Optical flow evaluation results |
Statistics:
Average
SD
R0.5
R1.0
R2.0
A50
A75
A95
Error type: endpoint angle interpolation normalized interpolation |
Average 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] | 3.5 | 0.07 1 | 0.20 2 | 0.05 1 | 0.15 2 | 0.51 4 | 0.12 5 | 0.18 2 | 0.37 2 | 0.14 2 | 0.10 2 | 0.49 5 | 0.06 2 | 0.41 1 | 0.61 1 | 0.21 2 | 0.23 2 | 0.66 3 | 0.19 2 | 0.10 6 | 0.12 12 | 0.17 15 | 0.34 2 | 0.80 6 | 0.23 2 |
PMMST [114] | 9.5 | 0.09 32 | 0.21 5 | 0.07 16 | 0.18 11 | 0.51 4 | 0.16 28 | 0.21 9 | 0.42 8 | 0.17 16 | 0.10 2 | 0.33 1 | 0.08 13 | 0.51 5 | 0.74 4 | 0.28 7 | 0.24 3 | 0.65 2 | 0.20 5 | 0.11 20 | 0.12 12 | 0.17 15 | 0.37 4 | 0.74 2 | 0.35 4 |
OFLAF [77] | 10.2 | 0.08 9 | 0.21 5 | 0.06 6 | 0.16 6 | 0.53 6 | 0.12 5 | 0.19 3 | 0.37 2 | 0.14 2 | 0.14 9 | 0.77 30 | 0.07 5 | 0.51 5 | 0.78 7 | 0.25 4 | 0.31 12 | 0.76 4 | 0.25 17 | 0.11 20 | 0.12 12 | 0.21 43 | 0.42 10 | 0.78 4 | 0.63 18 |
MDP-Flow2 [68] | 10.9 | 0.08 9 | 0.21 5 | 0.07 16 | 0.15 2 | 0.48 1 | 0.11 1 | 0.20 5 | 0.40 5 | 0.14 2 | 0.15 23 | 0.80 38 | 0.08 13 | 0.63 19 | 0.93 19 | 0.43 20 | 0.26 6 | 0.76 4 | 0.23 10 | 0.11 20 | 0.12 12 | 0.17 15 | 0.38 6 | 0.79 5 | 0.44 6 |
NN-field [71] | 12.2 | 0.08 9 | 0.22 17 | 0.05 1 | 0.17 8 | 0.55 10 | 0.13 11 | 0.19 3 | 0.39 4 | 0.15 7 | 0.09 1 | 0.48 4 | 0.05 1 | 0.41 1 | 0.61 1 | 0.20 1 | 0.52 61 | 0.64 1 | 0.26 20 | 0.13 44 | 0.13 38 | 0.20 37 | 0.35 3 | 0.83 8 | 0.21 1 |
ComponentFusion [96] | 14.2 | 0.07 1 | 0.21 5 | 0.05 1 | 0.16 6 | 0.55 10 | 0.12 5 | 0.20 5 | 0.44 9 | 0.15 7 | 0.11 4 | 0.65 9 | 0.06 2 | 0.71 35 | 1.07 40 | 0.53 37 | 0.32 16 | 1.06 27 | 0.28 23 | 0.11 20 | 0.13 38 | 0.15 9 | 0.41 9 | 0.88 13 | 0.54 9 |
TC/T-Flow [76] | 19.7 | 0.07 1 | 0.21 5 | 0.05 1 | 0.19 18 | 0.68 34 | 0.12 5 | 0.28 26 | 0.66 31 | 0.14 2 | 0.14 9 | 0.86 48 | 0.07 5 | 0.67 29 | 0.98 29 | 0.49 31 | 0.22 1 | 0.82 9 | 0.19 2 | 0.11 20 | 0.11 2 | 0.30 88 | 0.50 27 | 1.02 30 | 0.64 20 |
WLIF-Flow [93] | 20.2 | 0.08 9 | 0.21 5 | 0.06 6 | 0.18 11 | 0.55 10 | 0.15 23 | 0.25 18 | 0.56 20 | 0.17 16 | 0.14 9 | 0.68 11 | 0.08 13 | 0.61 16 | 0.91 17 | 0.41 18 | 0.43 36 | 0.96 15 | 0.29 29 | 0.13 44 | 0.12 12 | 0.21 43 | 0.51 32 | 1.03 33 | 0.72 39 |
NNF-EAC [103] | 21.7 | 0.09 32 | 0.22 17 | 0.07 16 | 0.17 8 | 0.53 6 | 0.13 11 | 0.23 11 | 0.49 12 | 0.15 7 | 0.16 37 | 0.80 38 | 0.09 28 | 0.60 13 | 0.89 13 | 0.40 16 | 0.38 26 | 0.78 6 | 0.28 23 | 0.12 33 | 0.12 12 | 0.18 29 | 0.57 45 | 1.24 49 | 0.69 33 |
Layers++ [37] | 22.3 | 0.08 9 | 0.21 5 | 0.07 16 | 0.19 18 | 0.56 13 | 0.17 34 | 0.20 5 | 0.40 5 | 0.18 27 | 0.13 8 | 0.58 7 | 0.07 5 | 0.48 3 | 0.70 3 | 0.33 9 | 0.47 48 | 1.01 19 | 0.33 48 | 0.15 68 | 0.14 60 | 0.24 56 | 0.46 17 | 0.88 13 | 0.72 39 |
IROF++ [58] | 23.2 | 0.08 9 | 0.23 24 | 0.07 16 | 0.21 32 | 0.68 34 | 0.17 34 | 0.28 26 | 0.63 24 | 0.19 39 | 0.15 23 | 0.73 23 | 0.09 28 | 0.60 13 | 0.89 13 | 0.42 19 | 0.43 36 | 1.08 30 | 0.31 38 | 0.10 6 | 0.12 12 | 0.12 4 | 0.47 19 | 0.98 23 | 0.68 32 |
LME [70] | 23.3 | 0.08 9 | 0.22 17 | 0.06 6 | 0.15 2 | 0.49 2 | 0.11 1 | 0.30 35 | 0.64 26 | 0.31 90 | 0.15 23 | 0.78 34 | 0.09 28 | 0.66 25 | 0.96 24 | 0.53 37 | 0.33 17 | 1.18 45 | 0.28 23 | 0.12 33 | 0.12 12 | 0.18 29 | 0.44 12 | 0.91 16 | 0.61 14 |
nLayers [57] | 24.2 | 0.07 1 | 0.19 1 | 0.06 6 | 0.22 40 | 0.59 16 | 0.19 57 | 0.25 18 | 0.54 16 | 0.20 48 | 0.15 23 | 0.84 44 | 0.08 13 | 0.53 7 | 0.78 7 | 0.34 11 | 0.44 40 | 0.84 10 | 0.30 34 | 0.13 44 | 0.13 38 | 0.20 37 | 0.47 19 | 0.97 22 | 0.67 30 |
HAST [109] | 25.5 | 0.07 1 | 0.20 2 | 0.05 1 | 0.18 11 | 0.54 8 | 0.13 11 | 0.17 1 | 0.32 1 | 0.12 1 | 0.15 23 | 0.90 59 | 0.06 2 | 0.49 4 | 0.74 4 | 0.22 3 | 0.58 70 | 1.09 31 | 0.44 70 | 0.19 98 | 0.17 89 | 0.47 116 | 0.32 1 | 0.64 1 | 0.33 3 |
PH-Flow [101] | 26.2 | 0.08 9 | 0.24 31 | 0.07 16 | 0.21 32 | 0.68 34 | 0.17 34 | 0.23 11 | 0.49 12 | 0.19 39 | 0.16 37 | 0.83 42 | 0.09 28 | 0.56 9 | 0.83 9 | 0.38 13 | 0.30 10 | 0.81 7 | 0.24 15 | 0.15 68 | 0.13 38 | 0.30 88 | 0.43 11 | 0.85 9 | 0.66 28 |
FC-2Layers-FF [74] | 26.3 | 0.08 9 | 0.21 5 | 0.07 16 | 0.21 32 | 0.70 42 | 0.17 34 | 0.20 5 | 0.40 5 | 0.18 27 | 0.15 23 | 0.76 29 | 0.08 13 | 0.53 7 | 0.77 6 | 0.37 12 | 0.49 54 | 1.02 20 | 0.33 48 | 0.16 79 | 0.13 38 | 0.29 83 | 0.44 12 | 0.87 12 | 0.64 20 |
Correlation Flow [75] | 27.0 | 0.09 32 | 0.23 24 | 0.07 16 | 0.17 8 | 0.58 15 | 0.11 1 | 0.43 67 | 0.99 69 | 0.15 7 | 0.11 4 | 0.47 3 | 0.08 13 | 0.75 41 | 1.08 41 | 0.56 42 | 0.41 32 | 0.92 13 | 0.30 34 | 0.14 55 | 0.13 38 | 0.27 72 | 0.40 8 | 0.85 9 | 0.42 5 |
AGIF+OF [85] | 28.4 | 0.08 9 | 0.22 17 | 0.07 16 | 0.23 54 | 0.73 47 | 0.18 46 | 0.28 26 | 0.66 31 | 0.18 27 | 0.14 9 | 0.70 14 | 0.08 13 | 0.57 10 | 0.85 10 | 0.38 13 | 0.47 48 | 0.97 16 | 0.31 38 | 0.13 44 | 0.13 38 | 0.22 49 | 0.51 32 | 0.99 26 | 0.74 48 |
RNLOD-Flow [121] | 28.5 | 0.07 1 | 0.20 2 | 0.06 6 | 0.19 18 | 0.68 34 | 0.13 11 | 0.33 50 | 0.79 51 | 0.17 16 | 0.14 9 | 0.73 23 | 0.07 5 | 0.69 33 | 1.03 33 | 0.48 27 | 0.37 25 | 0.99 17 | 0.29 29 | 0.16 79 | 0.16 80 | 0.29 83 | 0.45 14 | 0.88 13 | 0.65 25 |
ProbFlowFields [128] | 28.8 | 0.10 47 | 0.31 72 | 0.08 44 | 0.19 18 | 0.63 23 | 0.17 34 | 0.27 22 | 0.63 24 | 0.22 59 | 0.11 4 | 0.49 5 | 0.07 5 | 0.82 51 | 1.22 56 | 0.59 45 | 0.25 4 | 1.05 26 | 0.21 6 | 0.09 4 | 0.12 12 | 0.17 15 | 0.58 47 | 1.33 51 | 0.62 16 |
FESL [72] | 30.4 | 0.08 9 | 0.21 5 | 0.07 16 | 0.25 67 | 0.75 54 | 0.19 57 | 0.27 22 | 0.61 22 | 0.18 27 | 0.14 9 | 0.68 11 | 0.08 13 | 0.61 16 | 0.89 13 | 0.44 21 | 0.47 48 | 1.03 23 | 0.32 43 | 0.14 55 | 0.15 71 | 0.25 62 | 0.50 27 | 0.96 20 | 0.63 18 |
Classic+CPF [83] | 30.4 | 0.08 9 | 0.23 24 | 0.07 16 | 0.22 40 | 0.73 47 | 0.17 34 | 0.30 35 | 0.70 35 | 0.18 27 | 0.14 9 | 0.72 22 | 0.08 13 | 0.63 19 | 0.93 19 | 0.45 24 | 0.51 59 | 1.03 23 | 0.32 43 | 0.14 55 | 0.12 12 | 0.30 88 | 0.48 21 | 0.93 17 | 0.72 39 |
ALD-Flow [66] | 30.5 | 0.07 1 | 0.21 5 | 0.06 6 | 0.19 18 | 0.64 27 | 0.13 11 | 0.30 35 | 0.73 40 | 0.15 7 | 0.17 48 | 0.92 64 | 0.07 5 | 0.78 44 | 1.14 45 | 0.59 45 | 0.33 17 | 1.30 57 | 0.21 6 | 0.12 33 | 0.12 12 | 0.28 78 | 0.54 39 | 1.19 45 | 0.73 44 |
COFM [59] | 31.1 | 0.08 9 | 0.26 45 | 0.06 6 | 0.18 11 | 0.62 21 | 0.14 19 | 0.30 35 | 0.74 42 | 0.19 39 | 0.15 23 | 0.86 48 | 0.07 5 | 0.79 45 | 1.14 45 | 0.74 73 | 0.35 22 | 0.87 12 | 0.28 23 | 0.14 55 | 0.12 12 | 0.28 78 | 0.49 23 | 0.94 18 | 0.71 38 |
Sparse-NonSparse [56] | 31.3 | 0.08 9 | 0.23 24 | 0.07 16 | 0.22 40 | 0.73 47 | 0.18 46 | 0.28 26 | 0.64 26 | 0.19 39 | 0.14 9 | 0.71 19 | 0.08 13 | 0.67 29 | 0.99 32 | 0.48 27 | 0.49 54 | 1.06 27 | 0.32 43 | 0.14 55 | 0.11 2 | 0.28 78 | 0.49 23 | 0.98 23 | 0.73 44 |
TC-Flow [46] | 31.6 | 0.07 1 | 0.21 5 | 0.06 6 | 0.15 2 | 0.59 16 | 0.11 1 | 0.31 42 | 0.78 48 | 0.14 2 | 0.16 37 | 0.86 48 | 0.08 13 | 0.75 41 | 1.11 43 | 0.54 39 | 0.42 34 | 1.40 68 | 0.25 17 | 0.11 20 | 0.12 12 | 0.29 83 | 0.62 55 | 1.35 53 | 0.93 72 |
Efficient-NL [60] | 32.0 | 0.08 9 | 0.22 17 | 0.06 6 | 0.21 32 | 0.67 32 | 0.17 34 | 0.31 42 | 0.73 40 | 0.18 27 | 0.14 9 | 0.71 19 | 0.08 13 | 0.59 12 | 0.88 12 | 0.39 15 | 1.30 100 | 1.35 62 | 0.67 94 | 0.14 55 | 0.13 38 | 0.26 66 | 0.45 14 | 0.85 9 | 0.55 11 |
LSM [39] | 33.0 | 0.08 9 | 0.23 24 | 0.07 16 | 0.22 40 | 0.73 47 | 0.18 46 | 0.28 26 | 0.64 26 | 0.19 39 | 0.14 9 | 0.70 14 | 0.09 28 | 0.66 25 | 0.97 26 | 0.48 27 | 0.50 56 | 1.06 27 | 0.33 48 | 0.15 68 | 0.12 12 | 0.29 83 | 0.50 27 | 0.99 26 | 0.73 44 |
Ramp [62] | 33.6 | 0.08 9 | 0.24 31 | 0.07 16 | 0.21 32 | 0.72 44 | 0.18 46 | 0.27 22 | 0.62 23 | 0.19 39 | 0.15 23 | 0.71 19 | 0.09 28 | 0.66 25 | 0.97 26 | 0.49 31 | 0.51 59 | 1.09 31 | 0.34 54 | 0.15 68 | 0.12 12 | 0.30 88 | 0.48 21 | 0.96 20 | 0.72 39 |
Classic+NL [31] | 36.2 | 0.08 9 | 0.23 24 | 0.07 16 | 0.22 40 | 0.74 52 | 0.18 46 | 0.29 32 | 0.65 30 | 0.19 39 | 0.15 23 | 0.73 23 | 0.09 28 | 0.64 22 | 0.93 19 | 0.47 25 | 0.52 61 | 1.12 37 | 0.33 48 | 0.16 79 | 0.13 38 | 0.29 83 | 0.49 23 | 0.98 23 | 0.74 48 |
OAR-Flow [125] | 36.5 | 0.08 9 | 0.25 38 | 0.07 16 | 0.26 74 | 0.81 71 | 0.18 46 | 0.38 59 | 0.93 62 | 0.20 48 | 0.16 37 | 0.88 54 | 0.08 13 | 0.83 53 | 1.21 53 | 0.61 49 | 0.31 12 | 1.28 53 | 0.18 1 | 0.08 2 | 0.10 1 | 0.17 15 | 0.52 37 | 1.13 40 | 0.69 33 |
TV-L1-MCT [64] | 36.9 | 0.08 9 | 0.23 24 | 0.07 16 | 0.24 61 | 0.77 59 | 0.19 57 | 0.32 46 | 0.76 46 | 0.19 39 | 0.14 9 | 0.69 13 | 0.09 28 | 0.72 37 | 1.03 33 | 0.60 47 | 0.54 64 | 1.10 35 | 0.35 58 | 0.11 20 | 0.12 12 | 0.20 37 | 0.54 39 | 1.04 35 | 0.84 62 |
PMF [73] | 37.5 | 0.09 32 | 0.25 38 | 0.07 16 | 0.19 18 | 0.60 19 | 0.14 19 | 0.23 11 | 0.46 11 | 0.17 16 | 0.17 48 | 0.87 52 | 0.09 28 | 0.58 11 | 0.86 11 | 0.26 5 | 0.82 86 | 1.17 42 | 0.54 83 | 0.21 109 | 0.22 118 | 0.36 104 | 0.39 7 | 0.75 3 | 0.59 13 |
FlowFields+ [130] | 37.5 | 0.11 58 | 0.35 85 | 0.08 44 | 0.23 54 | 0.73 47 | 0.19 57 | 0.30 35 | 0.72 38 | 0.25 72 | 0.14 9 | 0.65 9 | 0.09 28 | 0.85 57 | 1.25 58 | 0.62 51 | 0.25 4 | 1.09 31 | 0.21 6 | 0.10 6 | 0.12 12 | 0.16 12 | 0.59 49 | 1.35 53 | 0.65 25 |
FMOF [94] | 38.8 | 0.08 9 | 0.22 17 | 0.07 16 | 0.24 61 | 0.76 56 | 0.19 57 | 0.24 14 | 0.54 16 | 0.18 27 | 0.14 9 | 0.70 14 | 0.08 13 | 0.64 22 | 0.94 23 | 0.44 21 | 1.19 96 | 1.12 37 | 0.65 93 | 0.15 68 | 0.13 38 | 0.32 98 | 0.58 47 | 1.16 43 | 0.70 37 |
SVFilterOh [111] | 39.5 | 0.10 47 | 0.24 31 | 0.08 44 | 0.21 32 | 0.62 21 | 0.15 23 | 0.24 14 | 0.51 15 | 0.17 16 | 0.16 37 | 0.84 44 | 0.09 28 | 0.61 16 | 0.92 18 | 0.27 6 | 0.81 85 | 1.19 46 | 0.46 75 | 0.21 109 | 0.20 114 | 0.42 111 | 0.37 4 | 0.80 6 | 0.44 6 |
IROF-TV [53] | 40.1 | 0.09 32 | 0.25 38 | 0.08 44 | 0.22 40 | 0.77 59 | 0.19 57 | 0.30 35 | 0.70 35 | 0.19 39 | 0.18 55 | 0.93 67 | 0.11 57 | 0.73 39 | 1.04 36 | 0.56 42 | 0.44 40 | 1.69 91 | 0.31 38 | 0.09 4 | 0.11 2 | 0.12 4 | 0.50 27 | 1.08 38 | 0.73 44 |
S2F-IF [123] | 40.3 | 0.11 58 | 0.35 85 | 0.08 44 | 0.22 40 | 0.75 54 | 0.19 57 | 0.30 35 | 0.72 38 | 0.24 69 | 0.16 37 | 0.79 37 | 0.10 50 | 0.87 60 | 1.28 67 | 0.66 56 | 0.26 6 | 1.09 31 | 0.23 10 | 0.10 6 | 0.12 12 | 0.17 15 | 0.55 42 | 1.19 45 | 0.61 14 |
CombBMOF [113] | 40.9 | 0.10 47 | 0.29 61 | 0.07 16 | 0.22 40 | 0.65 29 | 0.16 28 | 0.25 18 | 0.55 18 | 0.17 16 | 0.16 37 | 0.74 27 | 0.11 57 | 0.67 29 | 0.98 29 | 0.44 21 | 0.60 72 | 1.04 25 | 0.54 83 | 0.17 88 | 0.17 89 | 0.25 62 | 0.51 32 | 1.06 37 | 0.64 20 |
MDP-Flow [26] | 41.3 | 0.09 32 | 0.25 38 | 0.08 44 | 0.19 18 | 0.54 8 | 0.18 46 | 0.24 14 | 0.55 18 | 0.20 48 | 0.16 37 | 0.91 60 | 0.09 28 | 0.74 40 | 1.06 39 | 0.61 49 | 0.46 44 | 1.02 20 | 0.35 58 | 0.12 33 | 0.14 60 | 0.17 15 | 0.78 81 | 1.68 85 | 0.97 77 |
2DHMM-SAS [92] | 43.2 | 0.08 9 | 0.24 31 | 0.07 16 | 0.23 54 | 0.78 62 | 0.17 34 | 0.42 66 | 0.90 59 | 0.22 59 | 0.15 23 | 0.75 28 | 0.09 28 | 0.65 24 | 0.96 24 | 0.48 27 | 0.56 67 | 1.13 40 | 0.34 54 | 0.15 68 | 0.13 38 | 0.30 88 | 0.56 44 | 1.13 40 | 0.79 55 |
EPPM w/o HM [88] | 43.4 | 0.11 58 | 0.30 69 | 0.08 44 | 0.19 18 | 0.67 32 | 0.13 11 | 0.29 32 | 0.71 37 | 0.17 16 | 0.17 48 | 0.78 34 | 0.11 57 | 0.63 19 | 0.93 19 | 0.33 9 | 0.60 72 | 1.35 62 | 0.40 69 | 0.19 98 | 0.15 71 | 0.45 115 | 0.45 14 | 0.94 18 | 0.64 20 |
FlowFields [110] | 43.5 | 0.12 80 | 0.35 85 | 0.08 44 | 0.23 54 | 0.76 56 | 0.20 72 | 0.31 42 | 0.75 43 | 0.25 72 | 0.15 23 | 0.73 23 | 0.10 50 | 0.87 60 | 1.28 67 | 0.66 56 | 0.27 8 | 1.12 37 | 0.23 10 | 0.10 6 | 0.12 12 | 0.17 15 | 0.60 52 | 1.37 56 | 0.64 20 |
MLDP_OF [89] | 43.7 | 0.11 58 | 0.28 54 | 0.09 66 | 0.18 11 | 0.56 13 | 0.13 11 | 0.34 52 | 0.79 51 | 0.17 16 | 0.16 37 | 0.82 41 | 0.09 28 | 0.72 37 | 1.05 38 | 0.50 33 | 0.34 20 | 1.10 35 | 0.27 22 | 0.18 95 | 0.15 71 | 0.44 114 | 0.76 74 | 1.09 39 | 0.69 33 |
Sparse Occlusion [54] | 45.0 | 0.09 32 | 0.24 31 | 0.08 44 | 0.22 40 | 0.63 23 | 0.19 57 | 0.38 59 | 0.91 60 | 0.18 27 | 0.17 48 | 0.85 47 | 0.09 28 | 0.75 41 | 1.09 42 | 0.47 25 | 0.34 20 | 1.00 18 | 0.26 20 | 0.22 113 | 0.22 118 | 0.28 78 | 0.53 38 | 1.13 40 | 0.67 30 |
NL-TV-NCC [25] | 45.5 | 0.10 47 | 0.26 45 | 0.08 44 | 0.22 40 | 0.72 44 | 0.15 23 | 0.35 54 | 0.85 55 | 0.16 12 | 0.15 23 | 0.70 14 | 0.09 28 | 0.79 45 | 1.16 48 | 0.51 34 | 0.78 82 | 1.38 65 | 0.48 77 | 0.16 79 | 0.15 71 | 0.26 66 | 0.55 42 | 1.16 43 | 0.55 11 |
CostFilter [40] | 45.6 | 0.10 47 | 0.27 52 | 0.08 44 | 0.20 30 | 0.63 23 | 0.15 23 | 0.22 10 | 0.45 10 | 0.18 27 | 0.19 61 | 0.88 54 | 0.12 65 | 0.60 13 | 0.90 16 | 0.28 7 | 0.75 81 | 1.19 46 | 0.50 79 | 0.21 109 | 0.24 125 | 0.40 110 | 0.46 17 | 1.02 30 | 0.62 16 |
S2D-Matching [84] | 46.0 | 0.09 32 | 0.26 45 | 0.07 16 | 0.23 54 | 0.80 68 | 0.18 46 | 0.38 59 | 0.93 62 | 0.20 48 | 0.15 23 | 0.70 14 | 0.09 28 | 0.70 34 | 1.03 33 | 0.51 34 | 0.55 66 | 1.17 42 | 0.35 58 | 0.17 88 | 0.13 38 | 0.32 98 | 0.51 32 | 1.01 28 | 0.81 58 |
OFH [38] | 46.2 | 0.10 47 | 0.25 38 | 0.09 66 | 0.19 18 | 0.69 39 | 0.14 19 | 0.43 67 | 1.02 73 | 0.17 16 | 0.17 48 | 1.08 77 | 0.08 13 | 0.87 60 | 1.25 58 | 0.73 70 | 0.43 36 | 1.69 91 | 0.32 43 | 0.10 6 | 0.13 38 | 0.18 29 | 0.59 49 | 1.40 61 | 0.74 48 |
AggregFlow [97] | 48.0 | 0.11 58 | 0.32 77 | 0.08 44 | 0.31 88 | 0.96 91 | 0.23 86 | 0.36 56 | 0.85 55 | 0.27 82 | 0.17 48 | 0.84 44 | 0.10 50 | 0.79 45 | 1.17 49 | 0.54 39 | 0.27 8 | 0.85 11 | 0.19 2 | 0.11 20 | 0.13 38 | 0.15 9 | 0.59 49 | 1.19 45 | 0.83 59 |
SimpleFlow [49] | 49.5 | 0.09 32 | 0.24 31 | 0.08 44 | 0.24 61 | 0.78 62 | 0.20 72 | 0.43 67 | 0.96 66 | 0.21 54 | 0.16 37 | 0.77 30 | 0.09 28 | 0.71 35 | 1.04 36 | 0.55 41 | 1.47 106 | 1.59 82 | 0.76 98 | 0.13 44 | 0.12 12 | 0.22 49 | 0.50 27 | 1.04 35 | 0.72 39 |
Aniso-Texture [82] | 50.6 | 0.08 9 | 0.21 5 | 0.07 16 | 0.19 18 | 0.60 19 | 0.17 34 | 0.50 77 | 1.11 79 | 0.21 54 | 0.12 7 | 0.58 7 | 0.07 5 | 0.93 79 | 1.28 67 | 0.92 85 | 0.46 44 | 1.27 52 | 0.38 68 | 0.20 102 | 0.20 114 | 0.30 88 | 0.68 62 | 1.37 56 | 0.88 67 |
Occlusion-TV-L1 [63] | 51.8 | 0.09 32 | 0.26 45 | 0.07 16 | 0.22 40 | 0.74 52 | 0.18 46 | 0.51 79 | 1.15 84 | 0.21 54 | 0.18 55 | 0.91 60 | 0.10 50 | 0.87 60 | 1.25 58 | 0.72 67 | 0.47 48 | 1.38 65 | 0.36 63 | 0.10 6 | 0.12 12 | 0.11 2 | 0.83 85 | 1.78 89 | 0.96 76 |
PGM-C [120] | 53.4 | 0.12 80 | 0.36 94 | 0.09 66 | 0.25 67 | 0.84 74 | 0.20 72 | 0.32 46 | 0.78 48 | 0.25 72 | 0.20 66 | 1.06 73 | 0.12 65 | 0.88 65 | 1.30 73 | 0.66 56 | 0.31 12 | 1.30 57 | 0.23 10 | 0.10 6 | 0.11 2 | 0.17 15 | 0.61 54 | 1.37 56 | 0.76 52 |
Adaptive [20] | 55.5 | 0.09 32 | 0.26 45 | 0.06 6 | 0.23 54 | 0.78 62 | 0.18 46 | 0.54 83 | 1.19 90 | 0.21 54 | 0.18 55 | 0.91 60 | 0.10 50 | 0.88 65 | 1.25 58 | 0.73 70 | 0.50 56 | 1.28 53 | 0.31 38 | 0.14 55 | 0.16 80 | 0.22 49 | 0.65 59 | 1.37 56 | 0.79 55 |
SRR-TVOF-NL [91] | 56.0 | 0.11 58 | 0.29 61 | 0.08 44 | 0.28 82 | 0.91 85 | 0.20 72 | 0.39 62 | 0.92 61 | 0.24 69 | 0.17 48 | 0.77 30 | 0.09 28 | 0.81 49 | 1.11 43 | 0.79 75 | 0.33 17 | 1.02 20 | 0.28 23 | 0.19 98 | 0.18 99 | 0.31 95 | 0.57 45 | 1.01 28 | 0.77 53 |
DPOF [18] | 56.7 | 0.12 80 | 0.33 80 | 0.08 44 | 0.26 74 | 0.80 68 | 0.20 72 | 0.24 14 | 0.49 12 | 0.20 48 | 0.19 61 | 0.83 42 | 0.13 73 | 0.66 25 | 0.98 29 | 0.40 16 | 1.11 95 | 1.41 70 | 0.57 88 | 0.25 119 | 0.14 60 | 0.55 119 | 0.51 32 | 1.02 30 | 0.54 9 |
ROF-ND [107] | 56.7 | 0.12 80 | 0.29 61 | 0.09 66 | 0.26 74 | 0.72 44 | 0.17 34 | 0.36 56 | 0.86 57 | 0.17 16 | 0.14 9 | 0.46 2 | 0.12 65 | 0.83 53 | 1.18 50 | 0.69 62 | 0.50 56 | 1.15 41 | 0.35 58 | 0.21 109 | 0.17 89 | 0.36 104 | 0.69 66 | 1.40 61 | 0.74 48 |
CPM-Flow [116] | 56.8 | 0.12 80 | 0.36 94 | 0.09 66 | 0.25 67 | 0.85 76 | 0.20 72 | 0.32 46 | 0.77 47 | 0.25 72 | 0.20 66 | 1.06 73 | 0.12 65 | 0.88 65 | 1.30 73 | 0.65 54 | 0.39 29 | 1.22 49 | 0.30 34 | 0.10 6 | 0.12 12 | 0.17 15 | 0.68 62 | 1.52 70 | 0.89 69 |
DeepFlow2 [108] | 56.9 | 0.10 47 | 0.29 61 | 0.09 66 | 0.25 67 | 0.79 67 | 0.19 57 | 0.40 65 | 0.96 66 | 0.23 65 | 0.21 74 | 1.08 77 | 0.12 65 | 0.80 48 | 1.18 50 | 0.62 51 | 0.36 24 | 1.45 73 | 0.24 15 | 0.11 20 | 0.11 2 | 0.24 56 | 0.82 84 | 1.68 85 | 1.00 80 |
ACK-Prior [27] | 57.4 | 0.11 58 | 0.25 38 | 0.09 66 | 0.18 11 | 0.59 16 | 0.13 11 | 0.27 22 | 0.64 26 | 0.16 12 | 0.15 23 | 0.78 34 | 0.09 28 | 0.82 51 | 1.14 45 | 0.71 66 | 1.90 119 | 1.90 100 | 0.99 114 | 0.23 117 | 0.17 89 | 0.49 118 | 0.77 78 | 1.44 65 | 0.91 70 |
TCOF [69] | 57.5 | 0.11 58 | 0.28 54 | 0.09 66 | 0.24 61 | 0.76 56 | 0.19 57 | 0.53 80 | 1.15 84 | 0.29 86 | 0.24 81 | 0.88 54 | 0.20 96 | 0.88 65 | 1.26 63 | 0.69 62 | 0.38 26 | 0.93 14 | 0.29 29 | 0.16 79 | 0.16 80 | 0.22 49 | 0.49 23 | 1.03 33 | 0.65 25 |
Kuang [131] | 57.5 | 0.11 58 | 0.35 85 | 0.08 44 | 0.24 61 | 0.85 76 | 0.18 46 | 0.33 50 | 0.84 54 | 0.23 65 | 0.18 55 | 0.91 60 | 0.11 57 | 0.90 72 | 1.33 85 | 0.66 56 | 0.69 78 | 1.36 64 | 0.50 79 | 0.10 6 | 0.12 12 | 0.17 15 | 0.65 59 | 1.38 60 | 1.05 84 |
RFlow [90] | 58.1 | 0.10 47 | 0.27 52 | 0.09 66 | 0.19 18 | 0.64 27 | 0.15 23 | 0.46 75 | 1.07 74 | 0.18 27 | 0.22 79 | 1.31 94 | 0.11 57 | 0.92 77 | 1.30 73 | 0.91 83 | 0.42 34 | 1.42 71 | 0.31 38 | 0.14 55 | 0.13 38 | 0.24 56 | 0.77 78 | 1.66 80 | 0.94 73 |
Complementary OF [21] | 58.4 | 0.11 58 | 0.28 54 | 0.10 85 | 0.18 11 | 0.63 23 | 0.12 5 | 0.31 42 | 0.75 43 | 0.18 27 | 0.19 61 | 0.97 68 | 0.12 65 | 0.97 86 | 1.31 80 | 1.00 92 | 1.78 118 | 1.73 94 | 0.87 106 | 0.11 20 | 0.12 12 | 0.22 49 | 0.68 62 | 1.48 66 | 0.95 74 |
EpicFlow [102] | 60.6 | 0.12 80 | 0.36 94 | 0.09 66 | 0.25 67 | 0.85 76 | 0.21 80 | 0.39 62 | 1.00 70 | 0.25 72 | 0.19 61 | 1.01 70 | 0.11 57 | 0.89 69 | 1.31 80 | 0.69 62 | 0.53 63 | 1.31 59 | 0.34 54 | 0.10 6 | 0.11 2 | 0.17 15 | 0.67 61 | 1.43 64 | 0.87 65 |
ComplOF-FED-GPU [35] | 61.5 | 0.11 58 | 0.29 61 | 0.10 85 | 0.21 32 | 0.78 62 | 0.14 19 | 0.32 46 | 0.79 51 | 0.17 16 | 0.19 61 | 0.99 69 | 0.11 57 | 0.89 69 | 1.29 70 | 0.73 70 | 1.25 98 | 1.74 95 | 0.64 92 | 0.14 55 | 0.13 38 | 0.30 88 | 0.64 57 | 1.50 68 | 0.83 59 |
Steered-L1 [118] | 62.0 | 0.09 32 | 0.22 17 | 0.08 44 | 0.14 1 | 0.49 2 | 0.12 5 | 0.28 26 | 0.69 34 | 0.16 12 | 0.18 55 | 1.06 73 | 0.09 28 | 0.89 69 | 1.24 57 | 0.91 83 | 1.71 115 | 1.68 89 | 0.94 109 | 0.26 121 | 0.18 99 | 0.71 123 | 1.06 98 | 1.80 91 | 1.64 105 |
Classic++ [32] | 62.3 | 0.09 32 | 0.25 38 | 0.07 16 | 0.23 54 | 0.78 62 | 0.19 57 | 0.43 67 | 1.00 70 | 0.22 59 | 0.20 66 | 1.11 79 | 0.10 50 | 0.87 60 | 1.30 73 | 0.66 56 | 0.47 48 | 1.62 83 | 0.33 48 | 0.17 88 | 0.14 60 | 0.32 98 | 0.79 82 | 1.64 78 | 0.92 71 |
HBM-GC [105] | 62.4 | 0.14 97 | 0.28 54 | 0.12 98 | 0.26 74 | 0.69 39 | 0.22 83 | 0.34 52 | 0.75 43 | 0.22 59 | 0.21 74 | 0.77 30 | 0.15 82 | 0.67 29 | 0.97 26 | 0.52 36 | 0.63 76 | 0.81 7 | 0.44 70 | 0.22 113 | 0.19 111 | 0.36 104 | 0.54 39 | 1.21 48 | 0.78 54 |
Aniso. Huber-L1 [22] | 64.0 | 0.10 47 | 0.28 54 | 0.08 44 | 0.31 88 | 0.88 82 | 0.28 93 | 0.56 86 | 1.13 80 | 0.29 86 | 0.20 66 | 0.92 64 | 0.13 73 | 0.84 56 | 1.20 52 | 0.70 65 | 0.39 29 | 1.23 50 | 0.28 23 | 0.17 88 | 0.15 71 | 0.27 72 | 0.64 57 | 1.36 55 | 0.79 55 |
TF+OM [100] | 64.8 | 0.10 47 | 0.26 45 | 0.07 16 | 0.22 40 | 0.66 31 | 0.19 57 | 0.36 56 | 0.78 48 | 0.39 93 | 0.20 66 | 0.89 57 | 0.13 73 | 0.98 90 | 1.31 80 | 1.03 93 | 0.56 67 | 1.55 80 | 0.33 48 | 0.16 79 | 0.17 89 | 0.27 72 | 0.76 74 | 1.59 76 | 0.98 78 |
CRTflow [80] | 66.1 | 0.11 58 | 0.30 69 | 0.08 44 | 0.24 61 | 0.77 59 | 0.17 34 | 0.50 77 | 1.13 80 | 0.21 54 | 0.23 80 | 1.24 88 | 0.12 65 | 0.86 59 | 1.27 65 | 0.65 54 | 0.60 72 | 1.95 105 | 0.50 79 | 0.12 33 | 0.14 60 | 0.21 43 | 0.79 82 | 1.77 88 | 0.98 78 |
DeepFlow [86] | 66.9 | 0.12 80 | 0.31 72 | 0.11 92 | 0.28 82 | 0.82 72 | 0.22 83 | 0.44 73 | 1.00 70 | 0.33 91 | 0.26 87 | 1.34 97 | 0.15 82 | 0.81 49 | 1.21 53 | 0.58 44 | 0.38 26 | 1.55 80 | 0.25 17 | 0.11 20 | 0.11 2 | 0.24 56 | 0.93 93 | 1.82 94 | 1.12 91 |
TriangleFlow [30] | 67.6 | 0.11 58 | 0.29 61 | 0.09 66 | 0.26 74 | 0.95 89 | 0.17 34 | 0.47 76 | 1.07 74 | 0.18 27 | 0.16 37 | 0.87 52 | 0.09 28 | 1.07 98 | 1.47 105 | 1.10 99 | 0.87 87 | 1.39 67 | 0.57 88 | 0.15 68 | 0.19 111 | 0.23 55 | 0.63 56 | 1.33 51 | 0.84 62 |
BriefMatch [124] | 68.4 | 0.09 32 | 0.24 31 | 0.07 16 | 0.21 32 | 0.68 34 | 0.16 28 | 0.25 18 | 0.59 21 | 0.16 12 | 0.20 66 | 1.11 79 | 0.10 50 | 0.93 79 | 1.29 70 | 0.98 87 | 1.69 114 | 1.63 85 | 1.07 118 | 0.25 119 | 0.18 99 | 0.73 124 | 1.25 109 | 1.94 100 | 2.15 118 |
TV-L1-improved [17] | 69.2 | 0.09 32 | 0.26 45 | 0.07 16 | 0.20 30 | 0.71 43 | 0.16 28 | 0.53 80 | 1.18 89 | 0.22 59 | 0.21 74 | 1.24 88 | 0.11 57 | 0.90 72 | 1.31 80 | 0.72 67 | 1.51 108 | 1.93 102 | 0.84 102 | 0.18 95 | 0.17 89 | 0.31 95 | 0.73 69 | 1.62 77 | 0.87 65 |
SIOF [67] | 71.6 | 0.11 58 | 0.28 54 | 0.09 66 | 0.27 80 | 0.95 89 | 0.20 72 | 0.60 95 | 1.17 86 | 0.48 96 | 0.25 85 | 1.13 81 | 0.16 84 | 0.97 86 | 1.33 85 | 1.03 93 | 0.43 36 | 1.32 60 | 0.36 63 | 0.13 44 | 0.13 38 | 0.18 29 | 0.76 74 | 1.52 70 | 1.14 95 |
LocallyOriented [52] | 73.1 | 0.12 80 | 0.35 85 | 0.08 44 | 0.33 93 | 1.01 94 | 0.25 89 | 0.61 98 | 1.30 99 | 0.28 83 | 0.18 55 | 0.80 38 | 0.13 73 | 0.93 79 | 1.29 70 | 0.79 75 | 0.98 91 | 1.48 76 | 0.56 87 | 0.12 33 | 0.14 60 | 0.21 43 | 0.73 69 | 1.48 66 | 0.95 74 |
CBF [12] | 73.5 | 0.10 47 | 0.28 54 | 0.09 66 | 0.34 94 | 0.80 68 | 0.37 98 | 0.43 67 | 0.95 65 | 0.26 78 | 0.21 74 | 1.14 82 | 0.13 73 | 0.90 72 | 1.27 65 | 0.82 78 | 0.41 32 | 1.23 50 | 0.30 34 | 0.23 117 | 0.19 111 | 0.39 109 | 0.76 74 | 1.56 72 | 1.02 83 |
DF-Auto [115] | 74.1 | 0.13 93 | 0.36 94 | 0.08 44 | 0.45 101 | 1.03 96 | 0.41 101 | 0.56 86 | 1.13 80 | 0.58 100 | 0.26 87 | 1.17 84 | 0.17 88 | 0.96 83 | 1.30 73 | 1.03 93 | 0.30 10 | 1.17 42 | 0.23 10 | 0.14 55 | 0.18 99 | 0.13 8 | 0.85 86 | 1.66 80 | 1.06 85 |
Local-TV-L1 [65] | 76.0 | 0.14 97 | 0.34 82 | 0.14 102 | 0.47 102 | 1.05 98 | 0.43 102 | 0.72 104 | 1.25 95 | 0.52 97 | 0.31 100 | 1.39 101 | 0.22 98 | 0.83 53 | 1.21 53 | 0.63 53 | 0.39 29 | 1.29 55 | 0.29 29 | 0.11 20 | 0.11 2 | 0.22 49 | 1.06 98 | 1.87 97 | 1.67 107 |
CLG-TV [48] | 76.0 | 0.11 58 | 0.29 61 | 0.09 66 | 0.32 91 | 0.86 81 | 0.30 94 | 0.55 84 | 1.17 86 | 0.28 83 | 0.25 85 | 1.05 72 | 0.17 88 | 0.92 77 | 1.30 73 | 0.79 75 | 0.47 48 | 1.72 93 | 0.35 58 | 0.17 88 | 0.17 89 | 0.25 62 | 0.74 72 | 1.57 74 | 0.88 67 |
Brox et al. [5] | 76.2 | 0.11 58 | 0.32 77 | 0.11 92 | 0.27 80 | 0.93 87 | 0.22 83 | 0.39 62 | 0.94 64 | 0.24 69 | 0.24 81 | 1.25 90 | 0.13 73 | 1.10 105 | 1.39 98 | 1.43 114 | 0.89 89 | 1.77 97 | 0.55 86 | 0.10 6 | 0.13 38 | 0.11 2 | 0.91 90 | 1.83 96 | 1.13 93 |
F-TV-L1 [15] | 76.7 | 0.14 97 | 0.35 85 | 0.14 102 | 0.34 94 | 0.98 92 | 0.26 91 | 0.59 93 | 1.19 90 | 0.26 78 | 0.27 92 | 1.36 99 | 0.16 84 | 0.90 72 | 1.30 73 | 0.76 74 | 0.54 64 | 1.62 83 | 0.36 63 | 0.13 44 | 0.15 71 | 0.20 37 | 0.68 62 | 1.56 72 | 0.66 28 |
Fusion [6] | 77.0 | 0.11 58 | 0.34 82 | 0.10 85 | 0.19 18 | 0.69 39 | 0.16 28 | 0.29 32 | 0.66 31 | 0.23 65 | 0.20 66 | 1.19 85 | 0.14 80 | 1.07 98 | 1.42 101 | 1.22 105 | 1.35 101 | 1.49 77 | 0.86 104 | 0.20 102 | 0.20 114 | 0.26 66 | 1.07 101 | 2.07 108 | 1.39 102 |
Rannacher [23] | 77.9 | 0.11 58 | 0.31 72 | 0.09 66 | 0.25 67 | 0.84 74 | 0.21 80 | 0.57 90 | 1.27 97 | 0.26 78 | 0.24 81 | 1.32 95 | 0.13 73 | 0.91 76 | 1.33 85 | 0.72 67 | 1.49 107 | 1.95 105 | 0.78 99 | 0.15 68 | 0.14 60 | 0.26 66 | 0.69 66 | 1.58 75 | 0.86 64 |
SuperFlow [81] | 78.2 | 0.11 58 | 0.29 61 | 0.08 44 | 0.34 94 | 0.85 76 | 0.33 96 | 0.53 80 | 1.08 77 | 0.59 101 | 0.28 95 | 1.23 87 | 0.21 97 | 0.99 91 | 1.32 84 | 1.21 104 | 0.46 44 | 1.49 77 | 0.36 63 | 0.15 68 | 0.16 80 | 0.19 33 | 0.90 89 | 1.81 92 | 1.07 87 |
Second-order prior [8] | 80.2 | 0.11 58 | 0.31 72 | 0.09 66 | 0.26 74 | 0.93 87 | 0.20 72 | 0.57 90 | 1.25 95 | 0.26 78 | 0.20 66 | 1.04 71 | 0.12 65 | 0.94 82 | 1.34 88 | 0.83 80 | 0.61 75 | 1.93 102 | 0.47 76 | 0.20 102 | 0.16 80 | 0.34 102 | 0.77 78 | 1.64 78 | 1.07 87 |
TriFlow [95] | 80.5 | 0.12 80 | 0.33 80 | 0.09 66 | 0.30 86 | 0.89 83 | 0.27 92 | 0.56 86 | 1.17 86 | 0.57 99 | 0.21 74 | 0.92 64 | 0.16 84 | 1.07 98 | 1.38 92 | 1.19 103 | 0.35 22 | 1.19 46 | 0.29 29 | 0.52 127 | 0.22 118 | 1.30 127 | 0.73 69 | 1.42 63 | 0.83 59 |
p-harmonic [29] | 82.0 | 0.12 80 | 0.36 94 | 0.11 92 | 0.25 67 | 0.82 72 | 0.21 80 | 0.57 90 | 1.24 92 | 0.28 83 | 0.26 87 | 1.20 86 | 0.19 95 | 1.07 98 | 1.39 98 | 1.31 109 | 0.44 40 | 1.65 87 | 0.37 67 | 0.15 68 | 0.16 80 | 0.21 43 | 0.87 87 | 1.76 87 | 1.06 85 |
Bartels [41] | 82.2 | 0.12 80 | 0.30 69 | 0.11 92 | 0.22 40 | 0.65 29 | 0.19 57 | 0.35 54 | 0.86 57 | 0.23 65 | 0.28 95 | 1.32 95 | 0.18 92 | 0.97 86 | 1.38 92 | 0.98 87 | 1.20 97 | 1.76 96 | 0.78 99 | 0.20 102 | 0.17 89 | 0.48 117 | 0.91 90 | 1.88 98 | 1.22 96 |
Dynamic MRF [7] | 83.3 | 0.12 80 | 0.34 82 | 0.11 92 | 0.22 40 | 0.89 83 | 0.16 28 | 0.44 73 | 1.13 80 | 0.20 48 | 0.24 81 | 1.29 93 | 0.14 80 | 1.11 106 | 1.52 112 | 1.13 101 | 1.54 109 | 2.37 118 | 0.93 107 | 0.13 44 | 0.12 12 | 0.31 95 | 1.27 111 | 2.33 118 | 1.66 106 |
FlowNetS+ft+v [112] | 84.4 | 0.11 58 | 0.31 72 | 0.09 66 | 0.30 86 | 0.91 85 | 0.25 89 | 0.62 101 | 1.27 97 | 0.44 95 | 0.27 92 | 1.26 91 | 0.18 92 | 1.04 97 | 1.38 92 | 1.10 99 | 0.46 44 | 1.66 88 | 0.34 54 | 0.17 88 | 0.18 99 | 0.35 103 | 0.75 73 | 1.67 84 | 1.00 80 |
SegOF [10] | 84.7 | 0.15 101 | 0.36 94 | 0.10 85 | 0.57 105 | 1.16 104 | 0.59 110 | 0.68 103 | 1.24 92 | 0.64 103 | 0.32 101 | 0.86 48 | 0.26 101 | 1.18 112 | 1.50 111 | 1.47 116 | 1.63 113 | 2.09 109 | 0.96 111 | 0.08 2 | 0.13 38 | 0.12 4 | 0.70 68 | 1.50 68 | 0.69 33 |
FlowNet2 [122] | 84.8 | 0.22 112 | 0.50 113 | 0.17 106 | 0.67 112 | 1.32 109 | 0.61 112 | 0.61 98 | 1.08 77 | 0.69 105 | 0.28 95 | 0.89 57 | 0.22 98 | 0.97 86 | 1.38 92 | 0.68 61 | 0.59 71 | 1.29 55 | 0.50 79 | 0.19 98 | 0.22 118 | 0.27 72 | 0.60 52 | 1.32 50 | 0.51 8 |
CNN-flow-warp+ref [117] | 86.1 | 0.13 93 | 0.38 100 | 0.11 92 | 0.31 88 | 0.85 76 | 0.30 94 | 0.59 93 | 1.31 101 | 0.41 94 | 0.28 95 | 1.39 101 | 0.17 88 | 1.09 104 | 1.42 101 | 1.33 110 | 0.80 83 | 1.94 104 | 0.48 77 | 0.10 6 | 0.12 12 | 0.17 15 | 1.35 114 | 2.18 115 | 1.72 110 |
LDOF [28] | 87.9 | 0.12 80 | 0.35 85 | 0.10 85 | 0.32 91 | 1.06 99 | 0.24 88 | 0.43 67 | 0.98 68 | 0.30 89 | 0.45 106 | 2.48 124 | 0.26 101 | 1.01 94 | 1.37 91 | 1.05 97 | 1.10 94 | 2.08 108 | 0.67 94 | 0.12 33 | 0.15 71 | 0.24 56 | 0.94 94 | 2.05 105 | 1.10 89 |
Ad-TV-NDC [36] | 88.2 | 0.23 115 | 0.40 105 | 0.31 122 | 0.92 120 | 1.42 114 | 0.93 119 | 1.05 115 | 1.60 112 | 0.74 110 | 0.48 107 | 1.27 92 | 0.49 110 | 0.85 57 | 1.25 58 | 0.60 47 | 0.44 40 | 1.47 74 | 0.32 43 | 0.12 33 | 0.13 38 | 0.19 33 | 1.59 120 | 2.06 107 | 2.87 125 |
StereoFlow [44] | 91.1 | 0.46 129 | 0.77 128 | 0.47 126 | 1.41 125 | 2.26 128 | 1.16 122 | 1.30 125 | 1.94 124 | 1.02 122 | 1.33 124 | 2.98 126 | 1.16 123 | 1.08 102 | 1.49 108 | 0.99 89 | 0.31 12 | 1.40 68 | 0.22 9 | 0.07 1 | 0.11 2 | 0.08 1 | 0.98 96 | 1.88 98 | 1.31 99 |
Shiralkar [42] | 92.7 | 0.13 93 | 0.39 102 | 0.10 85 | 0.28 82 | 1.08 100 | 0.19 57 | 0.61 98 | 1.33 104 | 0.25 72 | 0.27 92 | 1.35 98 | 0.18 92 | 1.01 94 | 1.47 105 | 0.90 82 | 0.88 88 | 2.04 107 | 0.54 83 | 0.20 102 | 0.16 80 | 0.42 111 | 1.04 97 | 2.13 112 | 1.10 89 |
Learning Flow [11] | 93.1 | 0.11 58 | 0.32 77 | 0.09 66 | 0.29 85 | 0.99 93 | 0.23 86 | 0.55 84 | 1.24 92 | 0.29 86 | 0.36 102 | 1.56 108 | 0.25 100 | 1.25 117 | 1.64 118 | 1.41 112 | 1.55 111 | 2.32 117 | 0.85 103 | 0.14 55 | 0.18 99 | 0.24 56 | 1.09 102 | 2.09 110 | 1.27 97 |
StereoOF-V1MT [119] | 93.9 | 0.13 93 | 0.40 105 | 0.10 85 | 0.34 94 | 1.33 111 | 0.19 57 | 0.60 95 | 1.42 106 | 0.22 59 | 0.26 87 | 1.38 100 | 0.16 84 | 1.21 114 | 1.65 119 | 1.22 105 | 1.60 112 | 2.42 119 | 0.93 107 | 0.12 33 | 0.14 60 | 0.25 62 | 1.46 118 | 2.52 120 | 1.70 109 |
IAOF2 [51] | 96.1 | 0.14 97 | 0.35 85 | 0.12 98 | 0.42 99 | 1.09 102 | 0.38 99 | 0.64 102 | 1.32 103 | 0.55 98 | 0.92 116 | 1.60 110 | 1.04 118 | 1.00 93 | 1.38 92 | 0.94 86 | 0.80 83 | 1.43 72 | 0.58 90 | 0.20 102 | 0.18 99 | 0.32 98 | 0.92 92 | 1.66 80 | 1.13 93 |
Filter Flow [19] | 97.8 | 0.17 103 | 0.39 102 | 0.13 100 | 0.43 100 | 1.09 102 | 0.38 99 | 0.75 105 | 1.34 105 | 0.78 113 | 0.70 114 | 1.54 107 | 0.68 113 | 1.13 109 | 1.38 92 | 1.51 117 | 0.57 69 | 1.32 60 | 0.44 70 | 0.22 113 | 0.23 123 | 0.26 66 | 0.96 95 | 1.66 80 | 1.12 91 |
Modified CLG [34] | 99.2 | 0.19 109 | 0.46 111 | 0.17 106 | 0.49 104 | 1.08 100 | 0.51 106 | 0.93 109 | 1.59 110 | 0.82 115 | 0.49 108 | 1.65 113 | 0.42 107 | 1.14 110 | 1.48 107 | 1.42 113 | 1.06 93 | 2.16 113 | 0.68 96 | 0.12 33 | 0.14 60 | 0.20 37 | 1.12 105 | 2.17 114 | 1.52 103 |
GraphCuts [14] | 99.4 | 0.16 102 | 0.38 100 | 0.14 102 | 0.59 108 | 1.36 113 | 0.46 103 | 0.56 86 | 1.07 74 | 0.64 103 | 0.26 87 | 1.14 82 | 0.17 88 | 0.96 83 | 1.35 89 | 0.84 81 | 2.25 126 | 1.79 98 | 1.22 121 | 0.22 113 | 0.17 89 | 0.43 113 | 1.22 108 | 2.05 105 | 1.78 112 |
GroupFlow [9] | 99.5 | 0.21 110 | 0.51 114 | 0.21 113 | 0.79 116 | 1.69 119 | 0.72 117 | 0.86 108 | 1.64 113 | 0.74 110 | 0.30 99 | 1.07 76 | 0.26 101 | 1.29 120 | 1.81 122 | 0.82 78 | 1.94 121 | 2.30 116 | 1.36 122 | 0.11 20 | 0.14 60 | 0.19 33 | 1.06 98 | 1.96 101 | 1.35 101 |
IAOF [50] | 99.9 | 0.17 103 | 0.39 102 | 0.18 109 | 0.61 109 | 1.23 106 | 0.55 109 | 1.20 119 | 1.87 122 | 0.73 108 | 0.66 113 | 1.46 103 | 0.72 114 | 0.99 91 | 1.36 90 | 0.99 89 | 0.73 80 | 1.83 99 | 0.45 73 | 0.18 95 | 0.15 71 | 0.27 72 | 1.30 112 | 1.81 92 | 2.09 117 |
Black & Anandan [4] | 100.0 | 0.18 107 | 0.42 108 | 0.19 110 | 0.58 107 | 1.31 108 | 0.50 105 | 0.95 111 | 1.58 109 | 0.70 106 | 0.49 108 | 1.59 109 | 0.45 108 | 1.08 102 | 1.42 101 | 1.22 105 | 1.43 104 | 2.28 115 | 0.83 101 | 0.15 68 | 0.17 89 | 0.17 15 | 1.11 103 | 1.98 102 | 1.30 98 |
SPSA-learn [13] | 100.4 | 0.18 107 | 0.45 110 | 0.17 106 | 0.57 105 | 1.32 109 | 0.51 106 | 0.84 107 | 1.50 107 | 0.72 107 | 0.52 110 | 1.64 112 | 0.49 110 | 1.12 108 | 1.42 101 | 1.39 111 | 1.75 117 | 2.14 111 | 1.06 117 | 0.13 44 | 0.13 38 | 0.19 33 | 1.32 113 | 2.08 109 | 1.73 111 |
BlockOverlap [61] | 101.2 | 0.17 103 | 0.35 85 | 0.16 105 | 0.48 103 | 1.02 95 | 0.46 103 | 0.75 105 | 1.31 101 | 0.59 101 | 0.40 105 | 1.47 104 | 0.33 106 | 0.96 83 | 1.26 63 | 1.14 102 | 1.40 103 | 1.47 74 | 0.86 104 | 0.31 124 | 0.22 118 | 0.86 126 | 1.20 107 | 1.78 89 | 2.19 119 |
UnFlow [129] | 101.7 | 0.38 125 | 0.70 124 | 0.25 119 | 0.76 114 | 1.46 115 | 0.70 115 | 0.98 113 | 1.75 116 | 0.73 108 | 0.55 111 | 1.52 106 | 0.48 109 | 1.47 123 | 1.83 124 | 1.61 120 | 0.91 90 | 2.19 114 | 0.72 97 | 0.13 44 | 0.16 80 | 0.12 4 | 0.87 87 | 2.03 103 | 1.00 80 |
HBpMotionGpu [43] | 101.8 | 0.17 103 | 0.41 107 | 0.13 100 | 0.61 109 | 1.34 112 | 0.59 110 | 0.95 111 | 1.68 114 | 0.76 112 | 0.38 103 | 1.63 111 | 0.27 104 | 1.11 106 | 1.49 108 | 1.27 108 | 0.66 77 | 1.53 79 | 0.45 73 | 0.20 102 | 0.18 99 | 0.28 78 | 1.12 105 | 2.04 104 | 1.67 107 |
2D-CLG [1] | 102.0 | 0.28 117 | 0.62 120 | 0.21 113 | 0.67 112 | 1.21 105 | 0.70 115 | 1.12 116 | 1.80 119 | 0.99 121 | 1.07 120 | 2.06 118 | 1.12 122 | 1.23 116 | 1.52 112 | 1.62 121 | 1.54 109 | 2.15 112 | 0.96 111 | 0.10 6 | 0.11 2 | 0.16 12 | 1.38 117 | 2.26 117 | 1.83 114 |
2bit-BM-tele [98] | 104.0 | 0.21 110 | 0.42 108 | 0.23 117 | 0.39 98 | 1.04 97 | 0.35 97 | 0.60 95 | 1.30 99 | 0.36 92 | 0.38 103 | 1.49 105 | 0.30 105 | 1.01 94 | 1.41 100 | 0.99 89 | 1.39 102 | 1.68 89 | 0.95 110 | 0.31 124 | 0.23 123 | 0.70 122 | 1.11 103 | 2.09 110 | 1.61 104 |
Nguyen [33] | 104.0 | 0.22 112 | 0.47 112 | 0.19 110 | 0.87 118 | 1.29 107 | 0.97 120 | 1.17 118 | 1.81 120 | 0.92 119 | 0.99 118 | 1.82 114 | 1.07 119 | 1.17 111 | 1.49 108 | 1.46 115 | 0.72 79 | 2.09 109 | 0.60 91 | 0.14 55 | 0.14 60 | 0.20 37 | 1.37 115 | 2.18 115 | 1.86 115 |
Horn & Schunck [3] | 108.3 | 0.22 112 | 0.55 115 | 0.22 115 | 0.61 109 | 1.53 117 | 0.52 108 | 1.01 114 | 1.73 115 | 0.80 114 | 0.78 115 | 2.02 116 | 0.77 115 | 1.26 118 | 1.58 116 | 1.55 118 | 1.43 104 | 2.59 122 | 1.00 115 | 0.16 79 | 0.18 99 | 0.15 9 | 1.51 119 | 2.50 119 | 1.88 116 |
TI-DOFE [24] | 111.6 | 0.38 125 | 0.64 121 | 0.47 126 | 1.16 123 | 1.72 120 | 1.26 125 | 1.39 127 | 2.06 129 | 1.17 124 | 1.29 123 | 2.21 120 | 1.41 126 | 1.27 119 | 1.61 117 | 1.57 119 | 1.28 99 | 2.57 121 | 1.01 116 | 0.13 44 | 0.15 71 | 0.16 12 | 1.87 124 | 2.71 124 | 2.53 123 |
SILK [79] | 112.7 | 0.25 116 | 0.55 115 | 0.29 120 | 0.77 115 | 1.49 116 | 0.79 118 | 1.14 117 | 1.83 121 | 0.84 116 | 0.59 112 | 1.82 114 | 0.55 112 | 1.36 121 | 1.69 120 | 1.82 123 | 1.92 120 | 2.65 123 | 1.15 120 | 0.16 79 | 0.13 38 | 0.36 104 | 1.69 121 | 2.54 121 | 2.30 122 |
Heeger++ [104] | 114.2 | 0.34 122 | 0.61 118 | 0.22 115 | 0.89 119 | 2.04 127 | 0.65 114 | 1.20 119 | 1.77 117 | 0.91 118 | 1.08 121 | 2.24 121 | 0.99 117 | 1.67 126 | 1.96 126 | 1.99 124 | 2.17 125 | 3.02 126 | 1.62 126 | 0.14 55 | 0.18 99 | 0.21 43 | 1.81 122 | 2.64 122 | 2.27 120 |
HCIC-L [99] | 116.0 | 0.43 128 | 0.64 121 | 0.29 120 | 1.90 129 | 1.89 123 | 2.31 129 | 1.20 119 | 1.51 108 | 1.44 128 | 1.49 127 | 2.58 125 | 1.55 127 | 1.21 114 | 1.52 112 | 1.03 93 | 1.01 92 | 1.63 85 | 0.98 113 | 0.83 129 | 0.55 129 | 1.52 129 | 1.26 110 | 1.82 94 | 1.34 100 |
Adaptive flow [45] | 118.0 | 0.36 123 | 0.59 117 | 0.37 125 | 1.21 124 | 1.60 118 | 1.23 124 | 1.21 122 | 1.77 117 | 1.18 125 | 0.94 117 | 2.03 117 | 0.97 116 | 1.20 113 | 1.57 115 | 1.08 98 | 1.73 116 | 1.90 100 | 1.12 119 | 0.59 128 | 0.37 128 | 1.37 128 | 1.37 115 | 2.16 113 | 1.81 113 |
SLK [47] | 118.3 | 0.30 119 | 0.70 124 | 0.36 124 | 1.09 122 | 1.77 121 | 1.21 123 | 1.25 124 | 1.98 126 | 1.03 123 | 1.56 128 | 2.26 122 | 1.71 128 | 1.54 125 | 1.82 123 | 2.14 126 | 2.02 122 | 2.79 125 | 1.36 122 | 0.17 88 | 0.16 80 | 0.26 66 | 2.43 126 | 3.18 126 | 3.31 127 |
FFV1MT [106] | 118.4 | 0.33 121 | 0.64 121 | 0.24 118 | 0.79 116 | 1.90 125 | 0.64 113 | 1.33 126 | 1.90 123 | 1.23 126 | 1.38 125 | 2.98 126 | 1.29 124 | 1.76 127 | 1.99 127 | 2.45 127 | 2.33 127 | 3.64 128 | 1.72 127 | 0.16 79 | 0.18 99 | 0.27 72 | 1.81 122 | 2.64 122 | 2.27 120 |
Periodicity [78] | 120.4 | 0.31 120 | 0.78 129 | 0.20 112 | 1.54 127 | 2.62 129 | 1.71 126 | 1.86 129 | 2.00 127 | 1.66 129 | 1.15 122 | 3.05 128 | 1.07 119 | 5.17 129 | 6.79 129 | 4.19 129 | 3.79 129 | 5.26 131 | 2.93 129 | 0.12 33 | 0.18 99 | 0.36 104 | 2.67 127 | 5.01 128 | 3.18 126 |
PGAM+LK [55] | 122.4 | 0.37 124 | 0.70 124 | 0.59 128 | 1.08 121 | 1.89 123 | 1.15 121 | 0.94 110 | 1.59 110 | 0.88 117 | 1.40 126 | 3.28 129 | 1.33 125 | 1.37 122 | 1.70 121 | 1.67 122 | 2.10 123 | 2.53 120 | 1.39 124 | 0.36 126 | 0.28 127 | 0.65 120 | 1.89 125 | 2.72 125 | 2.71 124 |
FOLKI [16] | 123.6 | 0.29 118 | 0.73 127 | 0.33 123 | 1.52 126 | 1.96 126 | 1.80 127 | 1.23 123 | 2.04 128 | 0.95 120 | 0.99 118 | 2.20 119 | 1.08 121 | 1.53 124 | 1.85 125 | 2.07 125 | 2.14 124 | 3.23 127 | 1.60 125 | 0.26 121 | 0.21 117 | 0.68 121 | 2.67 127 | 3.27 127 | 4.32 128 |
Pyramid LK [2] | 126.5 | 0.39 127 | 0.61 118 | 0.61 129 | 1.67 128 | 1.78 122 | 2.00 128 | 1.50 128 | 1.97 125 | 1.38 127 | 1.57 129 | 2.39 123 | 1.78 129 | 2.94 128 | 3.72 128 | 2.98 128 | 3.33 128 | 2.74 124 | 2.43 128 | 0.30 123 | 0.24 125 | 0.73 124 | 3.80 129 | 5.08 129 | 4.88 129 |
AdaConv-v1 [126] | 130.0 | 0.92 130 | 1.02 130 | 0.94 130 | 3.93 130 | 4.38 130 | 3.53 130 | 3.56 130 | 3.06 130 | 3.57 130 | 3.28 130 | 3.78 130 | 3.41 130 | 6.48 130 | 7.07 130 | 5.99 130 | 5.88 130 | 4.40 129 | 4.70 130 | 1.79 130 | 1.19 130 | 3.18 130 | 7.91 130 | 8.50 130 | 7.97 130 |
SepConv-v1 [127] | 130.0 | 0.92 130 | 1.02 130 | 0.94 130 | 3.93 130 | 4.38 130 | 3.53 130 | 3.56 130 | 3.06 130 | 3.57 130 | 3.28 130 | 3.78 130 | 3.41 130 | 6.48 130 | 7.07 130 | 5.99 130 | 5.88 130 | 4.40 129 | 4.70 130 | 1.79 130 | 1.19 130 | 3.18 130 | 7.91 130 | 8.50 130 | 7.97 130 |
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. |