Optical flow evaluation results |
Statistics:
Average
SD
R0.5
R1.0
R2.0
A50
A75
A95
Error type: endpoint angle interpolation normalized interpolation |
SD endpoint 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] | 7.1 | 0.20 19 | 0.44 21 | 0.10 4 | 0.38 2 | 0.90 2 | 0.17 1 | 0.48 3 | 0.87 4 | 0.14 3 | 0.41 3 | 1.24 6 | 0.09 2 | 0.82 1 | 1.02 1 | 0.34 2 | 0.73 2 | 1.77 4 | 0.54 2 | 0.11 20 | 0.12 45 | 0.13 16 | 0.80 3 | 1.37 3 | 0.45 2 |
PMMST [114] | 7.4 | 0.19 8 | 0.44 21 | 0.07 1 | 0.35 1 | 0.80 1 | 0.22 14 | 0.52 6 | 0.93 6 | 0.17 8 | 0.31 2 | 0.88 2 | 0.12 3 | 1.06 3 | 1.32 3 | 0.61 6 | 0.67 1 | 1.57 1 | 0.48 1 | 0.12 26 | 0.11 29 | 0.15 27 | 0.78 2 | 1.36 2 | 0.65 4 |
MDP-Flow2 [68] | 15.8 | 0.19 8 | 0.44 21 | 0.10 4 | 0.40 5 | 0.97 8 | 0.18 6 | 0.55 11 | 1.01 11 | 0.16 6 | 0.69 40 | 1.78 52 | 0.36 23 | 1.30 22 | 1.58 27 | 1.09 31 | 0.82 4 | 1.86 7 | 0.70 9 | 0.12 26 | 0.10 14 | 0.15 27 | 0.89 5 | 1.55 6 | 0.92 7 |
NN-field [71] | 16.1 | 0.21 33 | 0.47 54 | 0.11 16 | 0.38 2 | 0.92 5 | 0.17 1 | 0.51 5 | 0.93 6 | 0.15 5 | 0.43 7 | 1.30 9 | 0.08 1 | 0.83 2 | 1.03 2 | 0.32 1 | 1.72 67 | 1.98 11 | 1.12 63 | 0.12 26 | 0.12 45 | 0.13 16 | 0.85 4 | 1.46 4 | 0.43 1 |
OFLAF [77] | 17.4 | 0.19 8 | 0.44 21 | 0.10 4 | 0.45 14 | 1.08 26 | 0.19 7 | 0.53 8 | 0.98 10 | 0.13 2 | 0.76 61 | 2.04 88 | 0.40 53 | 1.15 8 | 1.43 10 | 0.63 7 | 0.78 3 | 1.71 2 | 0.66 5 | 0.11 20 | 0.10 14 | 0.13 16 | 1.00 7 | 1.50 5 | 1.44 19 |
ComponentFusion [96] | 20.8 | 0.19 8 | 0.44 21 | 0.10 4 | 0.47 22 | 1.15 41 | 0.20 11 | 0.58 12 | 1.06 12 | 0.20 11 | 0.51 9 | 1.52 14 | 0.15 5 | 1.50 51 | 1.81 64 | 1.29 46 | 0.95 11 | 2.13 14 | 0.80 20 | 0.13 35 | 0.10 14 | 0.16 33 | 1.05 11 | 1.72 17 | 1.29 12 |
FlowFields+ [130] | 24.0 | 0.22 56 | 0.43 14 | 0.16 87 | 0.43 11 | 0.90 2 | 0.30 34 | 0.66 15 | 1.13 14 | 0.32 36 | 0.42 4 | 1.14 5 | 0.24 8 | 1.44 42 | 1.69 40 | 1.18 39 | 0.93 8 | 2.13 14 | 0.65 3 | 0.10 14 | 0.09 3 | 0.12 10 | 1.32 41 | 2.10 45 | 1.55 31 |
S2F-IF [123] | 25.4 | 0.22 56 | 0.44 21 | 0.15 75 | 0.45 14 | 0.97 8 | 0.31 42 | 0.70 19 | 1.21 17 | 0.32 36 | 0.61 14 | 1.50 12 | 0.37 33 | 1.49 48 | 1.74 50 | 1.28 43 | 0.94 9 | 2.18 17 | 0.72 12 | 0.10 14 | 0.10 14 | 0.13 16 | 1.07 12 | 1.73 18 | 1.19 10 |
IROF++ [58] | 25.9 | 0.21 33 | 0.47 54 | 0.11 16 | 0.47 22 | 1.05 19 | 0.28 26 | 0.90 36 | 1.54 37 | 0.27 21 | 0.64 23 | 1.59 22 | 0.36 23 | 1.21 14 | 1.48 14 | 0.94 16 | 1.30 30 | 2.41 30 | 0.97 40 | 0.09 6 | 0.11 29 | 0.11 7 | 1.24 27 | 1.93 32 | 1.72 45 |
FlowFields [110] | 27.1 | 0.22 56 | 0.43 14 | 0.16 87 | 0.45 14 | 0.96 7 | 0.31 42 | 0.72 22 | 1.25 22 | 0.33 40 | 0.48 8 | 1.29 8 | 0.27 11 | 1.47 47 | 1.73 48 | 1.25 41 | 0.96 13 | 2.15 16 | 0.72 12 | 0.10 14 | 0.09 3 | 0.13 16 | 1.31 38 | 2.12 46 | 1.49 25 |
Correlation Flow [75] | 27.7 | 0.20 19 | 0.44 21 | 0.08 2 | 0.47 22 | 1.07 23 | 0.17 1 | 1.49 88 | 2.29 100 | 0.35 48 | 0.42 4 | 1.12 4 | 0.24 8 | 1.45 44 | 1.72 45 | 1.28 43 | 1.13 17 | 1.79 5 | 0.89 31 | 0.14 44 | 0.11 29 | 0.17 39 | 1.01 8 | 1.64 11 | 1.10 8 |
NNF-EAC [103] | 30.0 | 0.19 8 | 0.44 21 | 0.11 16 | 0.43 11 | 0.99 14 | 0.24 19 | 0.69 18 | 1.22 18 | 0.20 11 | 0.68 36 | 1.72 40 | 0.37 33 | 1.24 16 | 1.51 16 | 0.98 19 | 1.15 19 | 1.87 9 | 0.89 31 | 0.13 35 | 0.12 45 | 0.16 33 | 1.97 82 | 3.11 104 | 1.94 67 |
CombBMOF [113] | 30.0 | 0.21 33 | 0.46 42 | 0.10 4 | 0.49 31 | 1.10 27 | 0.20 11 | 0.70 19 | 1.24 20 | 0.16 6 | 0.61 14 | 1.58 21 | 0.36 23 | 1.29 20 | 1.56 21 | 0.99 21 | 1.33 32 | 1.85 6 | 1.13 64 | 0.16 57 | 0.15 87 | 0.18 43 | 1.43 49 | 2.33 56 | 1.30 14 |
WLIF-Flow [93] | 30.1 | 0.20 19 | 0.45 34 | 0.10 4 | 0.41 8 | 0.95 6 | 0.23 17 | 0.76 25 | 1.34 27 | 0.27 21 | 0.63 18 | 1.59 22 | 0.34 14 | 1.32 27 | 1.61 31 | 1.06 26 | 1.46 49 | 2.55 42 | 0.98 43 | 0.15 51 | 0.10 14 | 0.19 49 | 1.54 57 | 2.33 56 | 1.89 62 |
Layers++ [37] | 30.3 | 0.20 19 | 0.44 21 | 0.11 16 | 0.41 8 | 0.98 11 | 0.24 19 | 0.53 8 | 0.96 9 | 0.25 18 | 0.75 59 | 1.82 55 | 0.43 65 | 1.12 5 | 1.37 5 | 0.88 11 | 1.39 39 | 2.27 22 | 1.09 61 | 0.16 57 | 0.13 66 | 0.18 43 | 1.25 30 | 1.87 28 | 1.80 52 |
NL-TV-NCC [25] | 31.5 | 0.20 19 | 0.44 21 | 0.09 3 | 0.47 22 | 1.07 23 | 0.19 7 | 1.16 63 | 1.95 72 | 0.31 31 | 0.53 11 | 1.42 11 | 0.24 8 | 1.37 34 | 1.61 31 | 1.14 34 | 2.07 80 | 2.39 29 | 1.47 78 | 0.16 57 | 0.09 3 | 0.19 49 | 1.24 27 | 1.93 32 | 1.29 12 |
LME [70] | 32.8 | 0.21 33 | 0.48 69 | 0.11 16 | 0.40 5 | 0.99 14 | 0.17 1 | 0.89 34 | 1.50 35 | 0.68 93 | 0.68 36 | 1.63 30 | 0.50 85 | 1.33 28 | 1.56 21 | 1.29 46 | 1.17 20 | 2.77 54 | 0.83 23 | 0.13 35 | 0.11 29 | 0.17 39 | 1.07 12 | 1.68 14 | 1.39 16 |
FESL [72] | 33.4 | 0.20 19 | 0.46 42 | 0.10 4 | 0.54 49 | 1.16 47 | 0.31 42 | 0.82 31 | 1.43 30 | 0.27 21 | 0.64 23 | 1.60 27 | 0.34 14 | 1.23 15 | 1.49 15 | 0.96 18 | 1.38 38 | 2.51 39 | 0.92 35 | 0.16 57 | 0.13 66 | 0.22 62 | 1.33 43 | 1.98 38 | 1.53 27 |
FC-2Layers-FF [74] | 33.4 | 0.20 19 | 0.45 34 | 0.11 16 | 0.51 38 | 1.18 52 | 0.28 26 | 0.53 8 | 0.94 8 | 0.24 16 | 0.79 71 | 2.00 79 | 0.43 65 | 1.18 10 | 1.43 10 | 0.89 12 | 1.44 44 | 2.63 45 | 0.99 45 | 0.16 57 | 0.12 45 | 0.19 49 | 1.09 17 | 1.69 15 | 1.45 21 |
nLayers [57] | 34.2 | 0.20 19 | 0.46 42 | 0.11 16 | 0.43 11 | 0.98 11 | 0.25 21 | 0.67 16 | 1.18 16 | 0.29 27 | 0.90 87 | 2.34 109 | 0.52 88 | 1.25 17 | 1.53 19 | 0.93 15 | 1.33 32 | 2.21 18 | 0.94 39 | 0.14 44 | 0.12 45 | 0.18 43 | 1.22 24 | 1.96 35 | 1.51 26 |
AGIF+OF [85] | 34.2 | 0.21 33 | 0.46 42 | 0.11 16 | 0.54 49 | 1.16 47 | 0.29 30 | 0.98 46 | 1.65 45 | 0.34 44 | 0.63 18 | 1.57 18 | 0.35 17 | 1.19 13 | 1.45 12 | 0.91 14 | 1.41 41 | 2.25 21 | 1.01 52 | 0.15 51 | 0.10 14 | 0.19 49 | 1.41 47 | 2.04 41 | 1.85 60 |
PH-Flow [101] | 34.3 | 0.21 33 | 0.46 42 | 0.12 33 | 0.53 45 | 1.14 36 | 0.31 42 | 0.65 14 | 1.16 15 | 0.26 20 | 0.76 61 | 1.95 71 | 0.39 45 | 1.16 9 | 1.42 8 | 0.90 13 | 0.88 6 | 1.95 10 | 0.67 6 | 0.20 91 | 0.13 66 | 0.30 92 | 1.10 18 | 1.64 11 | 1.60 36 |
ProbFlowFields [128] | 35.2 | 0.23 66 | 0.48 69 | 0.15 75 | 0.48 28 | 1.12 30 | 0.32 48 | 0.74 24 | 1.32 24 | 0.32 36 | 0.42 4 | 1.25 7 | 0.22 6 | 1.60 72 | 1.91 92 | 1.33 53 | 1.04 14 | 2.47 36 | 0.70 9 | 0.09 6 | 0.08 1 | 0.11 7 | 1.46 50 | 2.43 63 | 1.46 24 |
LSM [39] | 35.8 | 0.20 19 | 0.45 34 | 0.12 33 | 0.54 49 | 1.16 47 | 0.32 48 | 0.91 37 | 1.53 36 | 0.34 44 | 0.62 16 | 1.55 17 | 0.36 23 | 1.31 26 | 1.57 23 | 1.07 28 | 1.42 42 | 2.41 30 | 0.99 45 | 0.17 72 | 0.11 29 | 0.22 62 | 1.25 30 | 1.83 24 | 1.72 45 |
Sparse-NonSparse [56] | 37.5 | 0.21 33 | 0.46 42 | 0.12 33 | 0.54 49 | 1.17 51 | 0.32 48 | 0.93 39 | 1.56 38 | 0.34 44 | 0.64 23 | 1.60 27 | 0.36 23 | 1.33 28 | 1.59 29 | 1.07 28 | 1.43 43 | 2.48 37 | 0.99 45 | 0.17 72 | 0.10 14 | 0.22 62 | 1.24 27 | 1.82 22 | 1.70 43 |
EPPM w/o HM [88] | 38.0 | 0.22 56 | 0.44 21 | 0.11 16 | 0.46 19 | 1.07 23 | 0.19 7 | 0.81 29 | 1.43 30 | 0.20 11 | 0.66 30 | 1.72 40 | 0.36 23 | 1.18 10 | 1.45 12 | 0.64 8 | 2.02 78 | 2.80 57 | 1.44 76 | 0.28 113 | 0.13 66 | 0.37 105 | 1.19 22 | 1.80 20 | 1.66 40 |
Classic+CPF [83] | 38.3 | 0.21 33 | 0.46 42 | 0.12 33 | 0.54 49 | 1.16 47 | 0.28 26 | 1.01 51 | 1.69 48 | 0.33 40 | 0.63 18 | 1.59 22 | 0.35 17 | 1.30 22 | 1.57 23 | 1.05 24 | 1.45 45 | 2.37 26 | 0.97 40 | 0.19 83 | 0.11 29 | 0.28 87 | 1.27 32 | 1.88 30 | 1.80 52 |
Ramp [62] | 39.0 | 0.21 33 | 0.46 42 | 0.12 33 | 0.50 35 | 1.10 27 | 0.31 42 | 0.87 33 | 1.49 34 | 0.34 44 | 0.63 18 | 1.57 18 | 0.35 17 | 1.30 22 | 1.57 23 | 1.09 31 | 1.52 56 | 2.75 51 | 1.02 54 | 0.20 91 | 0.12 45 | 0.33 98 | 1.22 24 | 1.81 21 | 1.72 45 |
TC/T-Flow [76] | 39.4 | 0.17 2 | 0.38 2 | 0.11 16 | 0.60 71 | 1.24 69 | 0.33 52 | 1.00 50 | 1.72 52 | 0.20 11 | 0.75 59 | 2.02 83 | 0.36 23 | 1.44 42 | 1.66 37 | 1.38 58 | 0.94 9 | 2.11 13 | 0.81 21 | 0.15 51 | 0.13 66 | 0.24 74 | 1.27 32 | 1.89 31 | 1.45 21 |
IROF-TV [53] | 39.5 | 0.21 33 | 0.46 42 | 0.13 52 | 0.51 38 | 1.13 33 | 0.33 52 | 0.98 46 | 1.60 41 | 0.30 30 | 0.71 44 | 1.77 50 | 0.40 53 | 1.33 28 | 1.60 30 | 1.10 33 | 1.51 52 | 3.56 94 | 0.99 45 | 0.08 4 | 0.10 14 | 0.10 5 | 1.31 38 | 2.07 42 | 1.74 48 |
PMF [73] | 39.5 | 0.21 33 | 0.47 54 | 0.11 16 | 0.45 14 | 1.03 16 | 0.20 11 | 0.48 3 | 0.85 3 | 0.14 3 | 0.80 74 | 2.12 94 | 0.38 42 | 1.12 5 | 1.38 7 | 0.54 3 | 3.68 94 | 2.46 34 | 2.72 95 | 0.21 96 | 0.21 117 | 0.28 87 | 1.04 10 | 1.55 6 | 1.55 31 |
CostFilter [40] | 40.0 | 0.21 33 | 0.47 54 | 0.11 16 | 0.47 22 | 1.05 19 | 0.22 14 | 0.43 1 | 0.75 1 | 0.18 9 | 0.76 61 | 1.95 71 | 0.38 42 | 1.12 5 | 1.37 5 | 0.56 5 | 3.23 89 | 2.45 32 | 2.50 92 | 0.22 99 | 0.20 116 | 0.31 95 | 1.20 23 | 1.85 26 | 1.55 31 |
HAST [109] | 40.1 | 0.21 33 | 0.47 54 | 0.10 4 | 0.49 31 | 1.04 18 | 0.33 52 | 0.44 2 | 0.80 2 | 0.11 1 | 0.89 86 | 2.41 114 | 0.37 33 | 1.08 4 | 1.34 4 | 0.54 3 | 1.99 75 | 2.75 51 | 1.58 80 | 0.26 107 | 0.16 98 | 0.37 105 | 0.72 1 | 1.29 1 | 0.55 3 |
Classic+NL [31] | 40.7 | 0.21 33 | 0.47 54 | 0.12 33 | 0.53 45 | 1.15 41 | 0.32 48 | 0.96 45 | 1.62 42 | 0.35 48 | 0.65 27 | 1.63 30 | 0.36 23 | 1.27 19 | 1.52 17 | 1.03 23 | 1.53 57 | 2.77 54 | 0.98 43 | 0.17 72 | 0.12 45 | 0.23 67 | 1.27 32 | 1.87 28 | 1.78 51 |
Efficient-NL [60] | 40.8 | 0.21 33 | 0.47 54 | 0.11 16 | 0.47 22 | 1.06 22 | 0.27 23 | 1.07 55 | 1.79 59 | 0.31 31 | 0.66 30 | 1.66 33 | 0.37 33 | 1.25 17 | 1.52 17 | 0.94 16 | 4.42 99 | 3.42 84 | 2.99 99 | 0.18 78 | 0.12 45 | 0.26 82 | 1.07 12 | 1.60 9 | 1.19 10 |
Sparse Occlusion [54] | 41.0 | 0.20 19 | 0.44 21 | 0.13 52 | 0.45 14 | 1.03 16 | 0.26 22 | 1.26 68 | 2.12 85 | 0.35 48 | 0.71 44 | 1.80 53 | 0.37 33 | 1.41 37 | 1.69 40 | 1.05 24 | 0.90 7 | 2.03 12 | 0.67 6 | 0.19 83 | 0.22 121 | 0.23 67 | 1.31 38 | 2.07 42 | 1.55 31 |
MDP-Flow [26] | 41.4 | 0.18 4 | 0.38 2 | 0.12 33 | 0.39 4 | 0.90 2 | 0.29 30 | 0.62 13 | 1.11 13 | 0.33 40 | 0.63 18 | 1.73 43 | 0.31 13 | 1.63 79 | 1.72 45 | 1.99 110 | 1.73 68 | 1.86 7 | 1.40 75 | 0.13 35 | 0.13 66 | 0.15 27 | 2.01 86 | 2.98 89 | 2.17 92 |
SRR-TVOF-NL [91] | 42.2 | 0.21 33 | 0.43 14 | 0.14 64 | 0.70 89 | 1.41 89 | 0.46 84 | 1.01 51 | 1.69 48 | 0.39 61 | 0.64 23 | 1.50 12 | 0.34 14 | 1.42 40 | 1.62 34 | 1.55 75 | 0.95 11 | 2.23 20 | 0.70 9 | 0.16 57 | 0.15 87 | 0.21 59 | 1.07 12 | 1.60 9 | 1.40 17 |
ROF-ND [107] | 42.9 | 0.23 66 | 0.46 42 | 0.11 16 | 0.56 61 | 1.13 33 | 0.31 42 | 1.15 61 | 1.90 68 | 0.33 40 | 0.27 1 | 0.76 1 | 0.14 4 | 1.49 48 | 1.76 53 | 1.39 59 | 1.09 15 | 2.21 18 | 0.83 23 | 0.19 83 | 0.13 66 | 0.23 67 | 1.58 58 | 2.17 50 | 1.82 55 |
TV-L1-MCT [64] | 43.0 | 0.21 33 | 0.48 69 | 0.12 33 | 0.56 61 | 1.19 56 | 0.30 34 | 1.05 54 | 1.79 59 | 0.35 48 | 0.62 16 | 1.54 15 | 0.35 17 | 1.35 33 | 1.61 31 | 1.14 34 | 1.46 49 | 2.50 38 | 1.00 51 | 0.15 51 | 0.10 14 | 0.36 103 | 1.33 43 | 1.94 34 | 1.83 56 |
PGM-C [120] | 45.0 | 0.24 82 | 0.47 54 | 0.17 93 | 0.55 56 | 1.22 62 | 0.36 60 | 0.78 27 | 1.35 28 | 0.36 54 | 0.77 66 | 1.83 57 | 0.47 78 | 1.52 56 | 1.80 58 | 1.31 51 | 1.17 20 | 2.64 46 | 0.76 16 | 0.09 6 | 0.09 3 | 0.12 10 | 1.27 32 | 1.97 36 | 1.54 29 |
COFM [59] | 45.6 | 0.23 66 | 0.52 88 | 0.13 52 | 0.52 42 | 1.14 36 | 0.30 34 | 0.89 34 | 1.56 38 | 0.31 31 | 0.84 83 | 2.18 99 | 0.42 63 | 1.50 51 | 1.75 52 | 1.51 69 | 0.86 5 | 1.73 3 | 0.75 15 | 0.19 83 | 0.11 29 | 0.25 78 | 1.07 12 | 1.65 13 | 1.44 19 |
FMOF [94] | 46.8 | 0.20 19 | 0.45 34 | 0.12 33 | 0.58 66 | 1.25 71 | 0.30 34 | 0.72 22 | 1.26 23 | 0.23 15 | 0.67 34 | 1.66 33 | 0.37 33 | 1.30 22 | 1.57 23 | 1.00 22 | 4.83 108 | 2.73 49 | 3.46 110 | 0.19 83 | 0.11 29 | 0.35 101 | 1.59 60 | 2.34 58 | 1.69 42 |
RNLOD-Flow [121] | 47.8 | 0.19 8 | 0.43 14 | 0.10 4 | 0.51 38 | 1.15 41 | 0.29 30 | 1.17 64 | 1.96 74 | 0.35 48 | 0.71 44 | 1.88 62 | 0.37 33 | 1.65 82 | 2.00 115 | 1.36 56 | 1.24 24 | 2.58 44 | 0.87 28 | 0.18 78 | 0.15 87 | 0.26 82 | 1.22 24 | 1.86 27 | 1.67 41 |
MLDP_OF [89] | 48.0 | 0.20 19 | 0.43 14 | 0.11 16 | 0.41 8 | 0.97 8 | 0.17 1 | 1.14 60 | 1.87 64 | 0.31 31 | 0.73 52 | 1.97 74 | 0.36 23 | 1.38 35 | 1.66 37 | 1.16 36 | 1.24 24 | 3.11 68 | 0.76 16 | 0.26 107 | 0.13 66 | 0.45 111 | 2.97 122 | 2.32 55 | 2.42 105 |
DPOF [18] | 48.4 | 0.23 66 | 0.47 54 | 0.14 64 | 0.60 71 | 1.21 60 | 0.41 77 | 0.52 6 | 0.92 5 | 0.19 10 | 0.71 44 | 1.73 43 | 0.44 70 | 1.18 10 | 1.42 8 | 0.73 10 | 3.53 93 | 2.98 65 | 2.25 89 | 0.38 118 | 0.13 66 | 0.47 113 | 0.95 6 | 1.58 8 | 0.78 5 |
SVFilterOh [111] | 48.7 | 0.22 56 | 0.49 80 | 0.12 33 | 0.51 38 | 1.18 52 | 0.23 17 | 0.68 17 | 1.24 20 | 0.24 16 | 0.76 61 | 2.02 83 | 0.37 33 | 1.33 28 | 1.65 36 | 0.66 9 | 3.09 88 | 3.20 72 | 2.02 86 | 0.27 111 | 0.16 98 | 0.35 101 | 1.01 8 | 1.71 16 | 1.17 9 |
Kuang [131] | 49.0 | 0.23 66 | 0.46 42 | 0.16 87 | 0.52 42 | 1.11 29 | 0.35 59 | 0.80 28 | 1.36 29 | 0.35 48 | 0.70 43 | 1.74 45 | 0.41 58 | 1.46 45 | 1.71 44 | 1.20 40 | 3.26 90 | 2.73 49 | 2.48 91 | 0.10 14 | 0.09 3 | 0.13 16 | 1.58 58 | 2.29 53 | 2.24 98 |
CPM-Flow [116] | 50.1 | 0.24 82 | 0.48 69 | 0.17 93 | 0.55 56 | 1.22 62 | 0.37 63 | 0.77 26 | 1.33 26 | 0.36 54 | 0.77 66 | 1.84 58 | 0.47 78 | 1.51 54 | 1.78 56 | 1.25 41 | 1.36 36 | 2.45 32 | 1.02 54 | 0.09 6 | 0.09 3 | 0.12 10 | 1.53 56 | 2.34 58 | 1.91 63 |
2DHMM-SAS [92] | 50.1 | 0.21 33 | 0.46 42 | 0.12 33 | 0.57 64 | 1.23 64 | 0.30 34 | 1.42 80 | 2.07 81 | 0.55 79 | 0.65 27 | 1.59 22 | 0.36 23 | 1.29 20 | 1.55 20 | 1.06 26 | 1.51 52 | 2.46 34 | 1.02 54 | 0.19 83 | 0.12 45 | 0.30 92 | 1.64 62 | 2.41 62 | 1.96 71 |
ALD-Flow [66] | 50.3 | 0.18 4 | 0.39 5 | 0.10 4 | 0.63 77 | 1.30 73 | 0.36 60 | 0.98 46 | 1.71 51 | 0.28 25 | 0.82 78 | 2.00 79 | 0.39 45 | 1.59 66 | 1.84 72 | 1.57 79 | 1.63 64 | 3.34 79 | 0.93 36 | 0.14 44 | 0.12 45 | 0.22 62 | 1.29 36 | 1.98 38 | 1.65 39 |
ACK-Prior [27] | 50.4 | 0.18 4 | 0.39 5 | 0.10 4 | 0.46 19 | 1.05 19 | 0.19 7 | 0.82 31 | 1.46 32 | 0.25 18 | 0.59 13 | 1.65 32 | 0.22 6 | 1.50 51 | 1.74 50 | 1.42 62 | 6.47 126 | 4.94 123 | 4.41 126 | 0.28 113 | 0.17 103 | 0.37 105 | 1.67 63 | 2.35 60 | 1.62 37 |
TCOF [69] | 50.9 | 0.21 33 | 0.44 21 | 0.13 52 | 0.52 42 | 1.15 41 | 0.30 34 | 1.59 95 | 2.33 109 | 0.58 82 | 0.68 36 | 1.72 40 | 0.39 45 | 1.58 64 | 1.82 67 | 1.55 75 | 1.24 24 | 2.33 24 | 0.86 26 | 0.22 99 | 0.12 45 | 0.39 108 | 1.13 20 | 1.75 19 | 1.45 21 |
OAR-Flow [125] | 52.2 | 0.20 19 | 0.43 14 | 0.14 64 | 0.85 102 | 1.47 96 | 0.61 101 | 1.15 61 | 1.89 66 | 0.40 62 | 0.80 74 | 1.97 74 | 0.39 45 | 1.62 76 | 1.87 79 | 1.53 73 | 1.51 52 | 3.29 78 | 0.83 23 | 0.09 6 | 0.10 14 | 0.12 10 | 1.17 21 | 1.83 24 | 1.43 18 |
EpicFlow [102] | 52.2 | 0.23 66 | 0.47 54 | 0.17 93 | 0.55 56 | 1.23 64 | 0.37 63 | 1.07 55 | 1.81 62 | 0.40 62 | 0.71 44 | 1.69 36 | 0.44 70 | 1.54 57 | 1.81 64 | 1.34 54 | 1.70 66 | 2.64 46 | 1.18 68 | 0.09 6 | 0.09 3 | 0.12 10 | 1.46 50 | 2.12 46 | 1.84 58 |
TC-Flow [46] | 53.0 | 0.17 2 | 0.38 2 | 0.10 4 | 0.50 35 | 1.15 41 | 0.27 23 | 1.09 57 | 1.88 65 | 0.29 27 | 0.77 66 | 2.01 81 | 0.39 45 | 1.54 57 | 1.80 58 | 1.43 64 | 2.11 81 | 3.56 94 | 1.17 67 | 0.15 51 | 0.12 45 | 0.25 78 | 1.69 64 | 2.50 67 | 2.24 98 |
Complementary OF [21] | 53.6 | 0.19 8 | 0.41 8 | 0.12 33 | 0.49 31 | 1.14 36 | 0.22 14 | 0.95 41 | 1.67 47 | 0.27 21 | 0.76 61 | 2.03 85 | 0.38 42 | 1.73 97 | 1.91 92 | 1.91 103 | 6.38 125 | 4.41 118 | 4.34 125 | 0.10 14 | 0.09 3 | 0.16 33 | 1.42 48 | 2.15 48 | 1.81 54 |
OFH [38] | 54.2 | 0.19 8 | 0.41 8 | 0.12 33 | 0.56 61 | 1.23 64 | 0.34 56 | 1.39 78 | 2.10 83 | 0.37 57 | 0.82 78 | 2.24 101 | 0.40 53 | 1.60 72 | 1.84 72 | 1.61 82 | 1.85 73 | 3.78 100 | 1.46 77 | 0.09 6 | 0.09 3 | 0.11 7 | 1.35 45 | 2.15 48 | 1.58 35 |
ComplOF-FED-GPU [35] | 55.3 | 0.19 8 | 0.41 8 | 0.12 33 | 0.59 68 | 1.24 69 | 0.36 60 | 0.95 41 | 1.66 46 | 0.28 25 | 0.79 71 | 2.03 85 | 0.40 53 | 1.57 62 | 1.80 58 | 1.58 81 | 4.00 97 | 3.34 79 | 2.67 94 | 0.14 44 | 0.11 29 | 0.23 67 | 1.46 50 | 2.27 52 | 1.74 48 |
SimpleFlow [49] | 55.8 | 0.21 33 | 0.47 54 | 0.12 33 | 0.55 56 | 1.20 57 | 0.34 56 | 1.45 81 | 2.19 90 | 0.41 66 | 0.66 30 | 1.68 35 | 0.35 17 | 1.41 37 | 1.69 40 | 1.16 36 | 5.06 112 | 4.23 114 | 3.40 107 | 0.16 57 | 0.12 45 | 0.23 67 | 1.29 36 | 1.97 36 | 1.70 43 |
Aniso-Texture [82] | 55.8 | 0.18 4 | 0.40 7 | 0.11 16 | 0.46 19 | 1.13 33 | 0.28 26 | 1.65 103 | 2.39 117 | 0.60 84 | 0.52 10 | 1.33 10 | 0.30 12 | 1.74 99 | 1.95 105 | 1.89 101 | 1.32 31 | 2.55 42 | 1.15 65 | 0.17 72 | 0.17 103 | 0.18 43 | 1.91 74 | 2.66 71 | 2.17 92 |
AggregFlow [97] | 57.9 | 0.29 101 | 0.63 114 | 0.16 87 | 0.86 103 | 1.58 103 | 0.60 99 | 0.93 39 | 1.62 42 | 0.42 68 | 0.74 54 | 1.91 67 | 0.37 33 | 1.51 54 | 1.80 58 | 1.29 46 | 1.09 15 | 2.36 25 | 0.65 3 | 0.13 35 | 0.13 66 | 0.21 59 | 1.32 41 | 2.02 40 | 1.62 37 |
HBM-GC [105] | 58.2 | 0.24 82 | 0.54 97 | 0.12 33 | 0.48 28 | 1.12 30 | 0.33 52 | 0.99 49 | 1.74 53 | 0.32 36 | 0.73 52 | 1.84 58 | 0.44 70 | 1.40 36 | 1.68 39 | 1.30 49 | 2.49 85 | 2.29 23 | 2.03 88 | 0.21 96 | 0.13 66 | 0.25 78 | 1.59 60 | 2.45 64 | 1.97 72 |
FlowNet2 [122] | 58.4 | 0.35 112 | 0.68 117 | 0.19 102 | 0.89 106 | 1.58 103 | 0.60 99 | 0.95 41 | 1.48 33 | 0.63 88 | 0.71 44 | 1.84 58 | 0.41 58 | 1.34 32 | 1.58 27 | 0.98 19 | 1.82 71 | 2.37 26 | 1.58 80 | 0.12 26 | 0.16 98 | 0.13 16 | 1.10 18 | 1.82 22 | 0.89 6 |
S2D-Matching [84] | 59.8 | 0.24 82 | 0.53 91 | 0.14 64 | 0.63 77 | 1.31 77 | 0.34 56 | 1.26 68 | 2.06 79 | 0.40 62 | 0.67 34 | 1.69 36 | 0.35 17 | 1.43 41 | 1.72 45 | 1.16 36 | 1.58 59 | 2.84 59 | 1.02 54 | 0.21 96 | 0.12 45 | 0.32 96 | 1.40 46 | 2.08 44 | 1.97 72 |
Occlusion-TV-L1 [63] | 61.3 | 0.22 56 | 0.47 54 | 0.13 52 | 0.53 45 | 1.18 52 | 0.38 68 | 1.58 93 | 2.36 116 | 0.54 78 | 0.74 54 | 1.76 48 | 0.42 63 | 1.55 61 | 1.81 64 | 1.35 55 | 1.60 60 | 2.98 65 | 1.22 69 | 0.09 6 | 0.11 29 | 0.09 3 | 2.10 91 | 3.06 99 | 2.16 91 |
RFlow [90] | 62.5 | 0.20 19 | 0.43 14 | 0.12 33 | 0.48 28 | 1.12 30 | 0.30 34 | 1.45 81 | 2.21 93 | 0.36 54 | 0.95 93 | 2.47 118 | 0.50 85 | 1.69 88 | 1.92 97 | 1.81 94 | 1.27 28 | 2.82 58 | 0.88 30 | 0.15 51 | 0.12 45 | 0.24 74 | 1.99 84 | 2.96 86 | 2.11 84 |
DeepFlow2 [108] | 63.1 | 0.21 33 | 0.45 34 | 0.14 64 | 0.71 90 | 1.32 79 | 0.52 91 | 1.17 64 | 1.92 69 | 0.41 66 | 0.83 82 | 1.99 77 | 0.47 78 | 1.54 57 | 1.80 58 | 1.44 66 | 1.45 45 | 3.38 83 | 0.91 34 | 0.12 26 | 0.11 29 | 0.18 43 | 2.00 85 | 2.93 83 | 2.04 79 |
SegOF [10] | 64.2 | 0.24 82 | 0.48 69 | 0.18 99 | 0.71 90 | 1.35 86 | 0.58 96 | 1.31 71 | 1.92 69 | 0.73 95 | 0.54 12 | 1.09 3 | 0.41 58 | 1.62 76 | 1.79 57 | 1.64 84 | 5.27 114 | 4.17 113 | 3.63 113 | 0.07 1 | 0.10 14 | 0.10 5 | 1.49 53 | 2.48 66 | 1.35 15 |
Aniso. Huber-L1 [22] | 65.0 | 0.23 66 | 0.48 69 | 0.15 75 | 0.60 71 | 1.26 72 | 0.39 70 | 1.66 104 | 2.29 100 | 0.50 74 | 0.69 40 | 1.59 22 | 0.39 45 | 1.59 66 | 1.82 67 | 1.55 75 | 1.21 23 | 2.86 60 | 0.72 12 | 0.19 83 | 0.14 83 | 0.29 91 | 1.72 66 | 2.64 69 | 1.84 58 |
CBF [12] | 65.5 | 0.19 8 | 0.41 8 | 0.12 33 | 0.55 56 | 1.14 36 | 0.45 81 | 1.36 76 | 2.04 77 | 0.47 71 | 0.81 77 | 2.01 81 | 0.44 70 | 1.65 82 | 1.91 92 | 1.68 85 | 1.13 17 | 2.54 41 | 0.76 16 | 0.29 116 | 0.17 103 | 0.42 109 | 1.91 74 | 2.85 79 | 2.10 83 |
Adaptive [20] | 68.0 | 0.23 66 | 0.50 83 | 0.14 64 | 0.57 64 | 1.30 73 | 0.39 70 | 1.72 110 | 2.49 124 | 0.56 81 | 0.74 54 | 1.71 39 | 0.43 65 | 1.49 48 | 1.76 53 | 1.30 49 | 1.45 45 | 2.53 40 | 0.99 45 | 0.17 72 | 0.17 103 | 0.20 56 | 1.91 74 | 2.83 77 | 2.00 76 |
Steered-L1 [118] | 68.3 | 0.16 1 | 0.34 1 | 0.11 16 | 0.40 5 | 0.98 11 | 0.27 23 | 0.92 38 | 1.62 42 | 0.31 31 | 0.78 69 | 2.05 90 | 0.40 53 | 1.62 76 | 1.86 77 | 1.77 89 | 6.13 124 | 3.96 106 | 4.33 124 | 0.40 119 | 0.17 103 | 0.79 121 | 2.44 107 | 3.04 96 | 2.98 117 |
TF+OM [100] | 69.2 | 0.21 33 | 0.45 34 | 0.13 52 | 0.54 49 | 1.18 52 | 0.39 70 | 0.95 41 | 1.56 38 | 0.76 98 | 0.71 44 | 1.76 48 | 0.45 75 | 1.85 119 | 2.02 120 | 1.99 110 | 2.23 82 | 3.65 97 | 1.29 72 | 0.16 57 | 0.15 87 | 0.19 49 | 1.86 71 | 2.74 74 | 2.13 89 |
CRTflow [80] | 70.0 | 0.21 33 | 0.42 13 | 0.14 64 | 0.67 83 | 1.40 87 | 0.37 63 | 1.59 95 | 2.32 108 | 0.55 79 | 0.97 94 | 2.37 112 | 0.53 90 | 1.63 79 | 1.91 92 | 1.54 74 | 1.60 60 | 3.74 99 | 0.89 31 | 0.11 20 | 0.11 29 | 0.16 33 | 1.93 78 | 2.97 87 | 2.01 77 |
SIOF [67] | 70.0 | 0.24 82 | 0.52 88 | 0.13 52 | 0.68 85 | 1.40 87 | 0.48 87 | 1.50 90 | 2.15 88 | 0.83 99 | 0.82 78 | 1.94 70 | 0.49 84 | 1.66 86 | 1.88 85 | 1.80 93 | 1.35 35 | 2.87 61 | 0.97 40 | 0.12 26 | 0.11 29 | 0.15 27 | 1.72 66 | 2.55 68 | 1.99 75 |
LocallyOriented [52] | 70.9 | 0.29 101 | 0.60 106 | 0.15 75 | 0.77 95 | 1.42 91 | 0.57 95 | 1.63 101 | 2.29 100 | 0.49 73 | 0.66 30 | 1.57 18 | 0.41 58 | 1.59 66 | 1.80 58 | 1.55 75 | 3.32 91 | 3.50 89 | 2.32 90 | 0.11 20 | 0.11 29 | 0.16 33 | 1.72 66 | 2.45 64 | 2.03 78 |
Brox et al. [5] | 72.5 | 0.22 56 | 0.45 34 | 0.15 75 | 0.68 85 | 1.58 103 | 0.40 73 | 1.11 58 | 1.89 66 | 0.40 62 | 0.93 91 | 2.16 97 | 0.51 87 | 1.78 108 | 1.95 105 | 2.00 113 | 2.88 87 | 3.52 92 | 2.02 86 | 0.08 4 | 0.11 29 | 0.09 3 | 1.97 82 | 2.81 76 | 1.94 67 |
DeepFlow [86] | 73.4 | 0.23 66 | 0.48 69 | 0.15 75 | 0.78 97 | 1.34 83 | 0.62 103 | 1.23 67 | 1.95 72 | 0.75 97 | 0.99 100 | 2.47 118 | 0.54 93 | 1.54 57 | 1.82 67 | 1.37 57 | 1.51 52 | 3.51 91 | 0.93 36 | 0.12 26 | 0.10 14 | 0.18 43 | 2.18 95 | 3.03 95 | 2.13 89 |
SuperFlow [81] | 74.0 | 0.25 91 | 0.48 69 | 0.15 75 | 0.63 77 | 1.21 60 | 0.50 88 | 1.30 70 | 1.97 75 | 0.88 104 | 0.80 74 | 1.88 62 | 0.46 77 | 1.69 88 | 1.92 97 | 1.79 90 | 1.61 62 | 3.20 72 | 1.09 61 | 0.13 35 | 0.15 87 | 0.14 24 | 2.09 90 | 2.93 83 | 1.93 66 |
p-harmonic [29] | 75.3 | 0.23 66 | 0.48 69 | 0.15 75 | 0.54 49 | 1.20 57 | 0.40 73 | 1.62 99 | 2.29 100 | 0.61 86 | 0.78 69 | 1.75 46 | 0.53 90 | 1.75 101 | 1.92 97 | 1.99 110 | 1.45 45 | 3.28 76 | 0.99 45 | 0.16 57 | 0.15 87 | 0.16 33 | 2.17 94 | 3.06 99 | 2.12 85 |
Classic++ [32] | 75.8 | 0.22 56 | 0.48 69 | 0.15 75 | 0.60 71 | 1.34 83 | 0.38 68 | 1.35 75 | 2.10 83 | 0.44 69 | 0.84 83 | 2.08 92 | 0.44 70 | 1.59 66 | 1.88 85 | 1.39 59 | 1.47 51 | 3.17 70 | 1.04 59 | 0.20 91 | 0.14 83 | 0.27 85 | 2.11 92 | 3.01 93 | 2.17 92 |
CLG-TV [48] | 76.2 | 0.23 66 | 0.49 80 | 0.14 64 | 0.59 68 | 1.30 73 | 0.37 63 | 1.67 106 | 2.39 117 | 0.48 72 | 0.74 54 | 1.70 38 | 0.41 58 | 1.64 81 | 1.90 90 | 1.52 71 | 1.55 58 | 3.54 93 | 0.93 36 | 0.23 102 | 0.19 114 | 0.36 103 | 1.89 72 | 2.88 81 | 1.95 70 |
DF-Auto [115] | 76.9 | 0.30 105 | 0.57 103 | 0.18 99 | 0.88 105 | 1.46 95 | 0.67 106 | 1.33 73 | 2.07 81 | 0.84 102 | 0.82 78 | 1.95 71 | 0.45 75 | 1.72 96 | 1.95 105 | 1.79 90 | 1.24 24 | 2.77 54 | 0.69 8 | 0.13 35 | 0.17 103 | 0.12 10 | 1.93 78 | 2.89 82 | 1.94 67 |
TriFlow [95] | 77.3 | 0.25 91 | 0.53 91 | 0.15 75 | 0.66 82 | 1.48 97 | 0.45 81 | 1.36 76 | 2.04 77 | 0.89 105 | 0.68 36 | 1.77 50 | 0.39 45 | 1.82 115 | 1.98 111 | 1.93 106 | 1.33 32 | 2.96 64 | 0.87 28 | 0.65 126 | 0.21 117 | 0.69 119 | 1.49 53 | 2.20 51 | 1.53 27 |
TriangleFlow [30] | 80.0 | 0.25 91 | 0.55 100 | 0.13 52 | 0.69 87 | 1.53 102 | 0.40 73 | 1.47 86 | 2.19 90 | 0.38 58 | 0.74 54 | 1.87 61 | 0.43 65 | 1.81 112 | 2.00 115 | 1.98 108 | 2.41 84 | 2.89 62 | 1.74 83 | 0.19 83 | 0.18 112 | 0.24 74 | 1.52 55 | 2.29 53 | 1.86 61 |
Filter Flow [19] | 81.1 | 0.28 99 | 0.54 97 | 0.19 102 | 0.69 87 | 1.31 77 | 0.51 90 | 1.45 81 | 1.98 76 | 1.02 112 | 1.13 108 | 1.82 55 | 1.04 111 | 1.71 92 | 1.83 70 | 2.04 115 | 1.28 29 | 2.37 26 | 1.01 52 | 0.16 57 | 0.16 98 | 0.17 39 | 2.29 99 | 2.98 89 | 2.12 85 |
Local-TV-L1 [65] | 81.3 | 0.28 99 | 0.53 91 | 0.17 93 | 1.04 110 | 1.49 99 | 0.91 118 | 1.80 113 | 2.24 97 | 1.00 111 | 0.92 90 | 2.16 97 | 0.54 93 | 1.94 123 | 1.87 79 | 1.41 61 | 1.18 22 | 2.67 48 | 0.81 21 | 0.12 26 | 0.10 14 | 0.14 24 | 2.34 102 | 3.52 120 | 2.36 101 |
Fusion [6] | 81.4 | 0.23 66 | 0.48 69 | 0.16 87 | 0.50 35 | 1.23 64 | 0.30 34 | 0.81 29 | 1.32 24 | 0.38 58 | 0.72 51 | 1.92 68 | 0.47 78 | 1.83 117 | 2.04 122 | 2.00 113 | 5.51 116 | 3.26 74 | 3.93 117 | 0.20 91 | 0.19 114 | 0.26 82 | 2.56 111 | 3.51 119 | 2.80 114 |
UnFlow [129] | 81.6 | 0.48 124 | 0.77 127 | 0.33 122 | 0.78 97 | 1.33 80 | 0.63 105 | 1.40 79 | 1.85 63 | 0.73 95 | 1.03 102 | 1.90 65 | 0.85 107 | 1.75 101 | 1.87 79 | 1.92 105 | 2.31 83 | 3.59 96 | 1.62 82 | 0.07 1 | 0.10 14 | 0.06 1 | 1.90 73 | 3.09 101 | 1.83 56 |
Second-order prior [8] | 82.0 | 0.22 56 | 0.47 54 | 0.14 64 | 0.65 81 | 1.33 80 | 0.46 84 | 1.64 102 | 2.30 105 | 0.52 77 | 0.79 71 | 1.89 64 | 0.48 83 | 1.66 86 | 1.89 89 | 1.70 86 | 1.78 70 | 3.44 86 | 1.47 78 | 0.23 102 | 0.14 83 | 0.32 96 | 2.04 87 | 2.83 77 | 2.44 106 |
StereoFlow [44] | 82.2 | 0.52 125 | 0.77 127 | 0.38 127 | 1.21 122 | 1.73 121 | 0.91 118 | 1.49 88 | 1.92 69 | 0.99 110 | 1.41 119 | 2.88 128 | 1.03 110 | 1.59 66 | 1.84 72 | 1.52 71 | 1.36 36 | 3.16 69 | 0.78 19 | 0.07 1 | 0.09 3 | 0.08 2 | 2.06 88 | 3.00 91 | 2.17 92 |
F-TV-L1 [15] | 82.5 | 0.26 94 | 0.54 97 | 0.15 75 | 0.86 103 | 1.50 100 | 0.61 101 | 1.73 111 | 2.31 106 | 0.65 90 | 1.00 101 | 2.46 116 | 0.52 88 | 1.57 62 | 1.84 72 | 1.42 62 | 1.66 65 | 3.42 84 | 1.15 65 | 0.14 44 | 0.18 112 | 0.13 16 | 1.95 80 | 2.97 87 | 1.76 50 |
TV-L1-improved [17] | 82.5 | 0.22 56 | 0.47 54 | 0.14 64 | 0.53 45 | 1.20 57 | 0.37 63 | 1.71 109 | 2.48 123 | 0.60 84 | 0.98 97 | 2.45 115 | 0.54 93 | 1.61 74 | 1.87 79 | 1.50 68 | 4.77 106 | 3.84 102 | 3.21 101 | 0.18 78 | 0.17 103 | 0.22 62 | 1.96 81 | 2.94 85 | 2.08 82 |
GraphCuts [14] | 82.7 | 0.24 82 | 0.47 54 | 0.17 93 | 1.17 120 | 1.72 119 | 0.90 117 | 1.12 59 | 1.70 50 | 0.83 99 | 0.69 40 | 1.60 27 | 0.39 45 | 1.46 45 | 1.69 40 | 1.31 51 | 6.11 123 | 3.97 107 | 4.27 122 | 0.24 104 | 0.13 66 | 0.34 100 | 2.36 104 | 3.16 107 | 2.54 110 |
Shiralkar [42] | 82.8 | 0.23 66 | 0.45 34 | 0.13 52 | 0.72 93 | 1.43 92 | 0.46 84 | 1.70 107 | 2.33 109 | 0.61 86 | 0.91 89 | 2.03 85 | 0.57 98 | 1.61 74 | 1.83 70 | 1.57 79 | 1.89 74 | 3.02 67 | 1.26 70 | 0.24 104 | 0.13 66 | 0.33 98 | 2.28 97 | 3.05 97 | 2.20 96 |
BriefMatch [124] | 82.9 | 0.19 8 | 0.41 8 | 0.12 33 | 0.58 66 | 1.14 36 | 0.44 80 | 0.70 19 | 1.22 18 | 0.29 27 | 0.90 87 | 2.24 101 | 0.47 78 | 1.87 121 | 2.09 123 | 2.21 125 | 5.67 119 | 3.28 76 | 4.16 121 | 0.52 123 | 0.25 124 | 1.06 128 | 3.09 125 | 3.30 115 | 4.34 129 |
Rannacher [23] | 84.9 | 0.23 66 | 0.50 83 | 0.15 75 | 0.59 68 | 1.33 80 | 0.41 77 | 1.76 112 | 2.51 127 | 0.64 89 | 0.97 94 | 2.38 113 | 0.53 90 | 1.59 66 | 1.86 77 | 1.43 64 | 4.79 107 | 3.91 104 | 3.22 102 | 0.17 72 | 0.13 66 | 0.24 74 | 1.83 70 | 2.86 80 | 2.06 81 |
Bartels [41] | 85.3 | 0.23 66 | 0.50 83 | 0.13 52 | 0.49 31 | 1.15 41 | 0.29 30 | 1.01 51 | 1.78 57 | 0.38 58 | 1.07 106 | 2.60 124 | 0.63 100 | 1.71 92 | 1.97 110 | 1.82 95 | 4.54 102 | 3.85 103 | 3.40 107 | 0.28 113 | 0.14 83 | 0.51 117 | 2.36 104 | 3.22 110 | 2.66 112 |
Dynamic MRF [7] | 88.8 | 0.24 82 | 0.52 88 | 0.14 64 | 0.62 76 | 1.43 92 | 0.40 73 | 1.33 73 | 2.16 89 | 0.45 70 | 0.94 92 | 2.23 100 | 0.58 99 | 1.81 112 | 2.01 118 | 1.91 103 | 4.83 108 | 4.05 111 | 3.55 112 | 0.14 44 | 0.09 3 | 0.23 67 | 2.75 116 | 3.57 123 | 2.97 116 |
IAOF2 [51] | 89.8 | 0.26 94 | 0.53 91 | 0.20 105 | 0.80 99 | 1.59 108 | 0.56 94 | 1.58 93 | 2.33 109 | 0.86 103 | 1.18 112 | 1.81 54 | 1.13 115 | 1.65 82 | 1.90 90 | 1.62 83 | 1.83 72 | 2.75 51 | 1.34 74 | 0.24 104 | 0.15 87 | 0.47 113 | 1.92 77 | 2.66 71 | 1.98 74 |
CNN-flow-warp+ref [117] | 90.0 | 0.26 94 | 0.53 91 | 0.18 99 | 0.60 71 | 1.23 64 | 0.50 88 | 1.48 87 | 2.21 93 | 0.67 91 | 0.98 97 | 2.29 107 | 0.54 93 | 1.77 107 | 1.96 108 | 1.93 106 | 2.58 86 | 3.93 105 | 1.75 84 | 0.11 20 | 0.13 66 | 0.19 49 | 2.88 120 | 3.40 117 | 3.07 118 |
Ad-TV-NDC [36] | 90.1 | 0.38 116 | 0.62 112 | 0.32 119 | 1.53 126 | 1.77 124 | 1.42 126 | 2.25 128 | 2.56 129 | 1.17 121 | 0.97 94 | 1.75 46 | 0.85 107 | 1.58 64 | 1.88 85 | 1.28 43 | 1.39 39 | 3.26 74 | 0.86 26 | 0.14 44 | 0.13 66 | 0.15 27 | 2.80 119 | 3.12 105 | 3.41 123 |
StereoOF-V1MT [119] | 90.2 | 0.26 94 | 0.53 91 | 0.13 52 | 0.93 107 | 1.74 122 | 0.53 92 | 1.60 97 | 2.25 98 | 0.50 74 | 0.85 85 | 1.90 65 | 0.55 97 | 1.76 104 | 1.92 97 | 1.84 99 | 4.73 104 | 4.08 112 | 3.26 103 | 0.11 20 | 0.11 29 | 0.20 56 | 3.05 124 | 3.57 123 | 3.12 119 |
FlowNetS+ft+v [112] | 90.4 | 0.24 82 | 0.50 83 | 0.16 87 | 0.77 95 | 1.48 97 | 0.58 96 | 1.66 104 | 2.31 106 | 0.83 99 | 0.98 97 | 2.15 95 | 0.65 101 | 1.70 91 | 1.92 97 | 1.74 87 | 1.62 63 | 3.49 88 | 1.08 60 | 0.26 107 | 0.25 124 | 0.47 113 | 1.70 65 | 2.64 69 | 1.91 63 |
HCIC-L [99] | 91.4 | 0.45 120 | 0.67 115 | 0.32 119 | 1.95 129 | 1.93 127 | 2.05 129 | 1.32 72 | 1.76 56 | 0.93 106 | 1.65 122 | 2.25 103 | 1.65 124 | 1.41 37 | 1.64 35 | 1.07 28 | 2.04 79 | 2.90 63 | 1.81 85 | 0.94 129 | 0.42 129 | 1.62 131 | 1.72 66 | 2.39 61 | 1.54 29 |
2D-CLG [1] | 93.8 | 0.47 122 | 0.75 126 | 0.32 119 | 0.74 94 | 1.30 73 | 0.62 103 | 1.89 118 | 2.29 100 | 1.15 120 | 1.26 115 | 1.97 74 | 1.16 116 | 1.80 111 | 1.92 97 | 2.04 115 | 4.75 105 | 3.83 101 | 3.27 104 | 0.10 14 | 0.08 1 | 0.14 24 | 2.34 102 | 3.00 91 | 2.44 106 |
IAOF [50] | 94.7 | 0.27 98 | 0.51 87 | 0.20 105 | 1.10 115 | 1.70 117 | 0.82 110 | 2.57 130 | 2.92 131 | 1.22 124 | 1.15 109 | 1.93 69 | 1.09 113 | 1.65 82 | 1.88 85 | 1.74 87 | 1.77 69 | 3.35 81 | 1.02 54 | 0.18 78 | 0.12 45 | 0.28 87 | 2.44 107 | 2.78 75 | 2.86 115 |
Nguyen [33] | 95.8 | 0.33 109 | 0.59 105 | 0.21 108 | 1.00 109 | 1.60 110 | 0.81 109 | 2.10 123 | 2.46 121 | 1.13 118 | 1.32 116 | 2.07 91 | 1.22 117 | 1.73 97 | 1.92 97 | 1.89 101 | 2.00 76 | 3.97 107 | 1.33 73 | 0.13 35 | 0.12 45 | 0.17 39 | 2.31 100 | 3.02 94 | 2.34 100 |
GroupFlow [9] | 96.2 | 0.33 109 | 0.60 106 | 0.22 113 | 1.08 112 | 1.71 118 | 0.85 111 | 1.61 98 | 2.13 86 | 0.94 107 | 0.65 27 | 1.54 15 | 0.43 65 | 2.01 125 | 2.22 126 | 1.45 67 | 5.60 117 | 4.02 109 | 3.86 115 | 0.26 107 | 0.16 98 | 0.46 112 | 2.08 89 | 2.72 73 | 2.41 104 |
Learning Flow [11] | 96.3 | 0.23 66 | 0.49 80 | 0.13 52 | 0.64 80 | 1.44 94 | 0.41 77 | 1.46 84 | 2.20 92 | 0.50 74 | 1.15 109 | 2.26 105 | 0.86 109 | 2.05 126 | 2.32 127 | 2.04 115 | 5.24 113 | 5.18 127 | 3.43 109 | 0.16 57 | 0.17 103 | 0.28 87 | 2.47 109 | 3.29 114 | 2.39 103 |
Heeger++ [104] | 99.5 | 0.37 115 | 0.61 108 | 0.26 115 | 1.16 118 | 1.68 115 | 0.88 115 | 1.53 91 | 2.06 79 | 0.68 93 | 1.50 121 | 2.25 103 | 1.31 120 | 1.76 104 | 1.73 48 | 2.06 118 | 4.14 98 | 3.70 98 | 2.93 98 | 0.16 57 | 0.12 45 | 0.21 59 | 3.19 126 | 3.52 120 | 3.50 125 |
Modified CLG [34] | 99.7 | 0.31 106 | 0.57 103 | 0.21 108 | 0.67 83 | 1.34 83 | 0.53 92 | 1.89 118 | 2.34 114 | 1.11 117 | 1.06 104 | 2.28 106 | 0.75 105 | 1.82 115 | 1.99 113 | 2.06 118 | 3.91 95 | 4.32 116 | 2.86 97 | 0.16 57 | 0.13 66 | 0.27 85 | 2.14 93 | 3.09 101 | 2.21 97 |
Horn & Schunck [3] | 100.2 | 0.36 113 | 0.62 112 | 0.23 114 | 0.98 108 | 1.58 103 | 0.77 108 | 1.88 117 | 2.26 99 | 1.13 118 | 1.36 118 | 2.15 95 | 1.22 117 | 1.69 88 | 1.84 72 | 1.83 97 | 3.92 96 | 4.34 117 | 2.78 96 | 0.16 57 | 0.15 87 | 0.23 67 | 2.40 106 | 3.05 97 | 2.38 102 |
SPSA-learn [13] | 100.8 | 0.33 109 | 0.61 108 | 0.21 108 | 1.09 114 | 1.83 125 | 0.85 111 | 1.82 116 | 2.33 109 | 1.05 113 | 1.23 113 | 2.31 108 | 1.06 112 | 1.79 110 | 1.96 108 | 1.98 108 | 5.61 118 | 4.52 120 | 3.84 114 | 0.12 26 | 0.10 14 | 0.15 27 | 2.48 110 | 3.21 109 | 2.45 108 |
TI-DOFE [24] | 101.6 | 0.46 121 | 0.69 119 | 0.35 125 | 1.16 118 | 1.59 108 | 1.03 121 | 2.17 126 | 2.39 117 | 1.27 125 | 1.43 120 | 2.04 88 | 1.34 121 | 1.71 92 | 1.87 79 | 1.83 97 | 3.36 92 | 4.03 110 | 2.58 93 | 0.13 35 | 0.12 45 | 0.20 56 | 2.67 113 | 3.12 105 | 2.76 113 |
LDOF [28] | 102.2 | 0.29 101 | 0.61 108 | 0.17 93 | 0.81 100 | 1.58 103 | 0.59 98 | 1.17 64 | 1.78 57 | 0.67 91 | 2.17 126 | 5.16 131 | 1.57 122 | 1.78 108 | 2.01 118 | 1.86 100 | 4.52 101 | 4.42 119 | 3.33 105 | 0.22 99 | 0.22 121 | 0.47 113 | 2.21 96 | 3.27 113 | 1.92 65 |
Black & Anandan [4] | 102.5 | 0.31 106 | 0.55 100 | 0.20 105 | 1.08 112 | 1.68 115 | 0.85 111 | 2.00 121 | 2.41 120 | 1.07 115 | 1.03 102 | 1.99 77 | 0.83 106 | 1.71 92 | 1.91 92 | 1.82 95 | 4.64 103 | 5.05 124 | 3.01 100 | 0.20 91 | 0.17 103 | 0.30 92 | 2.28 97 | 3.09 101 | 2.04 79 |
HBpMotionGpu [43] | 103.5 | 0.36 113 | 0.67 115 | 0.21 108 | 1.07 111 | 1.85 126 | 0.88 115 | 2.00 121 | 2.49 124 | 1.08 116 | 1.16 111 | 2.81 127 | 0.68 102 | 1.89 122 | 2.10 124 | 2.08 120 | 2.01 77 | 3.19 71 | 1.27 71 | 0.16 57 | 0.15 87 | 0.19 49 | 2.32 101 | 3.18 108 | 2.48 109 |
2bit-BM-tele [98] | 107.2 | 0.31 106 | 0.61 108 | 0.21 108 | 0.71 90 | 1.62 112 | 0.45 81 | 1.55 92 | 2.33 109 | 0.58 82 | 1.09 107 | 2.57 123 | 0.72 104 | 1.76 104 | 2.03 121 | 1.79 90 | 5.05 111 | 3.45 87 | 3.88 116 | 0.43 121 | 0.22 121 | 0.69 119 | 2.73 115 | 3.58 125 | 3.20 120 |
BlockOverlap [61] | 109.6 | 0.29 101 | 0.56 102 | 0.19 102 | 0.82 101 | 1.41 89 | 0.69 107 | 1.70 107 | 2.22 95 | 0.94 107 | 1.06 104 | 2.54 122 | 0.70 103 | 1.81 112 | 1.99 113 | 2.18 122 | 5.46 115 | 3.35 81 | 3.96 118 | 0.67 127 | 0.32 127 | 1.47 130 | 2.67 113 | 3.22 110 | 3.31 122 |
FFV1MT [106] | 110.4 | 0.41 118 | 0.72 121 | 0.27 117 | 1.11 116 | 1.76 123 | 0.87 114 | 1.62 99 | 2.13 86 | 0.94 107 | 2.02 125 | 2.68 126 | 1.87 126 | 2.00 124 | 1.87 79 | 2.60 127 | 4.48 100 | 4.68 121 | 3.35 106 | 0.18 78 | 0.15 87 | 0.25 78 | 3.19 126 | 3.52 120 | 3.50 125 |
Adaptive flow [45] | 112.7 | 0.44 119 | 0.69 119 | 0.26 115 | 1.43 123 | 1.72 119 | 1.31 125 | 1.91 120 | 2.22 95 | 1.33 126 | 1.35 117 | 2.10 93 | 1.28 119 | 1.74 99 | 2.00 115 | 1.51 69 | 4.85 110 | 3.50 89 | 3.49 111 | 0.70 128 | 0.36 128 | 0.97 127 | 2.62 112 | 3.33 116 | 2.57 111 |
PGAM+LK [55] | 115.2 | 0.66 130 | 0.72 121 | 1.02 131 | 1.64 127 | 2.08 129 | 1.67 127 | 1.46 84 | 1.79 59 | 1.05 113 | 2.59 131 | 3.23 129 | 2.39 131 | 1.75 101 | 1.76 53 | 2.09 121 | 6.62 127 | 4.24 115 | 5.02 127 | 0.40 119 | 0.21 117 | 0.57 118 | 2.95 121 | 3.22 110 | 3.48 124 |
SLK [47] | 116.0 | 0.47 122 | 0.73 123 | 0.44 129 | 1.13 117 | 1.51 101 | 1.02 120 | 2.12 124 | 2.34 114 | 1.19 122 | 1.72 123 | 2.35 110 | 1.60 123 | 1.86 120 | 1.92 97 | 2.20 124 | 5.75 120 | 4.91 122 | 4.02 119 | 0.32 117 | 0.12 45 | 0.44 110 | 3.61 128 | 4.02 128 | 3.71 127 |
SILK [79] | 116.2 | 0.39 117 | 0.68 117 | 0.30 118 | 1.19 121 | 1.60 110 | 1.10 122 | 2.19 127 | 2.47 122 | 1.19 122 | 1.25 114 | 2.36 111 | 1.10 114 | 1.84 118 | 1.98 111 | 2.18 122 | 5.84 121 | 5.17 126 | 4.03 120 | 0.45 122 | 0.12 45 | 0.96 126 | 3.03 123 | 3.47 118 | 3.24 121 |
AdaConv-v1 [126] | 120.2 | 0.54 126 | 0.74 124 | 0.34 123 | 1.43 123 | 1.67 113 | 1.23 123 | 1.80 113 | 1.75 54 | 1.42 128 | 2.28 128 | 2.51 120 | 2.32 129 | 3.78 128 | 3.95 128 | 3.85 129 | 7.12 129 | 5.61 129 | 5.15 128 | 1.19 130 | 0.66 130 | 0.88 122 | 2.75 116 | 3.69 126 | 2.12 85 |
SepConv-v1 [127] | 120.2 | 0.54 126 | 0.74 124 | 0.34 123 | 1.43 123 | 1.67 113 | 1.23 123 | 1.80 113 | 1.75 54 | 1.42 128 | 2.28 128 | 2.51 120 | 2.32 129 | 3.78 128 | 3.95 128 | 3.85 129 | 7.12 129 | 5.61 129 | 5.15 128 | 1.19 130 | 0.66 130 | 0.88 122 | 2.75 116 | 3.69 126 | 2.12 85 |
FOLKI [16] | 126.2 | 0.55 128 | 0.95 130 | 0.41 128 | 2.45 130 | 2.14 130 | 2.44 130 | 2.12 124 | 2.50 126 | 1.33 126 | 1.75 124 | 2.61 125 | 1.67 125 | 2.10 127 | 2.15 125 | 2.52 126 | 5.86 122 | 5.06 125 | 4.32 123 | 0.52 123 | 0.21 117 | 0.95 125 | 5.03 131 | 4.59 129 | 8.32 131 |
Periodicity [78] | 127.0 | 0.59 129 | 1.02 131 | 0.35 125 | 2.90 131 | 3.18 131 | 3.12 131 | 2.61 131 | 2.53 128 | 2.05 131 | 2.19 127 | 3.28 130 | 2.08 127 | 7.58 131 | 8.15 131 | 6.70 131 | 7.02 128 | 6.03 131 | 5.54 131 | 0.27 111 | 0.15 87 | 0.92 124 | 4.68 130 | 6.02 131 | 4.37 130 |
Pyramid LK [2] | 128.4 | 0.68 131 | 0.87 129 | 0.77 130 | 1.94 128 | 1.95 128 | 1.92 128 | 2.43 129 | 2.59 130 | 1.60 130 | 2.36 130 | 2.46 116 | 2.27 128 | 4.19 130 | 4.59 130 | 3.80 128 | 7.35 131 | 5.52 128 | 5.19 130 | 0.57 125 | 0.26 126 | 1.19 129 | 4.63 129 | 5.72 130 | 4.02 128 |
Method | time* | frames | color | 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] Kuang | 9.9 | 2 | gray | F. Kuang. PatchMatch algorithms for motion estimation and stereo reconstruction. Master thesis, University of Stuttgart, 2017. |