 112 17.9 95 16.5 79 47.2 96 14.2 73 19.9 104 30.4 90 11.5 75 5.79 79 14.1 79 6.98 68
TriFlow [95]89.0 21.3 99 44.9 73 13.5 88 16.0 99 36.5 56 18.7 100 18.2 99 38.6 76 27.8 108 7.35 74 30.3 62 5.59 87 21.2 89 27.3 78 18.5 98 15.1 68 37.1 41 15.0 77 49.5 144 41.7 125 95.6 147 6.38 88 13.3 72 9.77 89
p-harmonic [29]89.3 21.2 98 63.8 118 20.6 103 12.4 77 35.9 55 12.7 87 17.7 98 47.5 108 14.9 92 10.9 106 42.1 112 8.85 108 20.4 81 26.1 61 17.1 91 17.9 88 52.5 112 18.4 96 15.6 90 28.6 83 5.86 47 5.89 82 13.5 74 7.67 77
FlowNetS+ft+v [112]90.3 15.2 59 44.9 73 10.7 61 13.4 90 38.2 78 13.4 91 18.8 101 42.8 96 24.4 103 9.01 92 38.8 104 6.24 93 23.2 109 31.5 124 15.9 85 13.3 42 42.6 72 13.1 67 18.2 98 32.6 96 21.9 117 8.73 103 19.1 110 12.8 104
OFRF [134]90.6 20.1 95 40.7 59 18.8 98 25.4 114 43.7 106 27.8 113 20.4 104 39.7 85 25.4 104 12.5 109 34.3 85 11.1 113 17.0 38 23.7 43 8.99 14 17.6 86 40.0 58 16.7 88 15.0 87 28.4 81 28.3 122 16.1 127 19.4 113 34.6 132
Second-order prior [8]92.9 15.6 64 48.2 83 12.1 80 12.6 82 39.1 86 12.3 80 16.2 91 44.6 100 12.2 80 7.57 81 31.6 70 5.45 84 22.2 99 30.6 114 14.3 78 20.8 101 56.8 124 17.7 95 28.0 125 33.8 100 27.1 121 7.43 96 17.4 102 10.4 94
ContinualFlow_ROB [152]95.8 38.4 118 64.1 120 31.4 115 31.9 120 52.5 126 35.2 121 32.8 121 61.5 123 38.0 118 15.4 113 42.4 113 9.42 110 27.2 131 32.9 129 23.6 116 30.9 131 52.8 115 37.9 135 3.27 4 17.7 8 1.31 7 3.48 45 10.0 50 1.76 9
Fusion [6]96.9 17.9 80 57.7 108 18.6 97 9.42 37 32.3 37 10.2 50 11.4 62 34.8 69 11.7 77 8.57 90 40.2 108 6.89 98 25.0 121 30.8 118 24.9 125 23.9 118 52.3 111 25.0 122 33.3 136 43.4 130 19.3 113 9.01 104 18.8 109 13.4 106
Learning Flow [11]97.2 16.4 69 47.3 79 11.5 76 14.0 95 40.3 98 14.4 97 16.4 93 41.7 93 15.6 94 8.05 87 40.7 109 4.87 79 27.1 130 35.0 132 22.5 113 17.2 85 50.0 105 16.0 85 15.5 89 34.1 103 13.9 92 10.1 110 20.2 117 12.5 103
LiteFlowNet [142]97.4 32.9 112 71.7 132 19.9 100 18.2 101 45.2 111 17.8 99 21.8 107 55.4 114 17.5 98 10.7 105 34.2 84 7.20 102 24.3 117 30.7 117 20.9 108 21.5 105 52.5 112 18.4 96 11.3 66 33.3 99 3.15 15 6.04 84 13.5 74 7.98 79
StereoFlow [44]97.9 85.4 147 89.0 147 87.9 146 73.1 147 88.5 147 68.8 142 66.8 146 87.5 145 52.4 138 81.5 146 91.1 146 78.5 145 25.9 125 27.6 80 29.7 133 6.38 1 29.4 8 6.60 2 1.39 1 10.9 1 0.20 1 6.34 87 13.8 76 10.3 92
SegOF [10]100.9 28.8 108 51.1 93 13.2 86 37.3 126 51.8 124 44.6 128 30.0 118 53.0 112 43.3 127 27.0 127 49.6 122 22.4 123 24.0 116 27.6 80 28.4 132 24.9 122 58.5 128 24.4 120 2.04 2 16.2 4 0.47 2 10.0 109 16.5 98 16.7 114
EAI-Flow [151]101.6 40.7 120 62.9 115 39.5 120 21.2 106 43.8 107 21.4 103 25.8 111 57.4 115 28.2 109 11.7 108 38.0 102 9.37 109 21.9 98 29.0 100 15.1 80 22.2 109 50.8 108 20.5 107 29.0 126 36.6 111 10.6 71 5.07 68 13.3 72 6.83 64
Shiralkar [42]103.0 22.0 101 69.5 126 19.6 99 10.9 54 42.6 105 8.48 37 18.4 100 54.0 113 9.43 69 10.1 103 45.4 116 7.72 105 21.5 92 28.9 99 15.9 85 26.8 126 60.7 131 25.4 124 24.3 120 29.9 88 39.4 134 11.0 116 23.8 128 12.2 101
Ad-TV-NDC [36]103.0 44.8 129 63.0 116 69.1 138 40.3 129 48.4 118 48.3 131 34.8 125 58.5 117 39.9 120 26.5 126 47.8 120 27.7 127 20.2 76 28.5 94 11.9 44 15.2 70 40.9 65 14.7 75 8.46 55 21.0 35 5.69 46 23.9 137 28.3 137 41.9 143
AugFNG_ROB [143]103.3 44.1 127 62.1 114 25.2 110 42.2 131 56.7 129 50.8 133 37.7 129 66.7 128 42.9 125 19.0 117 41.2 110 13.4 114 26.5 126 32.3 127 23.4 114 22.6 112 57.7 126 21.1 110 6.07 31 27.3 78 0.91 3 5.54 76 15.1 88 3.80 22
StereoOF-V1MT [119]103.5 21.7 100 68.0 124 20.3 102 11.8 67 50.4 121 7.18 28 20.7 105 62.8 124 9.21 67 9.80 101 50.8 123 6.56 96 27.9 133 35.8 133 23.9 117 25.0 123 67.3 135 24.0 117 8.00 51 27.7 79 12.2 80 13.4 122 23.5 127 15.8 110
WOLF_ROB [148]104.3 26.7 107 70.6 128 21.9 106 21.8 108 51.3 123 19.4 101 28.0 115 61.1 122 26.7 107 12.6 110 42.5 114 10.5 112 22.2 99 28.7 98 19.4 100 21.3 103 55.6 121 19.6 103 7.42 46 23.5 58 8.87 64 11.1 117 20.5 119 20.0 122
FlowNet2 [122]107.5 47.2 131 61.0 112 42.4 125 44.5 133 57.5 130 51.3 134 37.6 128 64.7 126 43.1 126 21.0 120 35.8 93 17.9 119 25.8 122 30.6 114 24.6 121 20.4 98 49.7 103 21.0 108 32.5 133 53.3 143 4.06 30 4.13 54 13.0 69 1.49 8
HBpMotionGpu [43]109.4 32.0 111 50.0 87 22.9 109 36.1 125 47.0 115 43.9 127 29.2 117 51.9 111 38.6 119 13.0 111 37.1 98 10.2 111 23.5 111 29.5 105 24.2 118 18.9 93 44.9 83 15.9 84 33.2 135 41.2 124 12.6 83 11.8 119 18.5 106 22.7 124
IAOF2 [51]110.1 25.3 106 49.2 85 22.2 107 24.6 111 44.3 108 28.6 114 20.0 103 45.4 102 25.5 105 49.8 136 57.5 131 60.5 139 23.2 109 31.0 120 15.7 84 23.2 117 49.6 101 19.3 102 30.5 128 39.0 120 19.0 111 9.25 106 18.6 107 9.82 90
Modified CLG [34]110.9 34.8 116 61.1 113 35.3 117 33.3 123 46.5 114 41.7 126 36.8 127 63.0 125 45.1 131 22.1 121 55.4 127 18.7 121 23.9 114 31.2 121 21.7 112 15.8 75 51.5 110 14.8 76 9.01 58 24.6 68 11.1 74 17.6 131 25.7 133 29.6 129
LFNet_ROB [149]111.7 42.0 122 80.9 141 30.7 114 25.2 113 54.8 128 25.3 109 33.7 123 74.4 134 26.0 106 17.2 115 48.0 121 14.4 116 26.5 126 32.5 128 24.8 124 23.0 114 56.4 122 22.7 113 12.5 75 38.3 119 6.87 53 6.05 85 15.7 94 9.05 86
EPMNet [133]112.5 47.1 130 71.6 131 41.3 123 41.8 130 61.0 134 47.5 130 34.2 124 60.0 120 40.1 121 22.9 124 38.8 104 20.2 122 25.8 122 30.6 114 24.6 121 20.4 98 49.7 103 21.0 108 23.9 119 44.7 133 3.33 19 7.39 95 18.0 103 7.28 72
Filter Flow [19]112.6 33.3 113 51.7 97 20.1 101 25.0 112 47.2 117 27.7 112 27.7 114 50.0 109 37.9 117 31.7 129 54.1 126 29.9 128 25.8 122 31.2 121 28.3 131 26.4 125 52.9 116 24.7 121 42.3 141 61.5 145 13.6 91 6.09 86 12.1 63 6.88 66
2D-CLG [1]112.7 44.0 126 63.3 117 36.1 118 44.3 132 52.3 125 55.1 136 49.1 137 75.4 135 50.5 135 64.3 142 76.4 141 67.8 142 24.8 119 29.7 108 27.4 129 20.5 100 53.6 117 22.4 112 2.52 3 13.0 2 3.50 21 22.8 136 27.9 136 36.9 135
ResPWCR_ROB [144]112.8 47.7 132 78.6 135 41.4 124 20.9 103 45.6 112 22.0 104 26.3 113 59.1 119 28.2 109 16.7 114 47.4 118 13.7 115 23.7 112 28.3 92 27.2 128 25.0 123 57.6 125 25.4 124 22.7 112 42.0 126 10.0 67 8.31 100 16.9 99 12.2 101
TVL1_ROB [138]113.3 66.6 141 79.1 138 86.0 143 52.8 139 52.6 127 65.8 140 51.6 138 78.2 140 53.6 141 55.5 140 76.0 140 58.3 138 23.0 106 31.4 123 17.6 94 14.5 61 49.2 98 17.1 92 4.65 15 20.9 34 2.19 11 26.8 141 31.5 138 41.0 142
SPSA-learn [13]113.4 35.8 117 71.2 129 43.1 127 28.4 118 47.0 115 32.8 119 31.4 120 57.7 116 42.2 124 22.2 122 51.0 124 22.9 125 23.9 114 29.4 104 24.6 121 24.8 121 56.5 123 25.1 123 10.7 65 25.1 70 3.72 24 21.7 134 24.9 132 35.5 134
BlockOverlap [61]113.6 41.4 121 54.1 107 36.2 119 27.3 117 41.4 103 32.6 118 26.2 112 46.5 105 31.8 113 20.0 118 36.6 95 18.1 120 22.4 101 26.8 70 25.7 126 24.5 119 45.6 90 21.1 110 39.3 140 47.0 137 43.5 139 13.8 123 16.0 95 28.7 128
IAOF [50]114.8 33.8 114 58.3 109 40.6 121 33.0 122 44.5 109 39.5 124 30.6 119 58.7 118 33.8 115 34.1 132 52.4 125 40.8 132 23.1 108 29.9 110 19.7 103 22.5 111 53.7 118 16.8 90 22.3 109 34.0 102 10.0 67 19.5 132 23.9 129 37.1 137
GraphCuts [14]116.1 34.5 115 59.0 110 32.1 116 26.2 115 51.1 122 26.4 111 28.1 116 51.7 110 40.4 123 13.0 111 47.5 119 7.98 106 23.7 112 30.0 112 24.3 119 33.4 135 45.2 88 25.7 126 31.2 130 37.7 116 36.8 131 10.7 113 19.7 115 17.7 116
Black & Anandan [4]117.2 38.5 119 69.5 126 53.4 130 28.5 119 49.6 120 32.0 117 33.4 122 60.5 121 40.2 122 22.6 123 55.9 128 22.8 124 24.5 118 32.2 126 19.6 101 21.7 107 58.9 129 22.7 113 22.6 110 37.6 115 5.27 42 16.7 129 22.6 126 25.4 126
GroupFlow [9]117.5 42.7 123 67.1 123 53.4 130 44.8 134 63.8 139 50.2 132 36.7 126 69.4 131 43.9 128 17.2 115 46.0 117 16.7 117 27.8 132 34.9 131 21.2 111 36.7 138 67.0 134 43.6 139 6.40 34 21.7 45 7.17 55 16.6 128 25.9 134 25.0 125
2bit-BM-tele [98]119.7 55.1 136 64.1 120 69.1 138 21.4 107 38.7 83 25.3 109 23.3 108 46.7 106 21.2 102 26.2 125 38.7 103 25.2 126 24.9 120 29.9 110 27.7 130 31.3 133 52.6 114 34.6 132 43.3 142 51.7 142 54.5 143 10.3 112 20.1 116 16.8 115
Nguyen [33]122.4 43.9 125 66.0 122 42.8 126 54.0 140 49.4 119 70.1 143 42.9 132 67.4 129 47.3 133 55.4 139 65.7 135 64.4 140 27.0 129 31.8 125 31.0 134 22.8 113 54.8 119 27.3 127 13.1 78 25.2 71 6.08 49 22.2 135 26.9 135 38.9 139
SILK [79]125.1 49.5 135 69.2 125 69.3 140 39.9 128 60.6 133 47.0 129 40.4 131 70.7 132 45.6 132 32.0 130 56.5 129 31.2 130 31.4 135 36.9 136 33.3 136 31.1 132 63.2 132 32.3 131 10.3 62 23.0 52 17.3 100 25.0 138 31.9 139 36.9 135
UnFlow [129]125.8 70.9 143 78.9 136 58.5 133 51.3 138 67.4 142 56.9 137 54.4 140 83.6 143 52.8 139 33.4 131 60.2 132 30.1 129 36.7 143 38.4 139 46.2 144 38.2 139 69.6 138 42.8 138 26.2 121 40.3 122 1.60 9 7.20 94 18.3 104 8.75 85
Periodicity [78]127.5 48.9 134 63.9 119 41.0 122 34.5 124 60.0 132 37.5 123 55.4 141 67.4 129 56.6 142 20.4 119 56.9 130 17.5 118 53.2 146 66.7 147 46.5 145 48.3 144 76.0 145 46.4 141 9.14 59 34.4 104 9.98 66 28.3 142 48.2 146 40.6 141
Horn & Schunck [3]128.0 43.3 124 80.7 140 58.6 134 32.5 121 59.7 131 35.1 120 40.2 130 76.3 138 44.7 130 31.5 128 64.8 133 32.6 131 29.3 134 36.4 135 27.0 127 27.5 129 68.7 136 29.7 129 27.0 123 43.3 129 7.32 56 25.9 139 36.5 142 34.6 132
H+S_ROB [137]131.0 60.1 139 76.9 133 59.9 136 57.3 141 76.3 145 63.5 139 60.9 143 89.5 147 51.6 137 79.9 145 79.2 143 83.1 146 36.5 142 39.6 142 45.4 143 45.4 141 73.7 144 49.8 143 5.73 27 32.8 97 5.96 48 34.5 144 35.1 141 37.9 138
Heeger++ [104]131.1 61.9 140 80.2 139 47.4 129 44.8 134 77.8 146 40.9 125 68.0 147 84.7 144 62.1 145 43.6 135 69.1 136 41.9 133 32.6 137 37.9 137 32.0 135 51.1 145 78.4 146 54.2 145 13.3 80 43.4 130 10.4 70 15.0 124 21.4 124 19.6 120
SLK [47]131.8 44.7 128 78.9 136 59.1 135 58.2 143 71.2 143 70.9 144 47.5 135 83.5 142 50.6 136 65.0 143 69.5 137 73.4 143 34.7 139 38.9 141 42.9 142 34.8 136 70.9 141 39.4 136 12.1 72 29.8 87 11.5 75 34.4 143 40.1 143 48.8 144
FFV1MT [106]134.1 59.9 138 77.8 134 53.6 132 37.7 127 72.5 144 37.0 122 63.6 145 82.1 141 62.4 146 42.8 134 73.9 139 42.6 134 41.9 145 45.8 145 52.3 146 52.5 146 81.8 147 56.1 146 20.2 105 42.4 127 18.3 108 15.0 124 21.4 124 19.6 120
TI-DOFE [24]134.3 73.1 144 84.6 145 89.6 147 61.2 145 64.7 141 74.8 146 58.6 142 88.7 146 58.0 143 70.9 144 81.6 144 76.1 144 31.7 136 38.0 138 35.4 137 29.7 130 68.7 136 36.3 133 17.1 94 32.0 93 8.67 62 35.5 145 42.8 144 49.8 145
FOLKI [16]136.6 48.0 133 71.5 130 68.8 137 48.6 137 63.2 136 59.5 138 43.0 133 75.6 136 44.0 129 40.4 133 65.6 134 45.8 135 35.3 140 40.6 143 41.6 140 36.3 137 71.6 143 44.4 140 23.6 118 44.7 133 40.4 137 36.9 146 43.4 145 54.5 146
PGAM+LK [55]138.8 58.6 137 80.9 141 69.8 141 45.1 136 63.7 138 51.9 135 43.2 134 76.2 137 47.5 134 50.3 137 82.2 145 51.4 136 32.7 138 36.0 134 42.3 141 41.4 140 70.1 139 41.2 137 56.3 145 58.0 144 55.0 144 25.9 139 32.4 140 40.3 140
Adaptive flow [45]139.3 76.7 146 83.7 144 86.4 144 57.9 142 63.6 137 67.3 141 48.7 136 73.2 133 52.9 140 52.7 138 69.9 138 56.0 137 35.4 141 38.4 139 39.4 139 46.1 142 70.6 140 47.6 142 73.1 146 75.2 146 88.1 145 17.2 130 24.6 131 25.6 127
HCIC-L [99]141.6 76.5 145 86.4 146 73.3 142 70.1 146 62.5 135 85.3 147 63.5 144 66.1 127 79.5 147 83.3 147 91.8 147 86.5 147 39.0 144 42.8 144 38.7 138 46.4 143 66.6 133 52.3 144 89.6 147 85.9 147 94.0 146 19.8 133 24.2 130 29.6 129
Pyramid LK [2]143.3 68.1 142 83.5 143 86.8 145 59.4 144 64.5 140 73.1 145 52.8 139 76.3 138 61.4 144 60.2 141 79.0 142 65.9 141 53.8 147 61.8 146 64.5 147 59.4 147 71.1 142 63.0 147 43.9 143 49.4 141 39.5 135 50.2 147 60.2 147 70.8 147
AVG_FLOW_ROB [141]148.5 99.9 148 99.7 148 100.0 148 100.0 148 100.0 148 100.0 148 99.9 148 99.9 148 99.9 148 99.6 148 98.8 148 99.5 148 99.6 148 99.7 148 99.0 148 97.0 148 96.3 148 95.6 148 99.1 148 92.9 148 99.6 148 100.0 161 99.9 148 99.9 148
AdaConv-v1 [126]148.6 100.0 149 99.9 149 100.0 148 100.0 148 100.0 148 100.0 148 99.9 148 99.9 148 99.9 148 100.0 149 100.0 149 99.9 149 99.9 149 99.9 149 99.9 149 99.9 149 99.8 149 100.0 149 99.7 149 97.0 149 99.9 149 99.9 148 99.9 148 99.9 148
SepConv-v1 [127]148.6 100.0 149 99.9 149 100.0 148 100.0 148 100.0 148 100.0 148 99.9 148 99.9 148 99.9 148 100.0 149 100.0 149 99.9 149 99.9 149 99.9 149 99.9 149 99.9 149 99.8 149 100.0 149 99.7 149 97.0 149 99.9 149 99.9 148 99.9 148 99.9 148
SuperSlomo [132]148.6 100.0 149 99.9 149 100.0 148 100.0 148 100.0 148 100.0 148 99.9 148 99.9 148 99.9 148 100.0 149 100.0 149 99.9 149 99.9 149 99.9 149 99.9 149 99.9 149 99.8 149 100.0 149 99.7 149 97.0 149 99.9 149 99.9 148 99.9 148 99.9 148
CtxSyn [136]148.6 100.0 149 99.9 149 100.0 148 100.0 148 100.0 148 100.0 148 99.9 148 99.9 148 99.9 148 100.0 149 100.0 149 99.9 149 99.9 149 99.9 149 99.9 149 99.9 149 99.8 149 100.0 149 99.7 149 97.0 149 99.9 149 99.9 148 99.9 148 99.9 148
CyclicGen [153]148.6 100.0 149 99.9 149 100.0 148 100.0 148 100.0 148 100.0 148 99.9 148 99.9 148 99.9 148 100.0 149 100.0 149 99.9 149 99.9 149 99.9 149 99.9 149 99.9 149 99.8 149 100.0 149 99.7 149 97.0 149 99.9 149 99.9 148 99.9 148 99.9 148
TOF-M [154]148.6 100.0 149 99.9 149 100.0 148 100.0 148 100.0 148 100.0 148 99.9 148 99.9 148 99.9 148 100.0 149 100.0 149 99.9 149 99.9 149 99.9 149 99.9 149 99.9 149 99.8 149 100.0 149 99.7 149 97.0 149 99.9 149 99.9 148 99.9 148 99.9 148
MEMC-Net+ [155]148.6 100.0 149 99.9 149 100.0 148 100.0 148 100.0 148 100.0 148 99.9 148 99.9 148 99.9 148 100.0 149 100.0 149 99.9 149 99.9 149 99.9 149 99.9 149 99.9 149 99.8 149 100.0 149 99.7 149 97.0 149 99.9 149 99.9 148 99.9 148 99.9 148
CFRF [156]148.6 100.0 149 99.9 149 100.0 148 100.0 148 100.0 148 100.0 148 99.9 148 99.9 148 99.9 148 100.0 149 100.0 149 99.9 149 99.9 149 99.9 149 99.9 149 99.9 149 99.8 149 100.0 149 99.7 149 97.0 149 99.9 149 99.9 148 99.9 148 99.9 148
MPRN [157]148.6 100.0 149 99.9 149 100.0 148 100.0 148 100.0 148 100.0 148 99.9 148 99.9 148 99.9 148 100.0 149 100.0 149 99.9 149 99.9 149 99.9 149 99.9 149 99.9 149 99.8 149 100.0 149 99.7 149 97.0 149 99.9 149 99.9 148 99.9 148 99.9 148
DAIN [158]148.6 100.0 149 99.9 149 100.0 148 100.0 148 100.0 148 100.0 148 99.9 148 99.9 148 99.9 148 100.0 149 100.0 149 99.9 149 99.9 149 99.9 149 99.9 149 99.9 149 99.8 149 100.0 149 99.7 149 97.0 149 99.9 149 99.9 148 99.9 148 99.9 148
PyrWarp [159]148.6 100.0 149 99.9 149 100.0 148 100.0 148 100.0 148 100.0 148 99.9 148 99.9 148 99.9 148 100.0 149 100.0 149 99.9 149 99.9 149 99.9 149 99.9 149 99.9 149 99.8 149 100.0 149 99.7 149 97.0 149 99.9 149 99.9 148 99.9 148 99.9 148
InterpCNN [160]148.6 100.0 149 99.9 149 100.0 148 100.0 148 100.0 148 100.0 148 99.9 148 99.9 148 99.9 148 100.0 149 100.0 149 99.9 149 99.9 149 99.9 149 99.9 149 99.9 149 99.8 149 100.0 149 99.7 149 97.0 149 99.9 149 99.9 148 99.9 148 99.9 148
OFRI [161]148.6 100.0 149 99.9 149 100.0 148 100.0 148 100.0 148 100.0 148 99.9 148 99.9 148 99.9 148 100.0 149 100.0 149 99.9 149 99.9 149 99.9 149 99.9 149 99.9 149 99.8 149 100.0 149 99.7 149 97.0 149 99.9 149 99.9 148 99.9 148 99.9 148
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] CtxSyn 0.07 2 color S. Niklaus and F. Liu. (Interpolation results only.) Context-aware synthesis for video frame interpolation. CVPR 2018.
[137] 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.
[138] 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.
[139] 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.
[140] 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.
[141] AVG_FLOW_ROB N/A 2 N/A Average flow field of ROB 2018 training set.
[142] 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.
[143] AugFNG_ROB 0.10 all color Anonymous. FusionNet and AugmentedFlowNet: Selective proxy ground truth for training on unlabeled images. ECCV 2018 submission 2834.
[144] ResPWCR_ROB 0.2 2 color Anonymous. Learning optical flow with residual connections. ROB 2018 submission.
[145] 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.
[146] ProFlow_ROB 76 3 color Anonymous. ProFlow: Learning to predict optical flow. BMVC 2018 submission 277.
[147] 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.
[148] WOLF_ROB 0.02 2 color Anonymous. Reversed deep neural network for optical flow. ROB 2018 submission.
[149] LFNet_ROB 0.068 2 color Anonymous. Learning a flow network. ROB 2018 submission.
[150] 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.
[151] 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.
[152] ContinualFlow_ROB 0.5 all color M Neoral, J. Sochman, and J. Matas. Continual occlusions and optical flow estimation. ACCV 2018.
[153] CyclicGen 0.088 2 color Anonymous. (Interpolation results only.) Deep video frame interpolation using cyclic frame generation. AAAI 2019 submission 323.
[154] TOF-M 0.393 2 color T. Xue, B. Chen, J. Wu, D. Wei, and W. Freeman. Video enhancement with task-oriented flow. arXiv 1711.09078, 2017.
[155] MEMC-Net+ 0.16 2 color W. Bao, W.-S. Lai, X. Zhang, Z. Gao, and M.-H. Yang. (Interpolation results only.) MEMC-Net: Motion estimation and motion compensation driven neural network for video interpolation and enhancement. Submitted to TPAMI 2018.
[156] CFRF 0.128 2 color Anonymous. (Interpolation results only.) Coarse-to-fine refinement framework for video frame interpolation. CVPR 2019 submission 1992.
[157] MPRN 0.32 4 color Anonymous. (Interpolation results only.) Multi-frame pyramid refinement network for video frame interpolation. CVPR 2019 submission 1361.
[158] DAIN 0.13 2 color Anonymous. (Interpolation results only.) DAIN: Depth-aware video frame interpolation. CVPR 2019 submission 1769.
[159] PyrWarp 0.14 2 color Anonymous. (Interpolation results only.) Feature pyramid warping for video frame interpolation. CVPR 2019 submission 868.
[160] InterpCNN 0.65 2 color Anonymous. (Interpolation results only.) Video frame interpolation with a stack of synthesis networks and intermediate optical flows. CVPR 2019 submission 6533.
[161] OFRI 0.31 2 color Anonymous. (Interpolation results only.) Efficient video frame interpolation via optical flow refinement. CVPR 2019 submission 6743.
* 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.