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        
SD
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]7.4 6.84 15 15.1 23 4.48 10 6.28 5 15.5 7 3.00 4 7.42 3 13.5 6 2.39 3 7.71 5 22.7 8 2.48 2 4.26 1 5.09 1 2.83 1 6.52 2 14.9 2 4.55 4 1.74 15 3.32 45 1.23 5 5.63 3 10.7 3 2.48 5
PMMST [114]8.6 6.46 4 14.1 4 3.23 1 5.42 1 12.7 3 3.51 13 8.20 9 14.7 11 3.66 13 7.46 2 17.8 2 4.34 4 4.79 13 5.67 8 3.91 21 6.77 3 15.2 4 3.61 2 1.74 15 3.36 48 1.34 9 5.95 7 11.3 7 2.25 2
NN-field [71]11.0 7.29 34 16.0 47 4.74 22 6.15 2 15.2 6 3.02 6 7.77 7 14.1 10 2.71 6 7.56 3 22.8 9 1.96 1 4.49 2 5.36 2 3.09 3 5.72 1 14.1 1 1.96 1 1.96 32 3.34 46 1.32 8 5.70 4 10.8 4 2.49 6
OFLAF [77]13.7 6.75 12 14.9 16 4.44 8 7.07 11 17.5 17 3.10 7 8.45 10 15.5 13 2.50 5 13.0 40 35.0 72 6.38 17 4.57 3 5.57 5 3.12 4 7.63 7 15.9 7 5.90 13 1.68 12 2.86 13 1.43 17 5.83 6 11.0 6 2.79 7
MDP-Flow2 [68]20.2 6.66 7 14.7 13 4.53 12 6.79 8 17.0 11 3.23 8 8.68 11 15.8 14 2.90 7 13.4 50 33.7 59 6.89 42 4.95 25 5.84 23 4.15 41 8.49 17 18.7 27 7.72 39 1.64 9 3.08 22 1.27 7 6.77 10 12.8 12 3.33 12
FC-2Layers-FF [74]22.6 7.07 23 15.5 33 4.91 28 8.30 31 19.7 34 4.30 21 7.61 5 13.6 7 4.29 24 11.8 17 30.2 24 6.20 15 4.60 4 5.54 3 3.52 7 9.20 27 18.3 23 6.27 20 2.20 50 3.42 57 1.85 39 7.54 16 14.3 17 4.12 17
nLayers [57]22.8 7.20 29 16.0 47 4.66 19 6.25 4 14.7 4 3.70 18 7.72 6 13.7 8 4.81 32 13.1 44 34.8 70 6.69 28 4.76 10 5.75 14 4.12 38 7.19 5 14.9 2 4.40 3 1.99 33 3.10 27 1.80 38 8.22 21 15.6 22 6.10 26
ComponentFusion [96]26.9 7.22 30 15.9 44 4.61 17 7.60 17 19.2 29 3.30 9 9.70 16 17.6 19 3.77 15 11.1 12 31.0 32 4.45 5 4.96 26 5.88 26 4.25 45 10.9 50 23.7 53 9.40 60 1.90 29 3.08 22 1.69 32 8.00 19 15.1 20 4.57 19
FESL [72]29.7 6.97 18 15.3 29 4.47 9 9.74 53 21.4 50 5.49 52 11.4 28 20.2 29 4.25 23 12.5 24 31.5 38 6.81 33 4.72 7 5.71 11 3.72 15 7.03 4 16.0 8 4.81 5 2.21 52 3.49 64 1.94 42 11.1 42 17.6 34 10.1 42
Correlation Flow [75]30.2 6.66 7 14.5 9 3.81 2 7.78 22 17.5 17 2.85 2 18.0 75 29.0 83 4.31 26 9.28 7 22.1 7 5.57 10 5.12 45 6.13 58 3.98 25 11.0 52 23.3 48 10.5 72 2.07 40 3.08 22 2.32 56 6.79 11 12.5 10 4.83 20
AGIF+OF [85]30.5 7.17 27 15.6 35 4.93 29 10.2 62 22.1 57 5.16 39 12.5 39 21.2 39 4.88 34 12.5 24 31.7 43 6.76 30 4.78 12 5.73 12 3.91 21 7.85 10 16.3 10 5.20 7 1.83 24 3.10 27 1.71 33 10.8 39 17.6 34 11.0 44
Layers++ [37]31.2 7.19 28 15.7 37 5.08 35 6.15 2 14.8 5 3.42 12 7.83 8 14.0 9 4.84 33 10.9 11 26.9 14 6.19 14 4.83 17 5.84 23 4.36 49 12.4 68 25.2 62 10.5 72 2.43 58 3.56 69 1.92 41 8.66 24 16.1 24 7.77 33
NNF-EAC [103]31.5 6.83 14 14.9 16 4.80 23 7.59 16 18.0 19 4.31 22 9.03 13 16.1 15 3.09 9 13.2 47 32.2 47 7.10 49 5.06 41 5.96 35 4.02 28 8.19 14 17.1 11 6.04 17 1.79 21 3.30 44 1.38 15 17.0 73 27.6 72 18.0 94
LME [70]32.4 7.04 21 15.6 35 4.53 12 6.68 6 16.9 10 2.85 2 13.6 48 22.6 44 12.0 95 11.5 15 27.8 15 6.39 18 5.03 36 5.93 32 4.52 56 12.4 68 27.0 74 10.9 78 1.76 18 3.38 51 1.43 17 6.62 9 12.5 10 2.86 8
Efficient-NL [60]32.5 7.43 47 16.2 55 4.85 24 7.73 20 18.1 20 4.58 24 14.0 51 23.6 52 4.47 28 13.1 44 32.9 54 7.50 61 4.77 11 5.77 18 3.65 11 8.08 12 16.1 9 5.25 8 2.48 62 3.37 50 3.12 76 6.91 12 12.1 8 5.78 24
NL-TV-NCC [25]33.8 6.92 17 14.6 12 3.96 3 8.32 32 19.8 37 2.84 1 15.4 60 26.0 64 3.92 18 10.8 10 26.6 12 5.58 11 5.09 42 6.00 41 4.07 31 11.1 54 23.5 49 10.5 72 2.09 41 3.06 21 2.27 54 11.6 45 20.4 45 9.14 38
IROF++ [58]33.8 7.44 48 16.1 51 5.11 37 8.61 37 19.4 30 5.12 37 12.3 37 21.0 38 5.12 36 12.8 34 32.1 45 7.13 51 4.88 19 5.76 16 3.89 20 9.01 21 18.9 31 6.76 26 1.78 20 3.22 40 1.23 5 10.4 36 18.5 40 13.3 57
TC/T-Flow [76]34.5 6.31 2 13.3 2 4.85 24 11.5 79 23.6 72 6.67 68 13.4 47 23.2 50 3.00 8 14.2 61 36.9 88 6.33 16 4.72 7 5.64 6 3.64 10 7.70 8 17.1 11 5.61 12 2.00 34 3.45 62 2.86 70 10.9 41 18.2 38 3.59 13
PH-Flow [101]34.8 7.38 42 15.9 44 5.22 42 9.30 45 19.5 31 5.71 54 9.62 15 17.2 17 5.09 35 13.4 50 34.5 65 7.10 49 4.82 14 5.73 12 3.82 18 8.36 16 17.1 11 5.35 10 2.66 67 3.43 59 3.42 85 7.13 15 13.3 15 5.33 23
LSM [39]35.2 7.06 22 15.2 27 5.21 41 9.65 51 21.2 49 5.48 51 11.9 33 20.2 29 5.33 41 12.0 18 30.0 22 6.84 37 5.14 49 6.13 58 4.62 62 9.12 24 18.1 20 6.53 22 2.13 46 3.11 29 2.11 46 8.09 20 14.3 17 6.76 30
Classic+CPF [83]35.3 7.31 36 15.8 40 5.09 36 9.93 57 22.1 57 5.05 35 13.3 45 22.4 43 4.64 31 12.5 24 32.0 44 6.79 32 4.87 18 5.83 22 3.99 26 7.43 6 15.6 6 5.32 9 2.26 54 3.21 39 2.78 68 9.89 33 16.5 26 13.8 60
CombBMOF [113]35.5 7.30 35 15.1 23 4.61 17 8.08 27 18.2 22 3.69 17 10.2 19 18.1 22 2.47 4 11.2 13 28.1 16 6.67 26 4.82 14 5.76 16 4.13 39 13.3 79 22.1 40 14.1 101 2.90 76 4.33 93 2.14 47 11.8 47 21.0 48 3.23 11
MLDP_OF [89]35.8 7.02 19 14.5 9 4.89 27 7.00 10 17.0 11 3.34 10 14.3 53 24.0 53 3.73 14 12.9 37 34.8 70 5.88 12 4.89 21 5.69 9 3.92 24 8.18 13 17.4 15 7.22 37 3.64 95 3.68 71 5.99 109 13.2 55 20.5 46 9.16 39
Sparse-NonSparse [56]35.8 7.26 33 15.7 37 5.22 42 9.83 54 21.6 52 5.45 50 12.2 35 20.8 34 5.35 43 12.1 20 30.0 22 6.86 39 5.14 49 6.13 58 4.58 59 9.16 25 18.5 26 6.58 23 2.05 38 3.03 18 2.06 45 7.62 17 13.8 16 5.98 25
HAST [109]36.8 7.11 25 16.0 47 4.27 4 8.90 39 17.4 16 7.54 77 6.79 1 12.4 3 1.56 1 14.8 67 37.2 89 6.63 22 4.68 6 5.70 10 2.91 2 10.5 46 20.7 35 11.1 79 3.74 98 4.39 96 5.40 108 5.74 5 10.9 5 2.09 1
WLIF-Flow [93]36.9 6.80 13 14.9 16 4.60 16 6.93 9 16.7 9 4.11 20 10.3 20 18.2 23 4.11 22 12.7 33 30.9 31 6.67 26 6.60 115 7.87 119 5.60 89 8.54 18 17.4 15 6.03 16 1.85 25 3.18 36 1.67 30 14.4 58 23.2 58 15.3 68
Ramp [62]38.0 7.32 38 15.8 40 5.20 40 8.74 38 19.8 37 5.27 45 11.4 28 19.7 28 5.40 45 12.5 24 31.6 41 6.86 39 4.97 27 5.88 26 4.08 33 9.04 22 18.3 23 7.00 33 2.65 66 3.36 48 4.00 97 8.65 23 15.3 21 11.7 49
PMF [73]38.0 7.59 53 16.5 65 4.94 30 7.64 18 18.3 25 3.66 15 7.48 4 13.2 5 2.31 2 14.7 66 36.3 86 6.84 37 4.66 5 5.64 6 3.18 6 9.85 33 21.3 37 9.22 57 3.62 94 5.25 116 3.70 93 7.12 14 13.2 14 6.82 31
Classic+NL [31]39.3 7.40 45 16.1 51 5.36 51 9.49 50 20.9 45 5.44 49 12.3 37 20.8 34 5.20 39 12.5 24 31.3 37 6.82 34 5.03 36 5.98 39 4.18 43 8.94 20 17.5 17 5.91 14 2.29 56 3.44 61 2.17 48 9.88 32 17.0 31 11.9 51
FlowFields+ [130]39.5 8.95 86 17.2 78 7.26 93 7.52 15 17.3 14 5.15 38 10.4 21 17.6 19 6.16 51 10.1 8 24.5 10 6.49 20 5.11 44 6.00 41 4.57 58 9.28 28 22.2 41 6.70 25 1.77 19 3.20 37 1.51 25 13.1 54 21.8 51 15.6 71
CostFilter [40]40.1 7.36 40 15.7 37 4.96 32 7.76 21 18.1 20 3.83 19 7.07 2 12.4 3 3.12 10 14.6 65 36.3 86 6.57 21 4.82 14 5.81 20 3.54 8 12.4 68 20.5 34 10.3 70 3.86 103 5.96 119 4.36 100 8.86 27 16.7 29 3.72 15
FMOF [94]41.0 7.14 26 15.5 33 5.28 46 10.4 66 22.7 63 5.42 46 10.8 24 19.0 27 3.90 17 12.2 22 31.0 32 6.63 22 4.98 29 5.97 36 4.10 34 10.0 40 17.2 14 6.98 31 2.46 60 3.40 54 4.60 101 14.7 60 23.2 58 10.1 42
ProbFlowFields [128]42.1 8.84 84 17.9 90 6.90 88 7.20 13 17.3 14 4.75 31 11.6 31 20.5 32 6.31 54 8.48 6 22.0 6 5.26 8 5.25 63 6.24 73 4.67 71 9.96 38 23.9 57 6.99 32 1.64 9 2.80 7 1.41 16 14.7 60 25.5 64 14.7 64
TV-L1-MCT [64]42.2 7.42 46 16.1 51 5.22 42 9.84 55 21.6 52 5.20 41 14.3 53 24.7 56 5.27 40 12.5 24 31.2 36 7.21 54 5.01 32 5.90 28 4.41 52 9.10 23 18.8 30 7.14 35 2.17 49 2.78 5 4.69 102 9.81 31 16.8 30 11.5 46
IIOF-NLDP [131]42.3 7.31 36 15.4 30 4.42 7 8.37 33 19.9 40 3.01 5 15.4 60 25.9 62 4.00 20 7.56 3 18.6 5 4.73 6 6.04 101 7.28 113 5.15 81 10.1 41 21.3 37 9.54 61 1.74 15 2.90 15 1.50 23 15.8 64 21.6 50 20.5 107
RNLOD-Flow [121]42.5 6.72 10 14.9 16 4.39 6 9.09 41 21.0 46 5.06 36 15.2 59 25.8 60 5.42 46 12.6 31 32.2 47 6.64 24 5.46 78 6.48 88 4.34 48 8.35 15 17.8 19 6.16 19 2.82 73 3.90 82 3.36 83 9.02 28 16.1 24 9.83 40
SVFilterOh [111]42.8 7.94 59 17.5 84 4.94 30 8.27 30 19.8 37 3.66 15 9.83 18 17.7 21 4.59 29 13.4 50 34.5 65 6.82 34 4.89 21 5.93 32 3.15 5 10.4 45 22.7 43 9.35 59 3.51 91 4.72 105 4.17 98 7.68 18 14.3 17 4.90 21
Complementary OF [21]43.2 7.37 41 15.1 23 5.30 49 9.46 47 22.5 61 4.63 25 13.0 40 22.8 46 4.04 21 14.8 67 37.9 95 6.87 41 4.97 27 5.86 25 4.39 51 11.0 52 24.4 58 8.06 44 1.79 21 2.79 6 2.22 51 12.2 48 22.0 53 11.2 45
ALD-Flow [66]43.5 6.69 9 14.4 8 4.54 14 12.5 88 25.7 84 6.93 71 14.3 53 24.9 57 4.29 24 14.8 67 35.3 77 7.04 47 4.98 29 5.92 30 3.66 12 10.1 41 23.6 51 6.92 29 2.02 36 3.23 41 2.86 70 10.2 35 18.9 41 6.45 29
EPPM w/o HM [88]44.4 7.33 39 14.5 9 5.00 34 7.20 13 17.2 13 3.41 11 11.8 32 20.8 34 3.20 11 12.6 31 31.1 35 7.00 46 5.14 49 6.08 52 4.67 71 12.0 62 22.7 43 9.97 67 4.48 114 3.69 73 6.02 110 10.4 36 17.6 34 11.5 46
FlowFields [110]45.2 9.03 89 17.5 84 7.31 94 8.02 26 18.6 27 5.24 44 11.1 27 18.9 26 6.32 55 11.6 16 28.7 17 7.40 59 5.14 49 6.03 46 4.64 66 10.1 41 23.8 56 7.62 38 1.69 13 2.86 13 1.51 25 12.9 50 22.8 57 14.9 67
TC-Flow [46]45.3 6.56 6 14.1 4 4.48 10 9.25 44 21.4 50 5.03 34 14.9 57 25.8 60 3.93 19 14.5 64 35.9 81 6.97 45 5.01 32 5.97 36 3.63 9 9.98 39 22.8 46 6.83 28 2.11 43 3.25 42 3.61 92 16.7 71 26.8 70 20.0 105
MDP-Flow [26]45.4 6.73 11 14.1 4 5.59 59 6.70 7 16.0 8 4.65 27 9.78 17 17.2 17 6.61 59 13.0 40 34.7 69 6.48 19 5.54 82 6.17 67 5.84 93 10.5 46 23.6 51 7.81 40 1.89 27 3.42 57 1.37 12 20.0 86 32.5 92 19.1 99
S2F-IF [123]46.2 8.86 85 17.2 78 7.05 91 7.84 25 18.3 25 5.20 41 10.8 24 18.5 25 6.21 52 13.0 40 30.7 29 8.15 69 5.09 42 5.98 39 4.58 59 9.62 31 22.7 43 7.20 36 1.92 30 3.77 78 1.73 34 10.7 38 18.4 39 12.8 56
SRR-TVOF-NL [91]48.0 7.83 56 15.4 30 5.99 72 13.2 94 26.1 87 8.61 84 13.2 43 22.0 41 6.78 60 13.6 55 30.3 25 7.34 58 4.72 7 5.56 4 4.04 30 9.87 35 19.8 33 8.30 48 3.25 86 3.73 76 3.18 79 6.93 13 12.9 13 4.92 22
SimpleFlow [49]48.2 7.56 52 16.2 55 5.59 59 9.48 48 21.0 46 5.68 53 17.1 71 27.0 73 6.29 53 13.0 40 32.8 52 7.09 48 5.18 57 6.16 63 4.62 62 8.75 19 17.6 18 6.81 27 2.13 46 3.50 65 2.30 55 9.78 30 17.4 33 7.10 32
IROF-TV [53]48.4 7.55 51 16.1 51 5.46 56 9.23 43 21.8 54 5.88 56 13.3 45 22.2 42 5.50 48 12.8 34 31.6 41 7.28 55 5.12 45 6.09 53 4.67 71 13.3 79 29.6 91 9.97 67 1.60 5 3.12 32 1.06 3 11.5 44 21.3 49 11.5 46
ACK-Prior [27]49.1 6.39 3 13.4 3 4.38 5 8.58 36 19.7 34 3.63 14 11.4 28 20.3 31 3.82 16 12.5 24 33.7 59 5.09 7 5.53 80 6.42 85 4.76 74 15.0 93 28.4 80 11.4 81 3.54 92 4.36 95 4.84 104 13.2 55 20.2 44 8.94 36
HBM-GC [105]49.8 8.35 77 18.4 95 4.98 33 7.66 19 18.2 22 5.20 41 13.8 49 24.2 55 5.34 42 12.0 18 30.5 27 6.83 36 5.04 38 6.02 45 4.43 53 9.34 29 15.5 5 7.92 41 3.03 81 4.57 101 2.51 63 16.1 66 26.0 66 17.9 92
2DHMM-SAS [92]51.0 7.39 44 15.9 44 5.25 45 10.7 70 23.4 69 5.43 47 18.0 75 27.0 73 7.95 79 13.1 44 32.7 50 7.28 55 4.89 21 5.78 19 4.01 27 9.88 36 18.2 21 6.37 21 2.50 65 3.39 52 3.37 84 13.8 57 22.4 56 15.5 70
Sparse Occlusion [54]51.8 7.38 42 15.8 40 5.28 46 8.25 29 19.9 40 4.71 29 15.5 63 26.3 68 4.63 30 13.5 54 33.3 58 6.92 43 5.25 63 6.26 74 4.11 36 9.70 32 21.1 36 5.94 15 4.85 115 5.95 118 3.71 94 10.8 39 20.0 43 8.01 35
DPOF [18]52.2 8.43 79 16.8 68 6.36 80 10.7 70 20.7 43 9.52 90 8.72 12 15.4 12 3.63 12 11.3 14 29.1 19 6.09 13 5.34 70 6.18 68 5.38 86 12.2 64 22.4 42 8.06 44 5.27 116 3.55 66 6.79 115 8.85 26 16.5 26 4.24 18
COFM [59]52.8 8.49 81 18.5 97 5.95 68 8.44 34 18.7 28 4.71 29 13.0 40 22.9 47 5.84 50 13.8 57 35.2 75 6.78 31 5.63 87 6.58 91 5.94 94 11.7 59 23.5 49 9.80 64 2.29 56 3.20 37 2.72 66 6.42 8 12.1 8 3.07 10
OFH [38]52.8 7.25 32 15.0 22 5.29 48 11.1 74 25.0 80 7.19 74 18.6 80 29.1 85 5.56 49 16.0 79 40.8 112 7.44 60 5.05 39 5.92 30 4.28 46 11.4 56 26.0 67 8.73 52 1.64 9 3.04 19 1.36 11 12.5 49 23.3 60 7.79 34
ROF-ND [107]53.4 7.02 19 14.7 13 4.66 19 8.18 28 19.5 31 4.66 28 15.5 63 25.9 62 5.47 47 4.68 1 12.6 1 2.76 3 5.93 95 7.12 110 5.27 83 12.2 64 25.4 64 9.09 56 4.23 111 4.20 89 3.32 82 14.7 60 22.1 55 18.8 97
PGM-C [120]54.0 9.43 93 18.5 97 7.51 98 9.84 55 23.4 69 6.05 57 12.0 34 20.6 33 7.17 66 15.8 76 35.3 77 9.67 80 5.20 60 6.11 55 4.66 68 9.86 34 23.7 53 7.06 34 1.63 7 2.84 10 1.49 22 11.1 42 20.8 47 6.23 28
S2D-Matching [84]54.4 8.25 73 18.0 92 5.76 65 10.9 72 22.8 64 5.71 54 17.3 72 28.5 79 6.38 57 12.1 20 30.3 25 6.65 25 5.15 53 6.13 58 4.63 64 9.18 26 19.5 32 6.69 24 2.66 67 3.35 47 3.50 86 11.6 45 19.9 42 14.7 64
OAR-Flow [125]54.6 8.08 67 16.8 68 6.19 77 17.4 107 28.9 105 12.3 104 16.3 69 26.7 72 7.01 63 15.1 71 35.2 75 7.31 57 5.17 55 6.16 63 4.24 44 9.51 30 22.9 47 5.52 11 1.54 4 2.98 17 1.59 29 9.10 29 17.1 32 3.69 14
TCOF [69]55.1 7.46 49 15.1 23 5.70 62 9.13 42 21.1 48 5.43 47 19.6 85 29.8 90 8.31 82 12.9 37 31.0 32 7.17 53 6.02 100 7.00 105 4.59 61 7.92 11 18.4 25 6.11 18 3.78 100 4.09 86 5.18 106 8.51 22 15.9 23 3.80 16
ComplOF-FED-GPU [35]55.4 7.49 50 15.4 30 5.37 53 12.1 85 26.5 93 7.51 76 13.1 42 22.7 45 4.33 27 15.6 74 37.7 94 7.55 63 4.94 24 5.82 21 4.13 39 12.3 66 27.7 77 8.48 50 2.49 64 3.04 19 3.56 89 12.9 50 23.9 61 9.01 37
AggregFlow [97]58.2 10.3 102 21.3 120 6.91 89 15.1 100 27.6 100 10.7 95 14.0 51 24.0 53 8.70 84 14.1 59 35.0 72 7.15 52 5.05 39 6.03 46 4.02 28 7.73 9 18.2 21 5.09 6 2.13 46 4.40 97 1.67 30 9.91 34 18.1 37 6.13 27
CPM-Flow [116]60.7 9.46 95 18.6 100 7.51 98 10.0 59 23.7 73 6.20 59 12.2 35 20.9 37 7.13 64 15.8 76 35.6 80 9.71 81 5.20 60 6.12 57 4.65 67 10.9 50 23.7 53 8.90 55 1.73 14 3.15 34 1.51 25 14.5 59 26.0 66 13.6 59
EpicFlow [102]61.9 9.39 92 18.4 95 7.50 97 10.0 59 23.8 75 6.29 61 15.8 66 26.6 71 7.35 72 15.5 73 34.4 64 9.65 78 5.20 60 6.11 55 4.66 68 10.2 44 24.5 59 8.18 47 1.63 7 2.81 9 1.48 19 15.8 64 24.8 62 17.1 88
Aniso-Texture [82]62.8 6.49 5 14.2 7 4.54 14 7.83 24 19.5 31 4.37 23 22.0 109 33.6 120 8.47 83 10.3 9 25.4 11 5.55 9 5.19 59 6.10 54 4.45 54 18.7 114 33.0 114 20.1 123 4.07 109 4.62 103 2.47 62 20.1 87 28.5 74 20.7 108
Occlusion-TV-L1 [63]63.5 7.73 55 16.4 62 5.36 51 9.48 48 22.4 60 6.18 58 19.2 81 30.0 93 7.14 65 14.4 63 33.8 61 7.91 67 5.31 69 6.27 76 4.47 55 11.9 60 27.9 78 8.04 42 2.09 41 3.08 22 1.50 23 21.9 98 35.3 108 17.5 89
RFlow [90]64.1 7.10 24 14.9 16 5.40 54 8.98 40 21.9 55 5.19 40 18.5 79 29.3 87 5.35 43 18.1 98 44.9 130 9.86 84 5.12 45 6.01 44 4.38 50 13.0 76 29.3 89 9.93 66 2.28 55 2.85 11 3.52 87 20.3 89 33.3 98 16.1 79
DeepFlow2 [108]64.5 8.14 69 16.6 66 5.96 70 14.1 95 26.5 93 10.2 91 15.8 66 26.2 67 6.46 58 16.5 87 37.4 92 9.54 76 5.01 32 5.94 34 3.72 15 10.7 48 25.1 61 8.08 46 1.92 30 3.12 32 2.45 60 20.3 89 32.1 89 16.3 81
Adaptive [20]65.5 7.94 59 17.0 73 5.33 50 10.2 62 23.9 76 6.28 60 21.2 97 31.7 108 7.69 76 13.6 55 29.8 21 7.87 66 4.88 19 5.75 14 3.71 14 13.2 77 28.9 86 9.72 63 3.21 85 4.71 104 2.91 72 19.3 83 30.5 81 15.6 71
TF+OM [100]66.8 7.97 63 16.8 68 5.98 71 9.40 46 20.7 43 6.33 62 15.4 60 22.9 47 17.5 104 13.4 50 31.5 38 8.10 68 5.13 48 6.06 51 4.66 68 13.9 84 29.3 89 14.0 99 2.47 61 4.09 86 2.00 43 18.3 76 29.4 78 19.3 101
Steered-L1 [118]67.5 5.97 1 12.7 1 4.67 21 7.14 12 18.2 22 4.63 25 13.2 43 23.3 51 5.12 36 15.2 72 38.1 96 7.54 62 5.85 94 6.73 95 6.98 113 13.7 83 26.5 69 11.7 86 6.39 121 4.25 91 13.3 124 22.3 101 32.7 94 20.3 106
Aniso. Huber-L1 [22]68.3 7.96 62 16.3 60 6.10 75 11.4 76 24.7 78 6.77 69 20.6 92 29.6 89 7.26 68 13.2 47 29.2 20 7.77 65 5.52 79 6.58 91 4.29 47 12.4 68 26.7 72 8.83 54 2.93 79 3.68 71 3.10 75 16.5 70 27.4 71 13.9 61
LocallyOriented [52]70.2 9.89 98 19.9 111 6.55 82 14.7 99 27.7 101 11.0 97 21.7 103 31.9 110 7.32 71 13.3 49 30.5 27 8.32 70 5.16 54 6.04 48 4.11 36 9.90 37 21.6 39 8.75 53 2.20 50 3.43 59 2.17 48 18.3 76 26.0 66 19.6 102
SIOF [67]70.8 8.21 72 17.0 73 5.49 57 13.0 93 27.4 99 8.63 85 20.1 90 29.0 83 16.3 103 16.8 89 37.3 91 10.0 88 5.40 76 6.34 81 4.88 76 11.6 57 25.4 64 10.1 69 1.81 23 3.27 43 1.34 9 15.1 63 25.1 63 11.9 51
DeepFlow [86]72.5 8.73 82 17.1 77 6.26 78 15.3 103 26.7 96 11.9 103 17.3 72 26.5 70 14.1 101 18.6 104 42.8 124 11.0 94 5.00 31 5.91 29 3.75 17 11.2 55 26.2 68 8.35 49 1.88 26 2.80 7 2.60 64 22.5 103 33.6 99 17.5 89
SegOF [10]73.0 9.43 93 17.7 86 8.06 105 12.0 84 22.9 67 10.4 94 17.0 70 26.1 66 12.6 97 13.8 57 26.8 13 11.2 96 5.53 80 6.26 74 6.06 97 19.0 115 35.6 126 18.8 117 1.37 2 2.60 2 0.83 2 16.4 68 30.0 79 14.0 62
CRTflow [80]73.4 7.92 57 16.0 47 5.91 67 11.4 76 23.7 73 6.55 66 19.8 87 30.1 94 7.77 77 17.2 94 41.3 115 9.76 82 5.34 70 6.30 77 3.69 13 15.8 98 31.0 102 14.3 102 2.12 45 2.92 16 2.35 57 19.2 82 32.9 96 15.3 68
AdaConv-v1 [126]73.8 13.0 117 16.4 62 8.62 112 10.2 62 9.17 1 11.5 100 10.7 22 11.1 1 7.66 74 16.3 84 17.8 2 14.4 108 10.8 127 8.97 124 16.5 127 19.7 118 18.7 27 19.4 119 19.7 130 17.1 130 8.18 119 3.51 1 4.92 1 2.42 3
SepConv-v1 [127]73.8 13.0 117 16.4 62 8.62 112 10.2 62 9.17 1 11.5 100 10.7 22 11.1 1 7.66 74 16.3 84 17.8 2 14.4 108 10.8 127 8.97 124 16.5 127 19.7 118 18.7 27 19.4 119 19.7 130 17.1 130 8.18 119 3.51 1 4.92 1 2.42 3
TriangleFlow [30]73.9 8.08 67 17.0 73 5.12 38 11.7 82 26.1 87 6.98 73 19.5 83 30.3 95 6.34 56 12.9 37 33.0 55 6.71 29 7.00 118 8.16 122 6.63 110 12.8 75 24.5 59 10.5 72 3.59 93 5.17 114 3.27 81 12.9 50 21.9 52 12.1 53
TriFlow [95]75.4 9.01 88 18.5 97 6.60 84 11.4 76 26.4 92 7.40 75 20.9 94 29.9 92 20.9 112 12.2 22 30.8 30 6.95 44 5.26 67 6.16 63 4.88 76 12.0 62 26.5 69 11.6 84 6.97 122 4.54 99 6.57 114 13.0 53 22.0 53 9.86 41
Brox et al. [5]75.6 8.46 80 16.7 67 6.56 83 11.3 75 26.2 89 6.94 72 15.0 58 25.4 59 6.89 62 17.4 95 38.8 100 9.80 83 5.99 98 6.88 102 6.26 102 14.1 86 31.1 105 12.0 88 2.03 37 3.41 56 1.14 4 19.1 81 30.3 80 12.1 53
p-harmonic [29]77.2 8.15 71 16.2 55 6.50 81 9.66 52 22.6 62 6.57 67 21.2 97 31.2 101 9.65 87 15.7 75 33.9 62 10.0 88 5.18 57 6.04 48 5.31 84 14.3 88 30.3 97 12.2 91 3.10 82 3.55 66 1.91 40 23.1 106 34.8 106 17.8 91
Fusion [6]78.2 8.76 83 17.7 86 7.01 90 7.82 23 19.7 34 4.78 32 10.9 26 18.4 24 7.23 67 12.8 34 32.6 49 8.32 70 7.04 119 8.11 120 6.57 108 14.9 91 28.3 79 13.2 94 4.37 113 5.18 115 2.77 67 26.2 116 38.6 118 26.4 118
CBF [12]78.4 7.23 31 14.9 16 5.16 39 9.95 58 21.9 55 7.69 79 17.6 74 27.0 73 7.28 70 16.4 86 39.3 104 9.18 74 6.34 111 7.35 116 6.11 100 13.4 81 28.4 80 8.52 51 5.54 118 5.11 111 6.41 112 18.7 78 30.8 82 16.4 82
FlowNet2 [122]78.5 12.8 116 23.3 125 8.09 106 16.8 106 28.5 102 12.5 105 13.9 50 21.6 40 12.0 95 15.0 70 34.0 63 9.98 87 5.36 74 6.23 72 4.94 78 14.0 85 29.9 94 12.2 91 3.36 87 6.60 123 2.24 52 8.83 25 16.6 28 3.02 9
TV-L1-improved [17]79.2 7.64 54 16.2 55 5.67 61 10.1 61 23.4 69 6.38 63 21.3 99 32.0 113 9.27 86 17.5 96 42.1 119 9.54 76 5.25 63 6.13 58 4.07 31 14.5 89 30.4 99 11.4 81 3.38 88 5.02 109 2.99 74 19.6 85 32.1 89 16.6 84
Local-TV-L1 [65]80.0 9.74 97 17.9 90 6.89 87 18.4 111 29.4 107 14.9 111 24.4 114 30.8 99 20.2 110 19.4 109 42.4 122 12.7 104 5.35 73 6.00 41 4.10 34 13.6 82 28.9 86 9.26 58 1.62 6 2.58 1 1.48 19 20.3 89 32.0 88 16.4 82
SuperFlow [81]80.1 9.27 91 17.3 81 6.63 85 11.7 82 22.8 64 8.84 87 19.3 82 28.0 78 18.4 108 16.9 91 38.2 97 9.89 85 5.44 77 6.32 78 5.61 90 11.9 60 26.6 71 9.56 62 2.92 77 4.08 85 1.79 36 20.1 87 32.4 91 15.8 77
CLG-TV [48]80.3 7.94 59 16.2 55 5.80 66 10.5 67 24.0 77 6.44 64 19.9 88 29.8 90 6.83 61 14.1 59 31.5 38 7.73 64 5.98 96 7.01 106 5.15 81 14.8 90 31.0 102 12.2 91 4.20 110 4.80 106 5.22 107 19.3 83 32.6 93 15.7 75
Classic++ [32]80.9 8.05 65 17.4 83 6.09 74 11.5 79 26.3 91 6.91 70 18.1 77 28.7 81 8.18 80 16.2 80 39.0 102 8.57 72 5.36 74 6.33 79 4.54 57 15.0 93 30.4 99 11.6 84 2.70 70 3.55 66 2.94 73 21.9 98 34.0 102 17.9 92
Rannacher [23]82.0 8.07 66 16.8 68 6.15 76 10.6 68 24.8 79 6.51 65 21.9 107 32.6 119 10.9 90 18.4 101 43.3 126 10.5 90 5.27 68 6.18 68 4.17 42 15.5 96 32.3 111 12.0 88 2.69 69 3.57 70 2.68 65 17.8 75 31.0 83 16.1 79
F-TV-L1 [15]82.5 8.41 78 16.8 68 6.31 79 18.0 110 29.7 108 12.8 106 21.6 101 30.5 96 10.1 89 18.2 99 42.3 121 9.65 78 5.02 35 5.97 36 3.91 21 14.1 86 30.8 101 10.6 76 2.79 71 4.90 107 2.35 57 20.5 92 32.7 94 15.6 71
DF-Auto [115]85.3 10.7 107 19.8 110 7.42 96 16.3 105 25.3 81 13.1 108 19.5 83 28.5 79 17.8 106 16.9 91 37.2 89 10.8 92 6.76 117 8.12 121 5.56 88 10.8 49 25.2 62 6.95 30 3.69 97 4.92 108 1.48 19 17.7 74 27.9 73 14.6 63
BriefMatch [124]85.5 6.91 16 14.8 15 4.85 24 11.0 73 22.8 64 8.24 82 9.42 14 16.6 16 5.13 38 16.7 88 40.7 111 8.61 73 9.76 125 10.6 128 14.1 125 18.2 111 31.0 102 17.4 113 9.26 126 6.60 123 20.9 130 28.0 120 36.5 110 34.9 124
StereoOF-V1MT [119]85.6 8.14 69 15.8 40 5.42 55 17.6 108 34.3 126 10.3 92 21.6 101 31.5 105 7.27 69 15.8 76 32.7 50 10.7 91 5.62 86 6.40 83 6.00 96 16.6 101 30.0 95 15.5 104 2.01 35 3.08 22 3.15 77 33.6 125 44.0 126 32.8 122
Dynamic MRF [7]87.3 8.32 76 17.3 81 5.95 68 12.3 86 28.5 102 7.75 80 19.6 85 31.8 109 7.56 73 18.4 101 42.2 120 11.6 100 5.25 63 6.16 63 4.80 75 17.7 105 34.4 122 16.6 109 1.89 27 2.63 3 3.21 80 30.3 122 43.6 125 29.0 120
Bartels [41]87.6 8.31 75 17.7 86 5.51 58 8.46 35 20.6 42 4.89 33 14.8 56 26.0 64 7.89 78 18.5 103 43.6 127 11.1 95 6.18 104 6.51 90 8.63 121 15.6 97 32.6 113 13.9 98 3.86 103 4.56 100 7.09 116 22.5 103 36.5 110 18.4 95
GraphCuts [14]88.2 9.24 90 17.2 78 7.53 100 22.1 120 33.0 121 16.8 118 15.9 68 23.1 49 15.6 102 14.2 61 28.9 18 9.34 75 5.83 93 6.82 98 6.25 101 18.5 113 29.7 92 12.0 88 2.85 74 3.39 52 3.52 87 23.8 110 36.1 109 19.1 99
Shiralkar [42]88.6 7.92 57 15.2 27 5.73 63 14.6 98 30.3 109 9.04 88 21.7 103 31.2 101 9.77 88 19.6 111 41.5 116 13.2 106 5.17 55 6.04 48 4.63 64 18.0 110 31.1 105 14.7 103 3.67 96 3.40 54 4.74 103 23.9 111 36.6 112 18.9 98
StereoFlow [44]90.0 16.2 127 22.6 123 13.9 128 22.1 120 31.6 113 18.8 122 24.7 115 30.6 98 21.2 114 23.3 122 39.9 105 19.6 122 5.98 96 6.22 71 7.11 115 11.6 57 27.6 75 8.05 43 1.36 1 2.85 11 0.65 1 20.7 93 33.7 100 17.0 87
Filter Flow [19]90.1 10.5 105 19.4 106 8.33 108 12.5 88 25.8 85 8.69 86 19.9 88 27.0 73 21.7 115 19.0 106 33.1 57 15.9 112 5.34 70 6.21 70 5.37 85 16.2 100 26.7 72 15.5 104 3.46 90 4.48 98 2.26 53 23.3 109 31.5 86 18.5 96
Second-order prior [8]90.7 8.04 64 16.3 60 6.01 73 12.5 88 26.5 93 9.10 89 21.0 95 31.1 100 9.23 85 16.2 80 36.2 85 9.93 86 5.70 90 6.67 94 5.09 80 20.7 122 34.1 120 21.0 124 3.76 99 3.89 81 4.25 99 18.9 79 31.8 87 19.9 104
CNN-flow-warp+ref [117]91.4 9.91 99 19.6 108 7.85 102 10.6 68 22.9 67 8.55 83 21.3 99 32.1 115 11.9 94 18.7 105 42.0 118 11.3 97 5.56 83 6.34 81 6.08 99 12.3 66 27.6 75 10.8 77 2.06 39 3.69 73 3.16 78 33.3 123 40.4 121 34.7 123
IAOF2 [51]92.5 10.0 100 19.9 111 7.91 104 14.4 97 26.7 96 10.9 96 22.0 109 32.2 116 17.6 105 19.1 107 33.0 55 17.1 115 5.81 92 6.86 100 4.94 78 12.6 73 25.9 66 11.9 87 4.26 112 4.11 88 7.75 118 16.2 67 26.5 69 13.5 58
FlowNetS+ft+v [112]92.9 8.96 87 17.8 89 6.83 86 14.2 96 26.2 89 11.1 98 22.3 111 32.2 116 12.7 98 16.8 89 36.1 84 10.9 93 6.31 110 7.29 114 6.26 102 12.5 72 28.8 84 9.88 65 3.84 102 6.75 125 6.30 111 16.4 68 29.0 76 14.7 64
Ad-TV-NDC [36]94.7 12.1 114 18.6 100 10.7 121 25.5 125 32.0 117 22.2 125 29.3 126 34.3 124 22.8 119 16.9 91 32.8 52 12.2 102 5.99 98 7.20 112 3.83 19 12.6 73 28.5 82 10.3 70 2.79 71 4.07 84 1.78 35 24.1 112 31.4 85 25.1 116
2D-CLG [1]96.2 15.2 125 24.5 128 11.2 123 15.1 100 25.9 86 13.5 109 27.5 120 33.9 121 24.7 126 22.2 119 38.3 98 19.0 121 5.67 88 6.44 86 6.29 105 17.5 103 34.3 121 16.4 108 1.47 3 2.68 4 1.54 28 21.7 97 34.6 105 16.9 86
LDOF [28]97.4 9.66 96 18.9 103 7.13 92 15.1 100 29.1 106 11.3 99 15.6 65 25.1 58 10.9 90 20.9 115 43.1 125 14.9 110 6.12 102 7.01 106 6.26 102 16.0 99 31.2 107 13.2 94 3.89 105 5.61 117 8.96 121 19.0 80 33.0 97 11.7 49
UnFlow [129]97.5 16.0 126 25.1 129 11.3 124 12.7 92 22.2 59 11.7 102 20.4 91 27.6 77 13.6 99 20.8 113 36.0 82 17.9 118 6.34 111 6.89 103 7.74 119 17.7 105 33.5 117 17.1 112 2.92 77 4.33 93 1.37 12 20.7 93 37.3 116 15.6 71
Nguyen [33]98.2 11.4 109 19.6 108 8.44 110 21.0 117 31.7 115 17.7 120 29.8 128 36.3 128 24.1 125 18.2 99 34.6 68 13.9 107 6.28 106 6.85 99 7.51 118 15.4 95 33.0 114 14.0 99 2.24 53 3.11 29 1.79 36 21.6 96 33.9 101 15.8 77
IAOF [50]99.1 10.1 101 18.6 100 7.89 103 22.2 122 33.7 124 15.9 114 33.0 130 39.2 131 23.7 122 16.2 80 32.1 45 11.6 100 5.80 91 6.87 101 5.39 87 17.8 107 30.1 96 11.4 81 3.13 84 3.69 73 3.88 96 22.4 102 29.3 77 21.6 111
SPSA-learn [13]99.4 11.3 108 19.4 106 8.52 111 20.9 115 34.2 125 16.2 117 26.5 118 33.9 121 22.4 118 22.5 121 39.9 105 18.9 120 5.59 84 6.41 84 5.94 94 17.8 107 31.4 109 18.3 116 2.11 43 3.11 29 1.37 12 24.3 113 35.1 107 19.7 103
Learning Flow [11]100.2 8.28 74 17.0 73 5.75 64 12.3 86 28.5 102 7.79 81 18.4 78 28.7 81 8.29 81 22.2 119 40.6 110 17.3 116 8.52 120 10.3 127 7.04 114 19.9 120 35.1 124 16.8 111 3.11 83 4.59 102 3.59 91 26.9 118 40.0 120 20.9 110
HBpMotionGpu [43]100.8 12.2 115 22.1 121 8.34 109 17.9 109 31.4 112 14.7 110 29.2 125 37.8 130 20.6 111 19.1 107 44.6 129 11.5 99 5.60 85 6.45 87 6.06 97 13.2 77 28.8 84 11.1 79 3.45 89 3.97 83 2.04 44 23.1 106 34.1 103 20.7 108
Modified CLG [34]103.5 11.6 111 20.5 116 9.14 117 12.6 91 25.4 83 10.3 92 27.4 119 34.4 125 23.8 123 21.8 117 42.7 123 16.7 114 6.18 104 7.06 109 6.57 108 14.9 91 33.0 114 13.4 96 2.45 59 3.80 79 3.81 95 22.0 100 36.6 112 16.8 85
Black & Anandan [4]105.0 10.5 105 18.0 92 8.14 107 21.4 119 32.8 120 16.1 116 26.1 117 32.3 118 21.8 116 20.8 113 38.9 101 16.1 113 6.29 108 7.41 117 5.62 91 17.1 102 31.2 107 13.7 97 4.06 108 5.11 111 2.37 59 23.1 106 34.1 103 15.7 75
GroupFlow [9]105.9 11.5 110 21.2 119 8.71 114 20.5 113 33.0 121 15.7 113 23.6 112 31.6 106 20.1 109 16.2 80 34.5 65 11.3 97 8.86 121 9.84 126 6.50 107 18.2 111 30.3 97 17.4 113 3.82 101 5.05 110 6.55 113 21.1 95 28.8 75 22.4 114
Heeger++ [104]106.4 11.7 112 18.3 94 9.04 116 23.3 124 37.0 130 17.4 119 21.0 95 29.1 85 11.1 92 27.0 126 40.5 109 24.4 125 5.69 89 6.33 79 5.80 92 25.9 129 37.5 129 26.3 129 2.48 62 3.88 80 2.20 50 37.6 128 46.0 129 41.6 130
HCIC-L [99]106.7 14.6 122 21.1 117 9.67 118 42.7 131 36.7 129 46.6 131 21.8 105 30.5 96 17.9 107 21.1 116 35.1 74 18.6 119 6.42 113 6.58 91 7.35 117 17.8 107 28.6 83 16.6 109 12.8 128 14.3 129 15.3 126 16.8 72 25.7 65 12.7 55
2bit-BM-tele [98]106.7 10.3 102 20.1 113 7.40 95 11.5 79 26.7 96 7.57 78 20.6 92 32.0 113 11.4 93 17.9 97 40.3 108 12.3 103 6.29 108 6.80 97 6.93 112 21.0 123 32.0 110 18.0 115 7.61 123 6.57 122 11.1 123 27.5 119 39.4 119 29.2 121
TI-DOFE [24]109.1 14.6 122 21.1 117 11.7 125 22.4 123 31.6 113 19.6 123 28.6 123 31.2 101 26.3 127 26.3 125 36.0 82 25.1 126 6.43 114 7.34 115 6.67 111 19.1 117 33.8 118 18.9 118 2.88 75 3.15 34 2.82 69 25.7 115 36.6 112 22.3 113
Horn & Schunck [3]110.1 11.8 113 19.3 105 8.85 115 20.1 112 33.5 123 15.3 112 25.5 116 31.2 101 24.0 124 26.1 124 38.6 99 23.5 123 6.12 102 6.99 104 6.34 106 19.9 120 33.9 119 19.6 121 3.95 107 4.28 92 2.46 61 25.3 114 37.5 117 22.0 112
BlockOverlap [61]110.1 10.3 102 19.1 104 7.74 101 15.4 104 25.3 81 13.0 107 24.3 113 31.9 110 21.0 113 19.5 110 43.8 128 13.1 105 9.14 122 7.60 118 13.9 124 19.0 115 29.0 88 15.7 106 11.0 127 8.77 127 24.8 131 23.0 105 31.3 84 25.9 117
FFV1MT [106]115.5 13.9 121 23.4 126 10.8 122 20.9 115 34.5 127 16.0 115 21.8 105 29.5 88 13.8 100 31.0 130 41.7 117 29.4 129 9.97 126 6.78 96 17.9 129 23.8 126 35.3 125 25.0 128 2.99 80 4.24 90 3.57 90 37.6 128 46.0 129 41.6 130
SILK [79]116.5 13.2 119 22.4 122 10.2 120 20.5 113 32.2 119 17.7 120 29.0 124 34.1 123 23.4 121 22.0 118 40.8 112 17.3 116 6.28 106 7.17 111 7.12 116 21.7 124 35.8 127 19.6 121 5.44 117 3.45 62 10.8 122 29.3 121 40.6 123 26.7 119
Adaptive flow [45]117.1 13.2 119 20.2 114 9.78 119 27.3 127 31.9 116 25.4 126 28.3 122 31.6 106 30.5 130 19.7 112 37.6 93 15.3 111 11.2 129 12.6 129 8.93 122 17.6 104 29.8 93 15.8 107 13.1 129 11.0 128 18.3 127 26.4 117 36.8 115 22.7 115
PGAM+LK [55]118.0 14.9 124 22.8 124 14.1 129 25.5 125 31.3 111 26.6 127 21.9 107 26.4 69 22.2 117 26.0 123 39.1 103 24.0 124 9.61 124 6.49 89 14.1 125 24.3 127 35.0 123 23.0 126 6.19 120 6.24 121 7.26 117 34.3 126 40.4 121 40.8 129
SLK [47]118.2 17.2 130 24.0 127 18.2 130 21.3 118 30.4 110 20.1 124 27.9 121 31.9 110 23.3 120 31.4 131 39.9 105 30.0 131 6.60 115 7.04 108 8.37 120 22.8 125 36.4 128 21.6 125 3.90 106 3.75 77 5.02 105 33.3 123 41.7 124 35.2 125
Pyramid LK [2]124.7 17.0 129 20.4 115 19.4 131 31.7 128 32.1 118 32.5 128 32.9 129 34.8 126 29.9 129 29.4 128 35.3 77 29.9 130 31.7 130 35.5 130 29.8 130 29.9 130 32.3 111 28.0 130 9.01 125 7.93 126 18.9 128 39.9 130 45.4 128 39.0 127
FOLKI [16]125.7 16.2 127 25.9 130 12.7 127 32.5 129 35.6 128 34.7 129 29.4 127 35.1 127 26.9 128 29.1 127 41.0 114 27.9 128 9.58 123 8.90 123 13.5 123 25.7 128 38.3 130 24.5 127 7.71 124 5.13 113 14.5 125 36.5 127 44.4 127 36.1 126
Periodicity [78]129.2 18.0 131 30.5 131 12.3 126 34.0 130 41.4 131 35.6 130 36.5 131 36.5 129 35.2 131 29.7 129 46.6 131 26.8 127 51.8 131 56.5 131 45.5 131 36.7 131 42.4 131 37.1 131 5.99 119 6.16 120 19.3 129 40.1 131 51.1 131 40.4 128
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. Submitted to TIP 2016.
[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-invariantdescriptors in variational optical flow estimation. ICIP 2017.
* 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.