Optical flow evaluation results |
Statistics:
Average
SD
R0.5
R1.0
R2.0
A50
A75
A95
Error type: endpoint angle interpolation normalized interpolation |
A95 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] | 9.3 | 0.19 5 | 0.97 10 | 0.10 2 | 0.47 14 | 2.00 3 | 0.36 19 | 0.45 4 | 1.58 2 | 0.31 13 | 0.23 10 | 1.93 4 | 0.15 14 | 1.80 2 | 2.62 1 | 0.67 1 | 0.92 3 | 3.05 2 | 0.52 7 | 0.34 25 | 0.38 55 | 0.40 13 | 1.32 8 | 3.74 5 | 0.78 2 |
MDP-Flow2 [68] | 13.7 | 0.20 7 | 0.91 4 | 0.14 15 | 0.37 2 | 2.11 4 | 0.28 4 | 0.40 3 | 1.89 8 | 0.27 4 | 0.21 7 | 4.76 46 | 0.13 4 | 3.73 24 | 4.67 21 | 2.57 28 | 1.07 5 | 4.13 7 | 0.80 16 | 0.36 31 | 0.34 25 | 0.49 35 | 1.27 7 | 4.18 12 | 2.27 10 |
PMMST [114] | 14.6 | 0.21 11 | 0.78 1 | 0.15 20 | 0.56 26 | 1.85 1 | 0.47 40 | 0.61 14 | 1.73 5 | 0.44 40 | 0.21 7 | 1.58 2 | 0.18 24 | 2.48 3 | 3.84 3 | 1.13 9 | 1.10 7 | 2.57 1 | 0.69 10 | 0.37 39 | 0.37 48 | 0.48 30 | 1.13 3 | 3.60 2 | 1.15 5 |
OFLAF [77] | 14.8 | 0.18 2 | 0.97 10 | 0.12 7 | 0.44 9 | 2.29 6 | 0.35 18 | 0.35 1 | 1.67 4 | 0.27 4 | 0.18 4 | 7.02 101 | 0.13 4 | 3.14 13 | 4.38 13 | 0.79 5 | 1.49 22 | 3.88 5 | 0.82 17 | 0.35 26 | 0.30 8 | 0.47 25 | 1.21 6 | 4.04 10 | 4.87 36 |
NN-field [71] | 15.4 | 0.21 11 | 1.05 24 | 0.10 2 | 0.55 23 | 1.97 2 | 0.41 28 | 0.45 4 | 1.81 6 | 0.31 13 | 0.20 6 | 2.01 5 | 0.13 4 | 1.78 1 | 2.62 1 | 0.67 1 | 5.40 94 | 3.52 3 | 0.33 1 | 0.39 44 | 0.37 48 | 0.48 30 | 1.33 9 | 4.03 9 | 0.65 1 |
NNF-EAC [103] | 22.8 | 0.22 18 | 0.98 13 | 0.15 20 | 0.44 9 | 2.49 11 | 0.32 12 | 0.51 10 | 2.43 16 | 0.29 9 | 0.30 33 | 4.28 35 | 0.17 20 | 3.44 15 | 4.48 14 | 2.00 22 | 1.76 29 | 4.14 8 | 1.09 26 | 0.40 50 | 0.37 48 | 0.49 35 | 1.82 20 | 5.75 48 | 4.09 26 |
ComponentFusion [96] | 24.4 | 0.20 7 | 1.12 53 | 0.14 15 | 0.42 7 | 2.56 14 | 0.36 19 | 0.50 9 | 2.09 12 | 0.30 11 | 0.25 18 | 4.26 34 | 0.15 14 | 4.34 45 | 5.53 74 | 3.20 44 | 1.39 19 | 4.77 24 | 1.21 28 | 0.36 31 | 0.34 25 | 0.49 35 | 1.46 10 | 4.63 18 | 3.43 20 |
Correlation Flow [75] | 27.8 | 0.24 26 | 0.96 8 | 0.13 11 | 0.44 9 | 2.72 19 | 0.28 4 | 1.13 51 | 7.59 103 | 0.26 3 | 0.25 18 | 2.18 6 | 0.18 24 | 4.34 45 | 5.12 41 | 3.88 55 | 1.60 23 | 4.99 43 | 0.78 14 | 0.43 61 | 0.35 32 | 0.61 56 | 1.14 4 | 3.93 6 | 1.00 4 |
WLIF-Flow [93] | 27.8 | 0.20 7 | 0.99 15 | 0.14 15 | 0.58 33 | 2.33 7 | 0.48 43 | 0.60 13 | 3.98 29 | 0.36 22 | 0.24 13 | 3.87 27 | 0.16 18 | 3.48 17 | 4.59 17 | 1.94 21 | 1.98 36 | 4.68 20 | 1.21 28 | 0.41 52 | 0.35 32 | 0.67 70 | 1.98 30 | 5.84 50 | 5.79 53 |
TC/T-Flow [76] | 29.4 | 0.22 18 | 0.89 3 | 0.09 1 | 0.50 17 | 3.52 57 | 0.29 7 | 0.57 11 | 4.47 39 | 0.28 6 | 0.25 18 | 6.68 87 | 0.13 4 | 4.01 38 | 5.07 38 | 2.83 33 | 0.72 1 | 4.54 13 | 0.53 8 | 0.44 65 | 0.38 55 | 0.80 95 | 1.79 19 | 5.27 33 | 5.05 39 |
Layers++ [37] | 30.3 | 0.22 18 | 1.09 43 | 0.19 41 | 0.57 30 | 2.79 22 | 0.48 43 | 0.49 8 | 2.07 10 | 0.40 31 | 0.16 2 | 3.00 8 | 0.12 3 | 2.69 4 | 3.92 4 | 1.41 13 | 2.67 58 | 4.82 34 | 2.11 67 | 0.48 73 | 0.39 62 | 0.56 45 | 1.48 11 | 4.57 17 | 6.44 81 |
LME [70] | 31.5 | 0.19 5 | 1.18 67 | 0.14 15 | 0.40 5 | 2.36 8 | 0.31 10 | 1.12 50 | 4.78 43 | 2.16 98 | 0.28 27 | 4.22 33 | 0.18 24 | 3.83 27 | 4.64 19 | 3.19 43 | 1.35 16 | 5.14 49 | 1.20 27 | 0.39 44 | 0.36 42 | 0.51 38 | 1.74 18 | 4.69 19 | 4.21 28 |
HAST [109] | 31.5 | 0.16 1 | 1.11 48 | 0.10 2 | 0.60 35 | 2.45 10 | 0.38 24 | 0.37 2 | 1.66 3 | 0.23 1 | 0.16 2 | 7.63 114 | 0.11 1 | 2.69 4 | 4.02 5 | 0.75 3 | 2.28 44 | 5.16 51 | 2.11 67 | 0.76 116 | 0.50 105 | 1.22 114 | 0.80 1 | 2.97 1 | 0.88 3 |
FC-2Layers-FF [74] | 31.7 | 0.21 11 | 1.06 27 | 0.15 20 | 0.62 37 | 2.91 25 | 0.49 46 | 0.47 6 | 2.01 9 | 0.41 32 | 0.24 13 | 5.40 61 | 0.17 20 | 2.71 6 | 4.34 12 | 1.18 10 | 3.69 70 | 4.62 15 | 2.29 78 | 0.52 87 | 0.38 55 | 0.66 65 | 1.94 26 | 3.95 7 | 3.89 22 |
IROF++ [58] | 31.8 | 0.23 22 | 1.11 48 | 0.15 20 | 0.69 54 | 3.07 30 | 0.55 64 | 0.71 25 | 3.60 23 | 0.49 56 | 0.31 40 | 3.72 23 | 0.21 47 | 3.45 16 | 4.49 15 | 1.91 20 | 2.27 43 | 4.87 36 | 1.85 52 | 0.28 9 | 0.35 32 | 0.35 7 | 1.84 23 | 4.90 25 | 4.73 34 |
RNLOD-Flow [121] | 35.8 | 0.18 2 | 0.94 7 | 0.13 11 | 0.56 26 | 3.22 40 | 0.39 25 | 0.79 37 | 5.57 61 | 0.32 16 | 0.19 5 | 5.51 62 | 0.14 11 | 4.13 39 | 5.10 40 | 3.04 39 | 2.15 38 | 4.65 18 | 1.95 57 | 0.52 87 | 0.45 88 | 0.66 65 | 1.67 15 | 4.70 21 | 5.61 49 |
PH-Flow [101] | 36.1 | 0.25 34 | 1.08 39 | 0.19 41 | 0.67 51 | 2.99 28 | 0.54 58 | 0.59 12 | 2.14 13 | 0.47 53 | 0.32 42 | 6.83 92 | 0.21 47 | 2.89 8 | 4.26 8 | 1.23 11 | 1.69 24 | 3.71 4 | 1.30 30 | 0.50 80 | 0.40 70 | 0.67 70 | 1.64 14 | 4.06 11 | 4.13 27 |
Efficient-NL [60] | 36.2 | 0.21 11 | 1.07 34 | 0.13 11 | 0.68 52 | 2.73 20 | 0.54 58 | 0.76 33 | 5.11 52 | 0.42 34 | 0.25 18 | 4.36 36 | 0.17 20 | 3.55 18 | 4.65 20 | 1.75 17 | 13.6 112 | 4.78 26 | 2.28 75 | 0.43 61 | 0.38 55 | 0.64 59 | 1.85 24 | 3.96 8 | 2.86 15 |
PMF [73] | 36.5 | 0.24 26 | 1.20 72 | 0.12 7 | 0.55 23 | 2.53 12 | 0.36 19 | 0.73 31 | 2.07 10 | 0.33 18 | 0.23 10 | 7.68 116 | 0.16 18 | 2.88 7 | 4.18 7 | 0.81 6 | 2.19 40 | 4.86 35 | 1.77 49 | 0.57 101 | 0.69 125 | 0.88 101 | 1.16 5 | 3.64 3 | 4.75 35 |
nLayers [57] | 36.6 | 0.18 2 | 1.13 57 | 0.11 5 | 0.71 57 | 2.43 9 | 0.60 72 | 0.79 37 | 3.30 20 | 0.59 63 | 0.14 1 | 7.94 123 | 0.11 1 | 3.02 10 | 4.52 16 | 1.30 12 | 2.95 61 | 4.35 10 | 2.38 88 | 0.41 52 | 0.40 70 | 0.45 23 | 1.73 17 | 5.23 31 | 5.32 41 |
FESL [72] | 37.9 | 0.21 11 | 0.99 15 | 0.12 7 | 0.85 78 | 3.07 30 | 0.65 85 | 0.71 25 | 4.19 31 | 0.46 46 | 0.24 13 | 3.77 24 | 0.18 24 | 3.59 20 | 4.69 22 | 1.76 18 | 3.92 78 | 4.81 32 | 2.23 73 | 0.43 61 | 0.43 84 | 0.63 58 | 1.97 27 | 4.79 22 | 3.97 25 |
TC-Flow [46] | 38.5 | 0.21 11 | 0.92 5 | 0.13 11 | 0.37 2 | 3.53 58 | 0.25 1 | 0.69 24 | 5.91 64 | 0.25 2 | 0.25 18 | 6.61 86 | 0.13 4 | 4.42 51 | 5.37 55 | 3.69 49 | 1.20 13 | 5.45 57 | 0.50 5 | 0.39 44 | 0.37 48 | 0.88 101 | 2.91 56 | 6.70 68 | 6.54 92 |
ALD-Flow [66] | 38.8 | 0.20 7 | 0.96 8 | 0.11 5 | 0.43 8 | 3.49 55 | 0.32 12 | 0.76 33 | 5.29 55 | 0.29 9 | 0.21 7 | 6.51 84 | 0.13 4 | 4.62 66 | 5.54 78 | 4.19 73 | 1.14 10 | 5.03 46 | 0.47 4 | 0.42 57 | 0.37 48 | 0.80 95 | 2.48 49 | 6.03 52 | 6.25 67 |
AGIF+OF [85] | 39.1 | 0.22 18 | 1.08 39 | 0.16 26 | 0.73 61 | 3.41 47 | 0.60 72 | 0.68 21 | 4.45 37 | 0.42 34 | 0.29 28 | 3.90 28 | 0.21 47 | 3.23 14 | 4.32 11 | 1.63 14 | 3.43 66 | 4.63 17 | 2.00 61 | 0.42 57 | 0.35 32 | 0.67 70 | 2.05 34 | 5.63 44 | 5.98 61 |
MLDP_OF [89] | 40.5 | 0.29 54 | 1.08 39 | 0.20 57 | 0.46 13 | 2.23 5 | 0.33 14 | 0.74 32 | 5.53 60 | 0.33 18 | 0.26 24 | 6.85 93 | 0.19 32 | 3.95 31 | 4.95 33 | 2.16 23 | 1.08 6 | 4.99 43 | 0.74 12 | 0.66 111 | 0.40 70 | 1.14 111 | 2.33 45 | 5.26 32 | 2.66 13 |
LSM [39] | 40.7 | 0.24 26 | 1.02 22 | 0.19 41 | 0.65 44 | 3.08 33 | 0.52 53 | 0.72 27 | 3.97 28 | 0.43 37 | 0.33 47 | 3.60 18 | 0.20 37 | 3.70 22 | 4.69 22 | 2.45 26 | 3.76 72 | 4.77 24 | 2.29 78 | 0.51 86 | 0.34 25 | 0.66 65 | 2.17 40 | 5.39 38 | 6.15 65 |
Classic+CPF [83] | 40.7 | 0.24 26 | 1.07 34 | 0.18 37 | 0.68 52 | 3.41 47 | 0.54 58 | 0.72 27 | 4.90 44 | 0.41 32 | 0.29 28 | 4.02 30 | 0.20 37 | 3.82 26 | 4.85 29 | 2.18 24 | 4.18 81 | 4.72 22 | 2.08 66 | 0.49 75 | 0.35 32 | 0.66 65 | 1.82 20 | 5.09 28 | 5.91 57 |
EPPM w/o HM [88] | 41.2 | 0.39 82 | 1.06 27 | 0.21 67 | 0.56 26 | 2.53 12 | 0.34 17 | 0.83 42 | 4.71 41 | 0.38 28 | 0.43 62 | 4.42 37 | 0.22 56 | 3.08 12 | 4.30 10 | 1.11 8 | 2.35 51 | 4.79 29 | 1.58 43 | 0.75 112 | 0.38 55 | 1.07 106 | 1.49 12 | 4.37 13 | 5.32 41 |
CombBMOF [113] | 41.2 | 0.31 56 | 1.06 27 | 0.16 26 | 0.62 37 | 2.64 18 | 0.46 39 | 0.67 19 | 2.39 14 | 0.46 46 | 0.52 74 | 3.34 12 | 0.33 78 | 3.71 23 | 5.03 37 | 1.77 19 | 3.65 69 | 4.51 12 | 3.27 105 | 0.47 71 | 0.49 102 | 0.52 41 | 1.70 16 | 5.02 27 | 3.45 21 |
CostFilter [40] | 41.8 | 0.28 47 | 1.18 67 | 0.19 41 | 0.54 20 | 2.61 17 | 0.36 19 | 0.72 27 | 1.86 7 | 0.38 28 | 0.29 28 | 6.22 77 | 0.20 37 | 3.00 9 | 4.16 6 | 0.87 7 | 2.17 39 | 4.96 41 | 1.39 35 | 0.63 107 | 0.63 123 | 1.07 106 | 1.86 25 | 5.47 40 | 5.74 51 |
Ramp [62] | 42.1 | 0.25 34 | 1.07 34 | 0.19 41 | 0.66 49 | 3.12 35 | 0.53 56 | 0.67 19 | 3.70 24 | 0.44 40 | 0.32 42 | 3.63 20 | 0.20 37 | 3.95 31 | 4.83 28 | 2.60 31 | 3.80 75 | 4.78 26 | 2.29 78 | 0.50 80 | 0.38 55 | 0.69 81 | 2.10 36 | 4.69 19 | 5.27 40 |
Sparse-NonSparse [56] | 42.2 | 0.24 26 | 1.06 27 | 0.20 57 | 0.65 44 | 3.13 36 | 0.54 58 | 0.68 21 | 4.08 30 | 0.43 37 | 0.34 50 | 3.65 21 | 0.21 47 | 3.81 25 | 4.81 27 | 2.51 27 | 3.72 71 | 4.72 22 | 2.29 78 | 0.50 80 | 0.33 21 | 0.65 64 | 2.04 32 | 5.64 45 | 6.16 66 |
SVFilterOh [111] | 42.3 | 0.25 34 | 1.35 97 | 0.17 33 | 0.56 26 | 2.77 21 | 0.41 28 | 0.48 7 | 2.49 17 | 0.36 22 | 0.25 18 | 7.54 113 | 0.19 32 | 3.69 21 | 5.13 42 | 0.75 3 | 3.76 72 | 5.14 49 | 1.48 38 | 0.75 112 | 0.47 96 | 1.15 113 | 0.96 2 | 4.37 13 | 1.26 6 |
IROF-TV [53] | 42.5 | 0.25 34 | 1.23 76 | 0.19 41 | 0.71 57 | 3.17 39 | 0.55 64 | 0.79 37 | 4.71 41 | 0.45 43 | 0.39 60 | 4.60 41 | 0.23 61 | 3.97 34 | 4.88 30 | 2.57 28 | 1.89 33 | 6.80 90 | 1.56 42 | 0.27 4 | 0.32 14 | 0.34 6 | 2.08 35 | 5.76 49 | 5.97 60 |
FMOF [94] | 42.5 | 0.23 22 | 0.99 15 | 0.18 37 | 0.75 65 | 3.31 42 | 0.60 72 | 0.62 15 | 3.02 18 | 0.43 37 | 0.29 28 | 3.77 24 | 0.20 37 | 3.96 33 | 4.78 25 | 1.69 16 | 4.72 87 | 4.81 32 | 2.14 70 | 0.49 75 | 0.35 32 | 0.71 87 | 2.72 54 | 5.97 51 | 5.50 45 |
NL-TV-NCC [25] | 42.6 | 0.28 47 | 1.12 53 | 0.16 26 | 0.63 40 | 3.28 41 | 0.39 25 | 0.78 36 | 6.29 68 | 0.30 11 | 0.30 33 | 3.60 18 | 0.21 47 | 3.97 34 | 4.88 30 | 3.23 45 | 5.50 95 | 5.64 61 | 1.75 48 | 0.47 71 | 0.34 25 | 0.70 82 | 2.25 42 | 5.36 36 | 2.04 9 |
ProbFlowFields [128] | 42.8 | 0.36 72 | 1.14 59 | 0.25 91 | 0.59 34 | 3.46 51 | 0.53 56 | 0.80 40 | 3.81 27 | 0.62 65 | 0.32 42 | 1.74 3 | 0.21 47 | 4.78 79 | 5.78 115 | 3.78 52 | 0.82 2 | 4.95 40 | 0.50 5 | 0.28 9 | 0.28 2 | 0.38 11 | 2.60 52 | 6.16 55 | 3.28 17 |
FlowFields+ [130] | 42.9 | 0.48 96 | 1.17 65 | 0.24 86 | 0.78 71 | 2.58 16 | 0.62 78 | 1.16 52 | 3.17 19 | 0.85 77 | 0.49 70 | 2.34 7 | 0.34 80 | 4.25 44 | 5.29 50 | 2.97 36 | 1.12 8 | 4.78 26 | 0.86 19 | 0.30 12 | 0.30 8 | 0.40 13 | 2.26 43 | 5.53 41 | 2.56 12 |
COFM [59] | 43.0 | 0.23 22 | 1.28 87 | 0.16 26 | 0.55 23 | 3.02 29 | 0.40 27 | 1.27 57 | 4.99 48 | 0.48 55 | 0.23 10 | 7.47 109 | 0.14 11 | 4.36 48 | 5.28 49 | 4.20 74 | 1.75 27 | 4.98 42 | 1.37 34 | 0.49 75 | 0.35 32 | 0.72 88 | 1.50 13 | 4.54 16 | 4.36 31 |
TV-L1-MCT [64] | 43.3 | 0.24 26 | 1.09 43 | 0.20 57 | 0.78 71 | 3.45 50 | 0.62 78 | 0.86 43 | 5.03 49 | 0.46 46 | 0.27 26 | 3.48 16 | 0.20 37 | 3.92 30 | 4.94 32 | 3.06 40 | 4.00 80 | 4.80 31 | 2.05 63 | 0.33 23 | 0.32 14 | 0.64 59 | 2.47 48 | 5.20 30 | 5.59 48 |
Classic+NL [31] | 44.0 | 0.25 34 | 1.09 43 | 0.20 57 | 0.66 49 | 3.15 37 | 0.51 49 | 0.68 21 | 4.28 33 | 0.46 46 | 0.34 50 | 3.70 22 | 0.22 56 | 3.55 18 | 4.59 17 | 2.29 25 | 3.76 72 | 4.67 19 | 2.30 83 | 0.52 87 | 0.39 62 | 0.66 65 | 2.04 32 | 4.96 26 | 5.83 54 |
OAR-Flow [125] | 44.4 | 0.26 41 | 1.11 48 | 0.14 15 | 0.73 61 | 4.65 107 | 0.44 34 | 1.58 65 | 6.32 69 | 0.47 53 | 0.36 55 | 6.57 85 | 0.13 4 | 4.73 74 | 5.63 88 | 4.10 69 | 1.19 11 | 5.01 45 | 0.42 2 | 0.27 4 | 0.32 14 | 0.42 16 | 2.31 44 | 5.41 39 | 3.90 23 |
OFH [38] | 44.6 | 0.31 56 | 0.97 10 | 0.23 79 | 0.49 16 | 3.57 61 | 0.29 7 | 1.68 69 | 6.87 84 | 0.33 18 | 0.31 40 | 7.52 112 | 0.15 14 | 4.62 66 | 5.44 62 | 4.26 76 | 1.36 18 | 6.26 78 | 0.68 9 | 0.30 12 | 0.33 21 | 0.39 12 | 2.93 57 | 6.20 57 | 4.92 37 |
MDP-Flow [26] | 44.7 | 0.28 47 | 1.00 19 | 0.22 73 | 0.54 20 | 2.56 14 | 0.49 46 | 0.72 27 | 2.40 15 | 0.56 62 | 0.38 59 | 5.14 52 | 0.24 63 | 3.85 28 | 4.78 25 | 2.95 35 | 2.29 46 | 4.68 20 | 1.83 50 | 0.37 39 | 0.37 48 | 0.47 25 | 4.44 85 | 8.33 92 | 6.47 83 |
S2F-IF [123] | 44.8 | 0.46 93 | 1.19 70 | 0.23 79 | 0.74 63 | 3.08 33 | 0.57 70 | 1.09 49 | 3.74 25 | 0.78 73 | 0.48 69 | 3.19 11 | 0.32 77 | 4.51 55 | 5.40 58 | 3.90 59 | 1.13 9 | 4.91 38 | 0.82 17 | 0.32 20 | 0.32 14 | 0.44 21 | 2.11 37 | 4.89 24 | 2.32 11 |
FlowFields [110] | 46.2 | 0.48 96 | 1.18 67 | 0.24 86 | 0.77 69 | 2.87 23 | 0.62 78 | 1.17 53 | 3.79 26 | 0.85 77 | 0.49 70 | 3.12 10 | 0.33 78 | 4.45 52 | 5.33 54 | 3.53 47 | 1.19 11 | 4.94 39 | 0.86 19 | 0.32 20 | 0.31 11 | 0.41 15 | 2.49 50 | 5.61 43 | 2.75 14 |
Sparse Occlusion [54] | 46.5 | 0.24 26 | 1.01 21 | 0.18 37 | 0.62 37 | 2.87 23 | 0.54 58 | 0.93 45 | 6.27 67 | 0.38 28 | 0.30 33 | 5.16 53 | 0.21 47 | 4.14 40 | 5.21 45 | 3.08 41 | 1.70 25 | 4.62 15 | 1.31 31 | 0.60 105 | 0.62 121 | 0.68 77 | 2.46 47 | 5.70 46 | 5.56 47 |
Complementary OF [21] | 53.4 | 0.32 59 | 1.04 23 | 0.22 73 | 0.41 6 | 3.36 44 | 0.26 2 | 0.82 41 | 4.95 46 | 0.37 26 | 0.32 42 | 6.71 88 | 0.21 47 | 5.20 106 | 5.66 92 | 5.72 118 | 24.5 129 | 5.71 63 | 0.91 22 | 0.33 23 | 0.32 14 | 0.55 44 | 3.10 62 | 6.16 55 | 5.88 56 |
2DHMM-SAS [92] | 53.4 | 0.25 34 | 1.07 34 | 0.19 41 | 0.76 67 | 3.63 68 | 0.55 64 | 1.29 58 | 6.70 80 | 0.52 58 | 0.34 50 | 4.11 31 | 0.22 56 | 3.88 29 | 4.73 24 | 2.58 30 | 4.27 84 | 4.88 37 | 1.97 59 | 0.50 80 | 0.40 70 | 0.68 77 | 2.35 46 | 5.70 46 | 5.95 58 |
HBM-GC [105] | 53.5 | 0.28 47 | 1.33 95 | 0.20 57 | 0.63 40 | 3.44 49 | 0.55 64 | 0.62 15 | 5.03 49 | 0.46 46 | 0.33 47 | 4.59 40 | 0.28 67 | 3.99 36 | 4.98 35 | 2.68 32 | 2.71 60 | 4.10 6 | 0.72 11 | 0.64 108 | 0.42 81 | 0.90 103 | 2.20 41 | 6.62 66 | 6.50 89 |
S2D-Matching [84] | 53.9 | 0.24 26 | 1.35 97 | 0.20 57 | 0.65 44 | 3.62 65 | 0.51 49 | 1.29 58 | 6.38 71 | 0.44 40 | 0.30 33 | 3.45 15 | 0.20 37 | 4.17 42 | 5.29 50 | 2.86 34 | 4.25 83 | 4.79 29 | 2.29 78 | 0.54 95 | 0.39 62 | 0.70 82 | 1.97 27 | 5.09 28 | 6.52 91 |
Occlusion-TV-L1 [63] | 54.0 | 0.27 44 | 1.08 39 | 0.16 26 | 0.61 36 | 3.58 64 | 0.47 40 | 2.20 82 | 7.85 114 | 0.45 43 | 0.32 42 | 5.18 55 | 0.17 20 | 4.60 64 | 5.43 60 | 3.95 61 | 2.33 50 | 6.22 76 | 2.14 70 | 0.27 4 | 0.34 25 | 0.28 3 | 5.04 91 | 8.77 101 | 6.49 86 |
AggregFlow [97] | 54.0 | 0.34 65 | 1.48 110 | 0.17 33 | 0.94 85 | 4.38 99 | 0.60 72 | 1.65 68 | 5.13 53 | 1.01 86 | 0.30 33 | 6.85 93 | 0.20 37 | 4.45 52 | 5.53 74 | 3.36 46 | 1.04 4 | 4.58 14 | 0.46 3 | 0.35 26 | 0.42 81 | 0.47 25 | 2.50 51 | 5.60 42 | 5.50 45 |
ACK-Prior [27] | 54.3 | 0.27 44 | 0.98 13 | 0.20 57 | 0.44 9 | 2.96 26 | 0.29 7 | 0.63 18 | 4.42 36 | 0.28 6 | 0.24 13 | 5.07 50 | 0.18 24 | 4.46 54 | 5.22 46 | 4.13 70 | 23.3 128 | 6.26 78 | 3.16 104 | 0.83 119 | 0.57 116 | 1.12 110 | 4.06 75 | 6.70 68 | 4.42 32 |
DPOF [18] | 54.5 | 0.43 88 | 1.12 53 | 0.19 41 | 0.90 80 | 3.38 46 | 0.56 68 | 0.76 33 | 1.50 1 | 0.55 60 | 0.46 66 | 3.50 17 | 0.36 82 | 3.07 11 | 4.29 9 | 1.64 15 | 9.19 106 | 7.16 96 | 2.59 95 | 0.93 121 | 0.40 70 | 1.36 117 | 1.82 20 | 3.73 4 | 1.68 8 |
SRR-TVOF-NL [91] | 55.1 | 0.34 65 | 1.12 53 | 0.20 57 | 1.01 87 | 3.98 93 | 0.54 58 | 1.74 71 | 5.48 59 | 0.79 74 | 0.36 55 | 3.43 14 | 0.22 56 | 4.23 43 | 4.95 33 | 4.73 85 | 1.47 20 | 4.33 9 | 1.43 36 | 0.52 87 | 0.51 106 | 0.70 82 | 2.16 38 | 4.37 13 | 4.23 29 |
PGM-C [120] | 55.3 | 0.48 96 | 1.24 80 | 0.23 79 | 0.77 69 | 3.62 65 | 0.63 82 | 1.23 56 | 4.33 35 | 0.87 79 | 0.56 77 | 5.51 62 | 0.30 73 | 4.58 62 | 5.49 69 | 3.87 54 | 1.48 21 | 5.50 58 | 1.34 32 | 0.30 12 | 0.29 4 | 0.43 18 | 2.65 53 | 6.11 54 | 4.96 38 |
TCOF [69] | 55.8 | 0.34 65 | 1.06 27 | 0.19 41 | 0.71 57 | 3.50 56 | 0.51 49 | 1.88 74 | 7.86 117 | 0.87 79 | 0.64 81 | 4.88 47 | 0.59 97 | 4.59 63 | 5.39 57 | 4.34 78 | 2.28 44 | 4.41 11 | 2.05 63 | 0.38 42 | 0.39 62 | 0.57 50 | 1.97 27 | 4.87 23 | 4.27 30 |
ComplOF-FED-GPU [35] | 56.2 | 0.32 59 | 1.00 19 | 0.19 41 | 0.69 54 | 3.71 76 | 0.31 10 | 0.98 47 | 5.10 51 | 0.36 22 | 0.35 53 | 6.81 90 | 0.18 24 | 4.60 64 | 5.43 60 | 4.17 72 | 12.7 110 | 6.78 89 | 1.36 33 | 0.44 65 | 0.36 42 | 0.82 97 | 3.18 64 | 6.56 64 | 5.43 44 |
ROF-ND [107] | 56.5 | 0.38 77 | 1.16 63 | 0.21 67 | 0.91 82 | 3.47 53 | 0.33 14 | 0.95 46 | 6.42 73 | 0.32 16 | 0.43 62 | 1.11 1 | 0.37 83 | 4.34 45 | 5.20 44 | 4.09 68 | 2.35 51 | 5.23 52 | 1.91 53 | 0.62 106 | 0.44 87 | 0.79 92 | 3.43 67 | 5.34 35 | 3.34 19 |
Kuang [131] | 58.1 | 0.46 93 | 1.22 74 | 0.22 73 | 0.83 75 | 3.37 45 | 0.57 70 | 1.39 61 | 4.25 32 | 0.77 72 | 0.49 70 | 4.57 39 | 0.27 65 | 4.41 50 | 5.26 48 | 3.72 50 | 2.67 58 | 5.29 54 | 2.43 91 | 0.31 17 | 0.29 4 | 0.44 21 | 3.30 66 | 6.41 59 | 7.00 107 |
DeepFlow2 [108] | 58.4 | 0.34 65 | 1.10 46 | 0.17 33 | 0.69 54 | 3.81 82 | 0.45 38 | 1.45 62 | 6.41 72 | 0.75 69 | 0.71 86 | 6.00 72 | 0.22 56 | 4.53 58 | 5.44 62 | 3.90 59 | 1.75 27 | 5.96 69 | 0.89 21 | 0.35 26 | 0.33 21 | 0.64 59 | 4.40 84 | 7.85 84 | 6.67 97 |
CPM-Flow [116] | 58.5 | 0.48 96 | 1.25 84 | 0.23 79 | 0.78 71 | 3.67 72 | 0.63 82 | 1.21 55 | 4.28 33 | 0.87 79 | 0.55 76 | 5.69 67 | 0.30 73 | 4.55 59 | 5.47 65 | 3.79 53 | 1.93 35 | 5.05 47 | 1.51 40 | 0.30 12 | 0.29 4 | 0.43 18 | 3.04 60 | 6.58 65 | 6.38 78 |
RFlow [90] | 60.5 | 0.29 54 | 1.05 24 | 0.19 41 | 0.39 4 | 3.46 51 | 0.28 4 | 1.80 72 | 7.45 100 | 0.31 13 | 0.30 33 | 7.40 106 | 0.19 32 | 5.17 104 | 5.76 110 | 5.24 98 | 2.26 42 | 6.13 74 | 1.98 60 | 0.45 67 | 0.36 42 | 0.56 45 | 4.48 87 | 7.88 86 | 6.75 103 |
Steered-L1 [118] | 61.0 | 0.23 22 | 0.83 2 | 0.15 20 | 0.31 1 | 2.97 27 | 0.26 2 | 0.62 15 | 4.90 44 | 0.28 6 | 0.29 28 | 6.50 83 | 0.14 11 | 4.69 71 | 5.44 62 | 4.95 89 | 13.5 111 | 6.38 84 | 3.53 108 | 0.95 123 | 0.53 109 | 2.54 121 | 7.20 112 | 8.68 98 | 8.48 114 |
Aniso-Texture [82] | 61.1 | 0.21 11 | 0.92 5 | 0.18 37 | 0.48 15 | 3.62 65 | 0.44 34 | 2.19 81 | 7.77 109 | 0.42 34 | 0.26 24 | 3.01 9 | 0.19 32 | 5.36 117 | 5.86 119 | 5.81 120 | 2.36 53 | 6.34 82 | 1.59 45 | 0.55 96 | 0.53 109 | 0.64 59 | 3.48 68 | 6.76 70 | 6.32 72 |
SimpleFlow [49] | 61.4 | 0.27 44 | 1.10 46 | 0.22 73 | 0.76 67 | 3.63 68 | 0.62 78 | 1.38 60 | 7.03 90 | 0.53 59 | 0.42 61 | 3.90 28 | 0.25 64 | 4.16 41 | 5.08 39 | 2.98 37 | 20.7 126 | 6.30 81 | 2.31 85 | 0.41 52 | 0.40 70 | 0.59 54 | 2.00 31 | 5.38 37 | 6.47 83 |
Adaptive [20] | 61.8 | 0.26 41 | 1.15 60 | 0.12 7 | 0.65 44 | 3.63 68 | 0.50 48 | 2.48 87 | 8.26 129 | 0.45 43 | 0.36 55 | 4.65 43 | 0.19 32 | 4.57 61 | 5.38 56 | 4.08 66 | 3.92 78 | 6.03 73 | 1.94 54 | 0.46 70 | 0.47 96 | 0.58 53 | 3.22 65 | 7.41 74 | 6.43 80 |
CRTflow [80] | 62.6 | 0.37 76 | 1.06 27 | 0.19 41 | 0.63 40 | 3.57 61 | 0.44 34 | 1.71 70 | 7.76 108 | 0.50 57 | 0.49 70 | 7.44 107 | 0.20 37 | 4.77 77 | 5.70 97 | 4.00 63 | 1.84 30 | 8.40 120 | 1.58 43 | 0.36 31 | 0.33 21 | 0.56 45 | 4.13 78 | 8.33 92 | 6.37 77 |
EpicFlow [102] | 63.6 | 0.48 96 | 1.24 80 | 0.23 79 | 0.79 74 | 3.69 73 | 0.63 82 | 1.51 64 | 5.73 63 | 0.87 79 | 0.56 77 | 5.32 59 | 0.30 73 | 4.63 68 | 5.56 83 | 4.08 66 | 3.58 68 | 5.91 68 | 1.63 47 | 0.30 12 | 0.29 4 | 0.43 18 | 3.08 61 | 6.36 58 | 6.33 74 |
TF+OM [100] | 66.6 | 0.28 47 | 1.23 76 | 0.16 26 | 0.54 20 | 3.53 58 | 0.43 32 | 1.91 76 | 4.53 40 | 2.12 97 | 0.30 33 | 5.24 58 | 0.21 47 | 5.48 123 | 5.91 122 | 5.59 114 | 2.43 56 | 6.24 77 | 0.76 13 | 0.52 87 | 0.49 102 | 0.67 70 | 4.26 81 | 7.69 81 | 6.09 63 |
SIOF [67] | 68.2 | 0.26 41 | 1.29 89 | 0.16 26 | 1.09 91 | 3.92 91 | 0.47 40 | 3.40 97 | 6.92 88 | 2.58 104 | 0.77 88 | 5.51 62 | 0.39 85 | 4.87 86 | 5.54 78 | 5.00 90 | 1.88 31 | 5.61 60 | 1.83 50 | 0.39 44 | 0.39 62 | 0.47 25 | 3.64 69 | 6.69 67 | 6.33 74 |
DeepFlow [86] | 68.4 | 0.35 71 | 1.11 48 | 0.24 86 | 0.83 75 | 3.86 84 | 0.52 53 | 1.83 73 | 6.36 70 | 1.34 91 | 0.86 90 | 7.76 120 | 0.27 65 | 4.51 55 | 5.49 69 | 3.76 51 | 1.92 34 | 6.38 84 | 0.96 23 | 0.36 31 | 0.31 11 | 0.67 70 | 5.10 93 | 8.51 95 | 6.71 100 |
BriefMatch [124] | 68.8 | 0.25 34 | 1.06 27 | 0.15 20 | 0.65 44 | 3.15 37 | 0.36 19 | 0.98 47 | 3.39 21 | 0.33 18 | 0.24 13 | 7.01 100 | 0.15 14 | 4.70 72 | 5.31 53 | 5.47 111 | 8.12 104 | 7.15 95 | 3.69 111 | 0.84 120 | 0.47 96 | 3.37 128 | 9.00 122 | 10.4 116 | 11.5 128 |
Aniso. Huber-L1 [22] | 70.0 | 0.31 56 | 1.11 48 | 0.20 57 | 1.03 88 | 3.73 78 | 0.83 94 | 2.25 84 | 7.75 107 | 0.96 85 | 0.67 83 | 4.18 32 | 0.43 89 | 4.63 68 | 5.48 67 | 3.89 58 | 1.88 31 | 5.24 53 | 1.43 36 | 0.52 87 | 0.45 88 | 0.84 99 | 3.00 58 | 6.93 71 | 6.07 62 |
FlowNet2 [122] | 71.1 | 0.72 112 | 1.55 111 | 0.38 105 | 2.00 106 | 4.44 100 | 1.68 107 | 2.15 79 | 4.45 37 | 2.07 96 | 0.56 77 | 6.86 95 | 0.38 84 | 3.99 36 | 4.99 36 | 3.00 38 | 2.63 57 | 5.85 66 | 2.23 73 | 0.42 57 | 0.60 118 | 0.51 38 | 2.16 38 | 5.29 34 | 1.39 7 |
Classic++ [32] | 72.1 | 0.28 47 | 1.16 63 | 0.21 67 | 0.64 43 | 3.88 85 | 0.52 53 | 1.96 78 | 6.43 74 | 0.61 64 | 0.36 55 | 6.46 82 | 0.20 37 | 4.72 73 | 5.63 88 | 3.88 55 | 2.31 48 | 7.38 100 | 2.28 75 | 0.56 97 | 0.45 88 | 0.70 82 | 5.09 92 | 8.15 90 | 6.57 94 |
TriangleFlow [30] | 72.4 | 0.32 59 | 1.23 76 | 0.23 79 | 0.72 60 | 4.53 104 | 0.42 31 | 1.60 66 | 7.09 92 | 0.36 22 | 0.33 47 | 4.75 45 | 0.18 24 | 5.39 120 | 5.96 123 | 6.08 123 | 5.72 96 | 5.96 69 | 1.95 57 | 0.49 75 | 0.62 121 | 0.68 77 | 3.00 58 | 6.41 59 | 5.85 55 |
LocallyOriented [52] | 73.3 | 0.41 85 | 1.38 103 | 0.19 41 | 1.23 93 | 4.25 96 | 0.76 91 | 3.57 103 | 7.52 102 | 1.10 89 | 0.52 74 | 3.81 26 | 0.28 67 | 4.88 89 | 5.47 65 | 4.35 79 | 7.73 101 | 7.52 105 | 1.49 39 | 0.35 26 | 0.35 32 | 0.56 45 | 4.49 88 | 6.47 62 | 5.95 58 |
TV-L1-improved [17] | 73.9 | 0.28 47 | 1.05 24 | 0.17 33 | 0.57 30 | 3.55 60 | 0.44 34 | 2.24 83 | 8.19 125 | 0.46 46 | 0.35 53 | 7.45 108 | 0.18 24 | 4.86 85 | 5.68 94 | 4.21 75 | 17.3 119 | 7.54 106 | 2.47 93 | 0.56 97 | 0.48 100 | 0.72 88 | 4.10 77 | 7.73 82 | 6.51 90 |
SegOF [10] | 74.4 | 0.56 109 | 1.27 86 | 0.41 106 | 1.94 104 | 3.85 83 | 1.77 109 | 3.44 99 | 6.14 66 | 1.96 94 | 1.26 103 | 3.34 12 | 0.92 106 | 4.83 82 | 5.24 47 | 4.79 87 | 16.6 118 | 7.57 107 | 4.51 118 | 0.21 1 | 0.32 14 | 0.32 5 | 2.87 55 | 6.41 59 | 3.00 16 |
Brox et al. [5] | 76.0 | 0.38 77 | 1.15 60 | 0.27 95 | 0.84 77 | 3.90 88 | 0.69 87 | 1.45 62 | 5.36 56 | 0.87 79 | 0.98 92 | 6.75 89 | 0.28 67 | 5.17 104 | 5.73 103 | 5.35 105 | 6.14 98 | 7.37 99 | 2.79 99 | 0.27 4 | 0.36 42 | 0.30 4 | 4.93 89 | 7.45 75 | 6.32 72 |
F-TV-L1 [15] | 76.3 | 0.38 77 | 1.23 76 | 0.30 98 | 1.32 96 | 4.46 102 | 0.67 86 | 3.59 104 | 7.35 97 | 0.70 67 | 0.69 85 | 7.63 114 | 0.29 72 | 4.74 75 | 5.54 78 | 4.30 77 | 3.49 67 | 6.97 93 | 1.94 54 | 0.38 42 | 0.43 84 | 0.42 16 | 3.65 70 | 7.59 78 | 3.92 24 |
CBF [12] | 76.6 | 0.32 59 | 0.99 15 | 0.20 57 | 1.00 86 | 3.48 54 | 0.97 97 | 1.61 67 | 6.55 78 | 0.81 75 | 0.66 82 | 6.31 78 | 0.46 90 | 4.87 86 | 5.65 91 | 4.92 88 | 2.31 48 | 5.30 55 | 1.62 46 | 0.77 117 | 0.54 112 | 1.11 108 | 4.07 76 | 7.61 79 | 6.58 95 |
DF-Auto [115] | 77.8 | 0.47 95 | 1.40 105 | 0.22 73 | 1.58 99 | 3.91 90 | 1.18 102 | 2.42 85 | 6.56 79 | 2.21 100 | 1.12 96 | 6.32 79 | 0.51 93 | 4.99 90 | 5.70 97 | 5.07 91 | 1.29 15 | 5.54 59 | 1.03 25 | 0.40 50 | 0.51 106 | 0.36 8 | 3.91 74 | 7.83 83 | 6.33 74 |
p-harmonic [29] | 78.1 | 0.39 82 | 1.15 60 | 0.33 100 | 0.74 63 | 3.63 68 | 0.61 77 | 2.42 85 | 7.91 119 | 0.82 76 | 1.01 93 | 4.53 38 | 0.66 101 | 5.21 107 | 5.70 97 | 5.59 114 | 1.74 26 | 6.29 80 | 1.54 41 | 0.45 67 | 0.41 77 | 0.53 42 | 5.19 95 | 8.53 96 | 6.31 70 |
SuperFlow [81] | 78.3 | 0.34 65 | 1.13 57 | 0.21 67 | 1.31 95 | 3.57 61 | 1.11 98 | 2.62 91 | 6.44 75 | 2.67 105 | 1.19 100 | 5.82 70 | 0.77 103 | 5.01 92 | 5.61 87 | 5.26 99 | 2.39 54 | 5.84 65 | 2.34 86 | 0.43 61 | 0.47 96 | 0.48 30 | 4.46 86 | 7.92 87 | 5.62 50 |
Local-TV-L1 [65] | 79.4 | 0.43 88 | 1.24 80 | 0.29 97 | 1.98 105 | 4.55 105 | 1.15 101 | 4.88 111 | 7.48 101 | 2.20 99 | 1.35 104 | 6.92 97 | 0.63 100 | 4.63 68 | 5.50 72 | 4.04 65 | 2.22 41 | 5.75 64 | 1.94 54 | 0.36 31 | 0.31 11 | 0.47 25 | 5.53 98 | 7.86 85 | 6.89 104 |
CLG-TV [48] | 79.5 | 0.33 63 | 1.07 34 | 0.22 73 | 0.90 80 | 3.76 79 | 0.75 90 | 2.15 79 | 7.98 121 | 0.73 68 | 0.67 83 | 4.67 44 | 0.53 94 | 4.84 83 | 5.55 81 | 4.44 80 | 2.11 37 | 7.44 102 | 2.05 63 | 0.57 101 | 0.56 115 | 0.86 100 | 4.18 80 | 8.19 91 | 6.25 67 |
Fusion [6] | 80.2 | 0.38 77 | 1.17 65 | 0.28 96 | 0.53 19 | 3.35 43 | 0.48 43 | 0.90 44 | 3.47 22 | 0.76 71 | 0.63 80 | 5.66 66 | 0.39 85 | 5.21 107 | 5.79 116 | 5.33 104 | 4.41 86 | 5.98 71 | 2.82 100 | 0.56 97 | 0.51 106 | 0.75 90 | 6.49 109 | 11.3 120 | 7.02 108 |
TriFlow [95] | 80.4 | 0.38 77 | 1.41 106 | 0.19 41 | 0.92 84 | 3.99 94 | 0.77 93 | 3.14 95 | 6.44 75 | 2.81 109 | 0.47 67 | 5.07 50 | 0.39 85 | 5.37 118 | 5.86 119 | 5.27 101 | 1.22 14 | 5.68 62 | 0.79 15 | 1.96 127 | 0.71 126 | 2.60 122 | 3.11 63 | 6.09 53 | 4.44 33 |
Bartels [41] | 81.5 | 0.34 65 | 1.28 87 | 0.26 93 | 0.52 18 | 3.07 30 | 0.41 28 | 1.19 54 | 5.46 58 | 0.46 46 | 0.43 62 | 7.93 121 | 0.31 76 | 5.14 99 | 5.73 103 | 5.26 99 | 5.91 97 | 7.40 101 | 2.13 69 | 0.64 108 | 0.46 93 | 1.26 115 | 6.18 105 | 10.0 115 | 8.44 113 |
Dynamic MRF [7] | 83.6 | 0.36 72 | 1.26 85 | 0.24 86 | 0.57 30 | 4.28 98 | 0.33 14 | 1.88 74 | 6.72 81 | 0.37 26 | 0.45 65 | 6.86 95 | 0.23 61 | 5.44 122 | 5.89 121 | 5.60 116 | 12.3 109 | 11.8 128 | 4.03 116 | 0.42 57 | 0.30 8 | 0.79 92 | 7.94 117 | 11.0 118 | 9.07 115 |
Rannacher [23] | 83.6 | 0.33 63 | 1.19 70 | 0.25 91 | 0.75 65 | 3.88 85 | 0.60 72 | 2.69 92 | 8.38 130 | 0.63 66 | 0.47 67 | 7.51 111 | 0.28 67 | 4.78 79 | 5.64 90 | 4.00 63 | 17.7 123 | 7.45 103 | 2.85 101 | 0.50 80 | 0.40 70 | 0.70 82 | 3.80 72 | 7.61 79 | 6.49 86 |
Second-order prior [8] | 84.7 | 0.36 72 | 1.22 74 | 0.21 67 | 1.07 89 | 3.69 73 | 0.70 89 | 2.58 90 | 7.79 111 | 1.06 88 | 0.78 89 | 5.23 57 | 0.28 67 | 4.77 77 | 5.58 85 | 4.74 86 | 3.07 63 | 7.16 96 | 2.63 96 | 0.64 108 | 0.45 88 | 0.79 92 | 4.31 82 | 7.93 88 | 6.99 106 |
LDOF [28] | 84.9 | 0.42 86 | 1.29 89 | 0.21 67 | 1.35 97 | 4.55 105 | 0.69 87 | 1.91 76 | 4.97 47 | 1.24 90 | 1.56 106 | 11.8 131 | 0.49 91 | 5.15 100 | 5.75 109 | 5.19 95 | 4.76 88 | 8.27 118 | 2.28 75 | 0.32 20 | 0.35 32 | 0.56 45 | 5.18 94 | 8.80 103 | 6.49 86 |
StereoFlow [44] | 85.5 | 1.36 128 | 2.53 128 | 0.92 123 | 3.87 122 | 5.71 124 | 2.94 120 | 3.43 98 | 6.77 82 | 2.99 111 | 3.06 113 | 9.56 129 | 2.88 114 | 4.78 79 | 5.57 84 | 4.67 84 | 1.35 16 | 6.94 92 | 0.97 24 | 0.22 2 | 0.28 2 | 0.27 2 | 4.93 89 | 8.34 94 | 6.55 93 |
FlowNetS+ft+v [112] | 86.2 | 0.39 82 | 1.21 73 | 0.19 41 | 1.19 92 | 3.94 92 | 0.83 94 | 3.45 100 | 7.87 118 | 2.02 95 | 1.18 99 | 7.17 104 | 0.59 97 | 5.08 96 | 5.72 102 | 5.16 94 | 2.42 55 | 6.39 87 | 2.22 72 | 0.48 73 | 0.45 88 | 1.46 118 | 3.88 73 | 7.32 73 | 5.74 51 |
StereoOF-V1MT [119] | 89.0 | 0.43 88 | 1.30 93 | 0.23 79 | 1.53 98 | 5.30 115 | 0.43 32 | 3.31 96 | 7.14 93 | 0.55 60 | 0.97 91 | 5.55 65 | 0.35 81 | 5.24 111 | 5.73 103 | 5.22 96 | 18.1 124 | 7.92 109 | 3.97 115 | 0.35 26 | 0.35 32 | 0.67 70 | 8.71 121 | 11.5 121 | 9.09 116 |
Ad-TV-NDC [36] | 89.0 | 0.89 117 | 1.39 104 | 1.05 125 | 4.30 124 | 5.21 114 | 3.61 123 | 7.03 124 | 7.85 114 | 2.68 106 | 2.01 108 | 4.92 48 | 1.93 108 | 4.55 59 | 5.53 74 | 3.64 48 | 2.30 47 | 7.11 94 | 2.02 62 | 0.37 39 | 0.34 25 | 0.48 30 | 8.69 120 | 9.20 106 | 9.22 117 |
UnFlow [129] | 89.6 | 1.23 123 | 2.28 125 | 0.66 116 | 2.36 109 | 3.90 88 | 1.90 111 | 3.94 106 | 5.69 62 | 2.23 101 | 2.29 110 | 5.87 71 | 2.12 110 | 5.37 118 | 5.74 108 | 5.74 119 | 4.39 85 | 7.49 104 | 3.63 109 | 0.25 3 | 0.36 42 | 0.23 1 | 4.16 79 | 9.27 107 | 5.41 43 |
Filter Flow [19] | 90.8 | 0.51 103 | 1.41 106 | 0.37 103 | 1.86 103 | 3.79 81 | 1.12 99 | 4.19 109 | 6.46 77 | 3.20 117 | 3.16 114 | 5.17 54 | 3.07 115 | 5.03 93 | 5.48 67 | 5.54 112 | 3.06 62 | 5.33 56 | 2.41 90 | 0.57 101 | 0.60 118 | 0.60 55 | 5.49 97 | 7.53 77 | 6.27 69 |
Shiralkar [42] | 91.2 | 0.44 92 | 1.24 80 | 0.24 86 | 1.29 94 | 4.44 100 | 0.51 49 | 3.66 105 | 8.22 126 | 0.75 69 | 1.05 94 | 6.20 76 | 0.39 85 | 4.87 86 | 5.52 73 | 4.61 83 | 5.13 91 | 8.26 117 | 2.90 103 | 0.75 112 | 0.41 77 | 1.11 108 | 6.02 102 | 8.77 101 | 6.39 79 |
GroupFlow [9] | 91.5 | 0.78 114 | 1.73 115 | 0.60 114 | 3.11 116 | 5.38 118 | 2.49 117 | 4.27 110 | 6.91 87 | 2.76 108 | 1.21 101 | 4.61 42 | 0.79 104 | 5.09 97 | 5.76 110 | 3.88 55 | 7.80 103 | 7.20 98 | 5.19 123 | 0.31 17 | 0.42 81 | 0.45 23 | 4.34 83 | 8.00 89 | 6.31 70 |
Learning Flow [11] | 92.0 | 0.36 72 | 1.29 89 | 0.19 41 | 0.88 79 | 4.18 95 | 0.56 68 | 2.94 93 | 6.87 84 | 1.01 86 | 1.78 107 | 6.81 90 | 0.50 92 | 5.51 124 | 6.19 126 | 5.41 109 | 14.2 114 | 8.05 111 | 3.44 107 | 0.41 52 | 0.54 112 | 0.57 50 | 6.48 108 | 9.99 114 | 6.48 85 |
IAOF2 [51] | 92.3 | 0.42 86 | 1.36 100 | 0.30 98 | 1.58 99 | 4.27 97 | 1.14 100 | 3.56 102 | 7.66 105 | 2.70 107 | 3.55 118 | 5.20 56 | 3.60 119 | 4.84 83 | 5.67 93 | 4.15 71 | 4.20 82 | 6.02 72 | 2.87 102 | 0.56 97 | 0.46 93 | 0.98 104 | 5.41 96 | 7.19 72 | 6.10 64 |
CNN-flow-warp+ref [117] | 92.6 | 0.49 101 | 1.29 89 | 0.35 101 | 1.08 90 | 3.70 75 | 0.94 96 | 3.01 94 | 7.08 91 | 1.40 92 | 1.12 96 | 7.19 105 | 0.58 96 | 5.23 109 | 5.77 113 | 5.36 106 | 5.17 93 | 8.89 124 | 2.38 88 | 0.28 9 | 0.32 14 | 0.67 70 | 10.1 125 | 11.6 124 | 10.5 121 |
GraphCuts [14] | 93.2 | 0.53 105 | 1.30 93 | 0.26 93 | 2.63 112 | 5.49 120 | 1.53 105 | 2.57 89 | 5.15 54 | 2.57 103 | 1.14 98 | 4.94 49 | 0.61 99 | 4.40 49 | 5.13 42 | 3.95 61 | 17.3 119 | 6.38 84 | 5.59 125 | 0.75 112 | 0.43 84 | 1.14 111 | 6.31 107 | 9.98 113 | 7.32 109 |
HCIC-L [99] | 96.2 | 1.29 126 | 2.09 121 | 0.69 119 | 6.78 130 | 6.72 128 | 7.04 130 | 3.50 101 | 5.40 57 | 3.11 115 | 5.45 125 | 7.14 103 | 5.37 126 | 4.52 57 | 5.40 58 | 3.13 42 | 3.84 77 | 5.12 48 | 3.89 114 | 2.11 129 | 1.50 129 | 2.77 123 | 3.75 71 | 6.50 63 | 3.31 18 |
2D-CLG [1] | 96.4 | 1.19 122 | 2.21 123 | 0.70 120 | 2.17 107 | 3.78 80 | 1.93 113 | 6.02 116 | 7.44 99 | 3.48 124 | 3.42 116 | 6.03 73 | 3.32 116 | 5.24 111 | 5.68 94 | 5.39 108 | 15.6 116 | 8.14 113 | 3.77 112 | 0.31 17 | 0.26 1 | 0.48 30 | 5.99 101 | 8.71 100 | 6.74 101 |
Black & Anandan [4] | 96.5 | 0.54 107 | 1.34 96 | 0.48 110 | 2.79 114 | 4.85 109 | 1.86 110 | 6.61 120 | 7.85 114 | 2.99 111 | 2.16 109 | 5.71 68 | 1.95 109 | 5.15 100 | 5.71 101 | 5.30 102 | 11.7 107 | 8.34 119 | 3.41 106 | 0.36 31 | 0.41 77 | 0.37 10 | 5.98 100 | 8.99 104 | 6.44 81 |
IAOF [50] | 97.3 | 0.54 107 | 1.35 97 | 0.47 109 | 2.65 113 | 5.31 116 | 1.90 111 | 8.88 131 | 9.21 131 | 3.37 121 | 3.44 117 | 5.36 60 | 3.47 117 | 4.74 75 | 5.53 74 | 4.56 82 | 3.32 65 | 6.48 88 | 2.37 87 | 0.53 94 | 0.39 62 | 0.68 77 | 7.36 115 | 7.48 76 | 7.98 111 |
Nguyen [33] | 98.4 | 0.71 111 | 1.68 113 | 0.49 111 | 2.50 111 | 4.86 110 | 2.21 115 | 7.41 126 | 7.97 120 | 3.47 123 | 3.87 120 | 5.77 69 | 3.73 120 | 5.05 95 | 5.76 110 | 5.14 92 | 3.81 76 | 7.76 108 | 2.63 96 | 0.39 44 | 0.36 42 | 0.57 50 | 6.17 104 | 8.67 97 | 6.69 98 |
SPSA-learn [13] | 101.0 | 0.53 105 | 1.36 100 | 0.45 107 | 2.47 110 | 5.72 125 | 1.69 108 | 4.97 114 | 7.78 110 | 2.93 110 | 2.51 111 | 7.47 109 | 2.31 111 | 5.30 114 | 5.83 118 | 5.44 110 | 22.3 127 | 7.93 110 | 4.87 120 | 0.36 31 | 0.34 25 | 0.51 38 | 6.03 103 | 9.76 112 | 6.66 96 |
Modified CLG [34] | 101.4 | 0.66 110 | 1.45 109 | 0.55 112 | 1.60 102 | 3.72 77 | 1.40 104 | 5.89 115 | 8.02 123 | 3.40 122 | 1.52 105 | 7.10 102 | 1.29 107 | 5.23 109 | 5.80 117 | 5.36 106 | 5.05 90 | 8.25 116 | 2.71 98 | 0.39 44 | 0.39 62 | 0.82 97 | 5.67 99 | 9.44 109 | 6.69 98 |
Horn & Schunck [3] | 102.0 | 0.73 113 | 1.72 114 | 0.59 113 | 2.87 115 | 4.93 112 | 2.05 114 | 6.26 119 | 7.71 106 | 3.28 119 | 3.80 119 | 6.43 80 | 3.48 118 | 5.10 98 | 5.55 81 | 5.31 103 | 8.89 105 | 10.1 126 | 4.92 121 | 0.41 52 | 0.46 93 | 0.36 8 | 6.67 110 | 9.11 105 | 6.95 105 |
HBpMotionGpu [43] | 103.0 | 0.50 102 | 1.56 112 | 0.35 101 | 2.28 108 | 5.51 121 | 1.67 106 | 6.22 118 | 7.82 112 | 3.03 113 | 1.10 95 | 8.84 127 | 0.69 102 | 5.58 125 | 6.09 125 | 5.55 113 | 3.29 64 | 6.34 82 | 2.47 93 | 0.52 87 | 0.48 100 | 0.64 59 | 6.21 106 | 8.69 99 | 6.74 101 |
TI-DOFE [24] | 103.1 | 1.13 121 | 2.07 120 | 1.14 126 | 3.43 120 | 4.87 111 | 3.26 122 | 7.27 125 | 7.84 113 | 3.50 125 | 4.33 122 | 6.11 74 | 4.19 122 | 5.04 94 | 5.58 85 | 5.15 93 | 6.52 99 | 10.3 127 | 4.70 119 | 0.36 31 | 0.37 48 | 0.53 42 | 7.85 116 | 9.38 108 | 8.17 112 |
2bit-BM-tele [98] | 103.8 | 0.51 103 | 1.42 108 | 0.45 107 | 0.91 82 | 4.49 103 | 0.76 91 | 2.54 88 | 7.60 104 | 0.91 84 | 0.73 87 | 7.71 117 | 0.55 95 | 5.30 114 | 5.99 124 | 5.23 97 | 5.13 91 | 6.80 90 | 2.30 83 | 1.15 126 | 0.64 124 | 2.10 120 | 7.02 111 | 11.6 124 | 9.85 118 |
BlockOverlap [61] | 104.5 | 0.43 88 | 1.36 100 | 0.37 103 | 1.58 99 | 3.89 87 | 1.30 103 | 4.01 107 | 7.31 96 | 2.26 102 | 1.22 102 | 7.93 121 | 0.84 105 | 5.24 111 | 5.70 97 | 6.40 125 | 4.80 89 | 6.20 75 | 2.44 92 | 0.80 118 | 0.55 114 | 3.83 129 | 7.28 114 | 9.60 110 | 10.0 120 |
Heeger++ [104] | 106.2 | 0.85 116 | 1.84 118 | 0.60 114 | 3.55 121 | 5.57 123 | 2.47 116 | 4.90 112 | 6.81 83 | 1.89 93 | 3.39 115 | 7.75 119 | 2.84 113 | 5.15 100 | 5.29 50 | 6.01 122 | 17.5 122 | 8.65 122 | 5.32 124 | 0.45 67 | 0.41 77 | 0.61 56 | 9.73 123 | 11.5 121 | 10.9 123 |
Adaptive flow [45] | 110.3 | 0.94 119 | 2.01 119 | 0.71 121 | 4.43 125 | 5.33 117 | 4.12 124 | 6.07 117 | 6.90 86 | 3.85 127 | 4.09 121 | 6.17 75 | 4.06 121 | 5.00 91 | 5.73 103 | 4.48 81 | 6.57 100 | 5.86 67 | 3.68 110 | 2.03 128 | 1.05 128 | 3.31 127 | 7.27 113 | 11.0 118 | 7.57 110 |
SILK [79] | 111.9 | 0.80 115 | 1.82 117 | 0.80 122 | 3.23 117 | 5.00 113 | 2.92 119 | 7.68 129 | 8.24 127 | 3.11 115 | 3.00 112 | 6.98 98 | 2.43 112 | 5.34 116 | 5.77 113 | 5.92 121 | 15.0 115 | 8.08 112 | 4.36 117 | 0.50 80 | 0.38 55 | 1.01 105 | 8.67 119 | 10.9 117 | 9.92 119 |
PGAM+LK [55] | 115.3 | 1.38 129 | 2.26 124 | 2.75 131 | 4.11 123 | 5.46 119 | 4.31 125 | 4.04 108 | 5.95 65 | 3.10 114 | 6.71 129 | 9.38 128 | 6.38 128 | 5.15 100 | 5.49 69 | 5.71 117 | 7.75 102 | 8.68 123 | 3.81 113 | 1.11 124 | 0.72 127 | 1.89 119 | 8.55 118 | 9.73 111 | 10.6 122 |
FFV1MT [106] | 115.7 | 0.93 118 | 2.34 126 | 0.67 118 | 3.35 118 | 5.54 122 | 2.51 118 | 4.92 113 | 6.95 89 | 3.22 118 | 5.72 126 | 8.77 126 | 5.31 125 | 5.74 126 | 5.73 103 | 7.13 126 | 13.6 112 | 15.3 131 | 6.77 127 | 0.49 75 | 0.49 102 | 0.75 90 | 9.73 123 | 11.5 121 | 10.9 123 |
SLK [47] | 116.6 | 1.25 124 | 2.15 122 | 1.34 127 | 3.35 118 | 4.75 108 | 3.16 121 | 7.53 127 | 8.14 124 | 3.31 120 | 5.19 124 | 6.99 99 | 4.99 124 | 5.42 121 | 5.69 96 | 6.26 124 | 16.1 117 | 8.44 121 | 4.98 122 | 0.59 104 | 0.39 62 | 1.26 115 | 10.7 126 | 12.1 127 | 11.0 125 |
Periodicity [78] | 121.5 | 1.29 126 | 3.11 131 | 0.66 116 | 8.81 131 | 9.31 131 | 9.57 131 | 7.66 128 | 7.36 98 | 6.46 131 | 6.64 128 | 9.79 130 | 6.12 127 | 22.6 131 | 24.0 131 | 21.6 131 | 20.0 125 | 15.2 130 | 12.8 131 | 0.27 4 | 0.53 109 | 2.84 124 | 13.6 131 | 17.6 131 | 13.3 131 |
FOLKI [16] | 124.1 | 1.10 120 | 2.44 127 | 0.96 124 | 6.01 129 | 6.49 127 | 6.33 129 | 6.80 121 | 8.01 122 | 3.62 126 | 4.98 123 | 7.71 117 | 4.88 123 | 5.98 127 | 6.25 127 | 7.48 127 | 12.2 108 | 14.8 129 | 5.96 126 | 1.11 124 | 0.59 117 | 2.88 125 | 11.0 127 | 11.9 126 | 11.5 128 |
Pyramid LK [2] | 124.3 | 1.25 124 | 1.79 116 | 1.67 130 | 5.06 126 | 5.74 126 | 5.11 126 | 7.76 130 | 8.24 127 | 4.54 128 | 6.31 127 | 6.43 80 | 6.90 129 | 12.3 128 | 13.8 128 | 10.9 128 | 17.4 121 | 9.60 125 | 6.83 128 | 0.94 122 | 0.60 118 | 3.13 126 | 12.6 130 | 16.8 130 | 11.7 130 |
AdaConv-v1 [126] | 126.2 | 2.32 130 | 2.67 129 | 1.41 128 | 5.68 127 | 7.08 129 | 5.41 127 | 6.88 122 | 7.15 94 | 6.23 129 | 7.42 130 | 8.22 124 | 7.70 130 | 13.5 129 | 14.0 129 | 12.0 129 | 31.6 130 | 8.14 113 | 8.03 129 | 3.93 130 | 2.37 130 | 4.62 130 | 11.8 128 | 16.0 128 | 11.0 125 |
SepConv-v1 [127] | 126.2 | 2.32 130 | 2.67 129 | 1.41 128 | 5.68 127 | 7.08 129 | 5.41 127 | 6.88 122 | 7.15 94 | 6.23 129 | 7.42 130 | 8.22 124 | 7.70 130 | 13.5 129 | 14.0 129 | 12.0 129 | 31.6 130 | 8.14 113 | 8.03 129 | 3.93 130 | 2.37 130 | 4.62 130 | 11.8 128 | 16.0 128 | 11.0 125 |
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. |