Optical flow evaluation results |
Statistics:
Average
SD
R0.5
R1.0
R2.0
A50
A75
A95
Error type: endpoint angle interpolation normalized interpolation |
R1.0 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] | 6.7 | 0.82 12 | 4.87 13 | 0.37 16 | 1.75 7 | 12.1 8 | 0.53 6 | 2.22 2 | 7.90 2 | 0.57 7 | 1.07 3 | 9.10 5 | 0.17 2 | 9.77 1 | 16.5 1 | 2.56 2 | 4.53 3 | 15.6 2 | 3.00 3 | 0.00 1 | 0.02 30 | 0.00 1 | 5.99 9 | 19.5 22 | 3.94 2 |
OFLAF [77] | 8.5 | 0.82 12 | 4.86 12 | 0.38 18 | 1.74 5 | 11.1 5 | 0.62 12 | 2.08 1 | 7.42 1 | 0.57 7 | 1.61 10 | 12.0 10 | 0.48 12 | 11.2 6 | 19.0 6 | 3.96 6 | 6.81 21 | 19.8 11 | 4.79 21 | 0.00 1 | 0.00 1 | 0.00 1 | 5.80 6 | 15.6 3 | 9.76 15 |
PMMST [114] | 8.9 | 0.65 2 | 3.86 2 | 0.05 1 | 2.23 18 | 13.5 16 | 1.21 33 | 2.81 8 | 9.66 8 | 0.83 13 | 1.29 5 | 6.70 2 | 0.42 10 | 11.7 7 | 19.1 7 | 5.55 10 | 5.50 8 | 17.8 6 | 4.52 13 | 0.00 1 | 0.02 30 | 0.00 1 | 5.44 3 | 15.8 4 | 5.70 5 |
NN-field [71] | 11.4 | 0.89 25 | 5.29 27 | 0.40 25 | 2.06 13 | 14.1 21 | 0.62 12 | 2.49 4 | 8.79 5 | 0.68 10 | 0.99 2 | 8.66 4 | 0.09 1 | 9.99 2 | 16.8 2 | 2.51 1 | 6.53 19 | 11.2 1 | 2.42 2 | 0.01 36 | 0.02 30 | 0.00 1 | 5.86 7 | 19.6 23 | 2.84 1 |
MDP-Flow2 [68] | 12.6 | 0.77 6 | 4.59 6 | 0.31 4 | 1.46 2 | 9.56 1 | 0.39 1 | 2.59 5 | 9.00 6 | 0.91 18 | 2.48 45 | 17.7 51 | 0.70 44 | 14.1 20 | 23.0 19 | 8.20 23 | 5.27 5 | 18.2 7 | 4.66 16 | 0.00 1 | 0.00 1 | 0.00 1 | 5.91 8 | 16.7 6 | 8.80 7 |
Correlation Flow [75] | 16.1 | 0.81 9 | 4.81 10 | 0.22 2 | 2.03 10 | 13.0 11 | 0.42 2 | 5.14 49 | 15.7 47 | 0.55 6 | 1.09 4 | 8.36 3 | 0.28 6 | 16.6 38 | 26.1 38 | 10.8 41 | 7.92 30 | 22.7 23 | 4.18 10 | 0.00 1 | 0.02 30 | 0.00 1 | 5.54 4 | 17.2 7 | 5.00 4 |
WLIF-Flow [93] | 16.6 | 0.84 17 | 4.97 18 | 0.34 10 | 2.03 10 | 13.3 14 | 0.76 19 | 3.64 18 | 12.0 17 | 1.41 30 | 2.23 33 | 14.4 24 | 0.55 18 | 13.1 13 | 21.6 13 | 7.54 16 | 8.23 36 | 20.9 15 | 5.39 28 | 0.00 1 | 0.00 1 | 0.00 1 | 6.94 15 | 18.0 11 | 10.4 20 |
NNF-EAC [103] | 21.6 | 0.81 9 | 4.82 11 | 0.39 23 | 1.95 9 | 12.0 7 | 0.83 22 | 3.16 11 | 10.6 11 | 1.00 20 | 2.60 49 | 18.5 54 | 0.77 49 | 13.9 17 | 22.8 18 | 7.86 18 | 6.67 20 | 19.3 9 | 5.14 26 | 0.10 49 | 0.02 30 | 0.00 1 | 7.08 18 | 19.2 17 | 10.5 21 |
ComponentFusion [96] | 22.7 | 0.98 47 | 5.81 54 | 0.37 16 | 1.59 4 | 10.7 4 | 0.53 6 | 2.84 9 | 9.86 9 | 0.85 16 | 1.94 17 | 13.3 13 | 0.54 17 | 15.3 32 | 24.9 32 | 10.5 40 | 6.83 22 | 26.2 46 | 5.50 31 | 0.03 42 | 0.00 1 | 0.32 52 | 6.69 11 | 18.5 13 | 9.59 11 |
TC/T-Flow [76] | 23.4 | 0.71 3 | 4.20 3 | 0.40 25 | 2.67 32 | 15.4 28 | 0.77 20 | 3.30 14 | 11.2 12 | 0.44 3 | 2.33 37 | 15.9 39 | 0.60 27 | 14.8 25 | 23.6 24 | 8.02 19 | 3.70 1 | 15.8 3 | 2.27 1 | 0.13 52 | 0.02 30 | 1.23 70 | 7.85 31 | 21.7 34 | 10.9 29 |
Layers++ [37] | 23.9 | 0.91 31 | 5.39 33 | 0.43 39 | 2.18 17 | 13.9 20 | 0.96 24 | 2.73 7 | 9.43 7 | 1.40 29 | 1.70 12 | 10.5 7 | 0.56 19 | 10.2 3 | 16.8 2 | 6.50 13 | 9.09 48 | 22.7 23 | 5.92 42 | 0.21 59 | 0.02 30 | 0.69 56 | 6.88 13 | 17.6 10 | 10.9 29 |
FC-2Layers-FF [74] | 24.7 | 0.87 22 | 5.16 23 | 0.42 34 | 2.70 34 | 17.8 39 | 1.20 31 | 2.59 5 | 8.73 4 | 1.39 28 | 1.88 15 | 13.3 13 | 0.50 13 | 11.1 5 | 18.0 5 | 6.07 11 | 9.16 51 | 21.3 17 | 5.89 40 | 0.04 45 | 0.02 30 | 0.22 50 | 7.48 24 | 19.4 20 | 11.1 34 |
AGIF+OF [85] | 25.2 | 0.90 27 | 5.34 29 | 0.42 34 | 3.13 45 | 19.3 48 | 1.37 38 | 3.87 22 | 12.8 19 | 1.80 43 | 2.19 30 | 14.3 22 | 0.64 32 | 12.4 9 | 20.6 9 | 7.20 15 | 9.27 53 | 22.4 20 | 5.97 46 | 0.00 1 | 0.00 1 | 0.00 1 | 7.26 20 | 18.8 15 | 10.7 26 |
LME [70] | 27.4 | 0.95 43 | 5.67 48 | 0.38 18 | 1.45 1 | 9.68 2 | 0.43 3 | 5.19 50 | 13.3 27 | 6.57 91 | 2.44 43 | 18.3 52 | 0.68 38 | 15.2 30 | 24.4 28 | 10.2 37 | 6.17 17 | 21.9 18 | 5.18 27 | 0.00 1 | 0.02 30 | 0.00 1 | 7.05 16 | 19.3 18 | 10.2 18 |
ALD-Flow [66] | 28.7 | 0.79 7 | 4.72 9 | 0.38 18 | 2.44 28 | 13.5 16 | 0.80 21 | 4.33 37 | 14.7 40 | 0.88 17 | 2.92 64 | 19.4 58 | 0.82 53 | 17.5 41 | 28.1 42 | 10.0 35 | 5.61 10 | 24.7 39 | 3.10 4 | 0.00 1 | 0.00 1 | 0.00 1 | 9.09 45 | 26.3 51 | 11.9 50 |
nLayers [57] | 28.8 | 0.88 23 | 5.25 25 | 0.44 43 | 2.79 36 | 15.6 32 | 1.47 43 | 4.34 38 | 14.4 36 | 2.33 57 | 1.54 8 | 11.6 9 | 0.52 16 | 10.4 4 | 17.1 4 | 5.51 9 | 8.89 42 | 19.3 9 | 5.79 39 | 0.31 70 | 0.00 1 | 1.16 68 | 7.27 21 | 19.3 18 | 11.3 40 |
MLDP_OF [89] | 29.2 | 0.94 40 | 5.51 43 | 0.36 14 | 1.74 5 | 11.7 6 | 0.44 5 | 4.05 28 | 13.1 24 | 0.50 5 | 1.48 7 | 12.3 11 | 0.29 7 | 15.4 33 | 24.7 31 | 9.15 28 | 5.54 9 | 18.2 7 | 3.11 5 | 1.54 108 | 0.05 89 | 9.31 115 | 8.33 37 | 21.4 33 | 9.39 10 |
PH-Flow [101] | 30.3 | 0.93 38 | 5.49 41 | 0.42 34 | 2.87 37 | 17.6 38 | 1.33 37 | 2.99 10 | 10.1 10 | 1.76 42 | 2.27 35 | 14.6 29 | 0.68 38 | 12.5 10 | 20.8 10 | 6.28 12 | 7.79 27 | 20.9 15 | 5.39 28 | 0.39 78 | 0.02 30 | 1.63 81 | 6.88 13 | 19.0 16 | 10.3 19 |
RNLOD-Flow [121] | 30.4 | 0.79 7 | 4.69 7 | 0.34 10 | 2.67 32 | 17.2 36 | 1.09 28 | 4.46 42 | 14.5 38 | 1.53 34 | 2.01 20 | 14.3 22 | 0.60 27 | 14.2 22 | 23.1 21 | 8.72 26 | 8.21 34 | 19.9 12 | 5.90 41 | 0.35 74 | 0.03 85 | 1.48 79 | 6.51 10 | 17.2 7 | 9.80 16 |
HAST [109] | 31.2 | 0.92 37 | 5.41 35 | 0.35 12 | 3.21 48 | 13.6 18 | 1.99 68 | 2.45 3 | 8.47 3 | 0.29 1 | 2.24 34 | 14.5 27 | 0.40 9 | 11.7 7 | 19.4 8 | 3.63 3 | 11.0 83 | 24.2 35 | 6.87 77 | 2.75 117 | 0.00 1 | 11.6 118 | 4.05 1 | 13.1 1 | 4.43 3 |
ProbFlowFields [128] | 32.6 | 1.16 69 | 6.86 77 | 0.85 90 | 2.32 20 | 14.6 23 | 1.45 41 | 4.28 35 | 14.9 42 | 2.43 60 | 1.56 9 | 9.89 6 | 0.50 13 | 18.1 45 | 29.1 47 | 11.5 45 | 4.44 2 | 20.6 13 | 4.20 11 | 0.00 1 | 0.02 30 | 0.00 1 | 8.97 43 | 25.7 46 | 9.75 13 |
IROF++ [58] | 33.0 | 0.96 44 | 5.70 52 | 0.44 43 | 3.00 42 | 19.4 49 | 1.37 38 | 3.90 24 | 12.8 19 | 1.96 48 | 2.36 40 | 15.8 38 | 0.69 42 | 14.1 20 | 23.0 19 | 8.22 24 | 9.14 50 | 25.0 40 | 6.07 52 | 0.00 1 | 0.02 30 | 0.00 1 | 7.35 22 | 20.3 26 | 10.8 27 |
TC-Flow [46] | 33.3 | 0.75 5 | 4.45 5 | 0.38 18 | 2.04 12 | 12.6 10 | 0.70 15 | 4.23 34 | 14.4 36 | 0.77 11 | 2.56 47 | 17.5 50 | 0.63 30 | 17.1 40 | 27.8 41 | 9.45 30 | 5.73 12 | 25.6 43 | 3.12 6 | 0.22 61 | 0.02 30 | 2.41 92 | 10.1 52 | 25.9 48 | 15.4 71 |
Efficient-NL [60] | 33.3 | 0.93 38 | 5.47 40 | 0.39 23 | 2.76 35 | 18.0 41 | 1.11 29 | 4.12 32 | 13.3 27 | 1.15 23 | 2.15 27 | 14.1 20 | 0.66 34 | 13.0 12 | 21.3 12 | 7.16 14 | 10.6 77 | 23.4 29 | 6.41 66 | 0.26 64 | 0.02 30 | 1.13 66 | 7.35 22 | 17.4 9 | 10.9 29 |
SVFilterOh [111] | 33.9 | 1.07 57 | 6.27 64 | 0.44 43 | 2.07 14 | 13.1 13 | 0.72 16 | 3.24 12 | 11.2 12 | 1.05 21 | 1.99 19 | 13.8 17 | 0.56 19 | 12.6 11 | 21.1 11 | 3.80 4 | 10.5 75 | 22.4 20 | 5.97 46 | 2.31 113 | 0.39 108 | 6.95 108 | 4.87 2 | 14.8 2 | 6.01 6 |
FESL [72] | 34.0 | 0.83 15 | 4.91 16 | 0.36 14 | 3.90 73 | 21.6 67 | 1.75 56 | 4.06 29 | 13.4 29 | 1.61 37 | 2.02 22 | 14.1 20 | 0.56 19 | 13.3 14 | 21.7 14 | 8.08 21 | 9.19 52 | 22.0 19 | 6.25 58 | 0.34 73 | 0.02 30 | 1.16 68 | 7.51 25 | 18.3 12 | 11.0 33 |
Classic+CPF [83] | 34.2 | 0.89 25 | 5.26 26 | 0.41 30 | 3.03 43 | 19.4 49 | 1.27 36 | 4.14 33 | 13.6 30 | 1.64 40 | 2.12 26 | 14.4 24 | 0.64 32 | 13.6 15 | 22.2 16 | 7.82 17 | 9.85 66 | 22.6 22 | 6.18 55 | 0.36 75 | 0.02 30 | 1.50 80 | 7.07 17 | 18.5 13 | 10.5 21 |
FMOF [94] | 35.2 | 0.83 15 | 4.92 17 | 0.43 39 | 3.35 55 | 20.0 54 | 1.57 50 | 3.37 16 | 11.4 15 | 1.46 32 | 1.98 18 | 13.8 17 | 0.56 19 | 14.2 22 | 23.2 22 | 8.08 21 | 9.98 70 | 22.7 23 | 6.19 56 | 0.42 81 | 0.02 30 | 1.87 86 | 8.11 36 | 21.0 30 | 10.5 21 |
HBM-GC [105] | 36.3 | 1.24 75 | 7.38 79 | 0.52 57 | 2.50 30 | 15.5 31 | 1.40 40 | 4.06 29 | 14.1 34 | 1.32 26 | 1.77 13 | 13.2 12 | 0.61 29 | 13.7 16 | 22.1 15 | 8.06 20 | 8.98 45 | 16.5 4 | 4.42 12 | 1.30 105 | 0.02 30 | 3.28 99 | 7.20 19 | 19.8 24 | 10.8 27 |
Ramp [62] | 36.7 | 0.90 27 | 5.36 30 | 0.41 30 | 3.14 46 | 20.0 54 | 1.52 45 | 3.86 21 | 12.9 21 | 1.93 46 | 2.01 20 | 14.5 27 | 0.59 25 | 15.1 29 | 24.4 28 | 9.67 32 | 9.44 56 | 22.9 26 | 5.95 44 | 0.29 68 | 0.02 30 | 1.38 75 | 7.83 30 | 20.9 28 | 11.5 43 |
Aniso-Texture [82] | 36.7 | 0.73 4 | 4.33 4 | 0.33 7 | 1.83 8 | 12.4 9 | 0.91 23 | 6.29 60 | 18.2 54 | 1.57 35 | 1.35 6 | 11.4 8 | 0.18 4 | 19.7 57 | 29.7 49 | 16.8 76 | 9.10 49 | 26.1 45 | 5.78 38 | 0.26 64 | 0.18 96 | 0.07 42 | 9.32 47 | 24.3 45 | 12.2 51 |
Sparse-NonSparse [56] | 37.0 | 0.88 23 | 5.21 24 | 0.40 25 | 3.16 47 | 19.8 53 | 1.53 48 | 3.90 24 | 12.9 21 | 2.00 50 | 2.18 29 | 15.2 35 | 0.66 34 | 15.6 34 | 25.4 35 | 10.1 36 | 9.38 55 | 23.7 33 | 5.97 46 | 0.31 70 | 0.00 1 | 1.28 71 | 7.74 27 | 20.9 28 | 11.2 38 |
PMF [73] | 37.2 | 1.08 58 | 6.23 62 | 0.35 12 | 2.33 21 | 14.8 24 | 0.60 10 | 3.87 22 | 13.6 30 | 0.62 9 | 2.29 36 | 14.4 24 | 0.44 11 | 14.0 19 | 23.3 23 | 3.86 5 | 9.55 57 | 28.3 63 | 6.63 71 | 0.89 99 | 0.79 118 | 3.74 103 | 5.66 5 | 15.8 4 | 8.92 8 |
CombBMOF [113] | 37.7 | 0.91 31 | 5.38 31 | 0.33 7 | 2.30 19 | 13.4 15 | 0.64 14 | 3.33 15 | 11.5 16 | 0.78 12 | 2.08 24 | 15.3 36 | 0.77 49 | 13.9 17 | 22.3 17 | 8.24 25 | 13.0 101 | 26.2 46 | 11.4 108 | 0.56 86 | 0.02 30 | 0.86 60 | 8.93 42 | 21.1 31 | 15.6 72 |
NL-TV-NCC [25] | 38.5 | 0.96 44 | 5.68 49 | 0.22 2 | 2.93 39 | 18.4 43 | 0.59 9 | 4.37 40 | 14.6 39 | 0.47 4 | 1.63 11 | 14.6 29 | 0.17 2 | 18.6 47 | 29.8 50 | 9.76 34 | 11.8 92 | 31.2 85 | 7.70 91 | 0.12 50 | 0.00 1 | 0.30 51 | 9.40 49 | 26.0 50 | 9.75 13 |
LSM [39] | 39.0 | 0.86 21 | 5.13 22 | 0.40 25 | 3.22 49 | 20.3 57 | 1.54 49 | 4.08 31 | 13.6 30 | 1.93 46 | 2.09 25 | 14.9 33 | 0.63 30 | 15.6 34 | 25.3 34 | 10.2 37 | 9.58 58 | 24.6 36 | 5.95 44 | 0.30 69 | 0.02 30 | 1.43 76 | 7.97 32 | 21.7 34 | 11.1 34 |
OFH [38] | 40.3 | 0.81 9 | 4.70 8 | 0.31 4 | 2.96 41 | 17.3 37 | 1.20 31 | 6.37 62 | 19.7 60 | 1.51 33 | 2.92 64 | 20.6 66 | 0.91 55 | 20.7 66 | 32.4 69 | 14.2 58 | 6.39 18 | 31.5 87 | 3.74 9 | 0.00 1 | 0.00 1 | 0.00 1 | 11.0 58 | 33.0 76 | 12.8 54 |
Sparse Occlusion [54] | 40.4 | 0.90 27 | 5.06 21 | 0.46 48 | 2.35 23 | 14.9 26 | 1.01 25 | 4.83 45 | 15.7 47 | 1.09 22 | 2.38 41 | 17.2 48 | 0.66 34 | 16.7 39 | 26.9 39 | 8.75 27 | 7.98 31 | 24.6 36 | 5.42 30 | 0.60 90 | 0.61 114 | 0.84 58 | 8.41 40 | 22.7 38 | 10.5 21 |
Classic+NL [31] | 40.8 | 0.91 31 | 5.38 31 | 0.45 47 | 3.22 49 | 20.4 58 | 1.49 44 | 3.97 26 | 13.1 24 | 1.97 49 | 2.33 37 | 15.0 34 | 0.68 38 | 14.9 26 | 24.0 26 | 10.2 37 | 9.83 63 | 23.9 34 | 6.24 57 | 0.33 72 | 0.02 30 | 1.28 71 | 7.80 29 | 21.2 32 | 11.1 34 |
EPPM w/o HM [88] | 41.5 | 1.16 69 | 5.61 46 | 0.33 7 | 2.33 21 | 15.4 28 | 0.60 10 | 4.28 35 | 14.7 40 | 0.32 2 | 2.20 31 | 14.7 31 | 0.59 25 | 14.3 24 | 23.6 24 | 5.47 8 | 12.2 93 | 29.9 77 | 7.04 79 | 2.28 112 | 0.03 85 | 6.80 107 | 6.72 12 | 19.4 20 | 8.96 9 |
MDP-Flow [26] | 41.6 | 0.84 17 | 5.01 19 | 0.47 49 | 2.37 24 | 13.0 11 | 1.76 57 | 4.04 27 | 14.0 33 | 2.72 65 | 2.70 53 | 21.0 67 | 0.98 60 | 18.0 44 | 28.5 44 | 13.1 54 | 8.58 40 | 26.6 50 | 5.71 35 | 0.00 1 | 0.02 30 | 0.00 1 | 12.4 74 | 31.9 68 | 16.2 76 |
CostFilter [40] | 44.2 | 1.14 65 | 6.62 70 | 0.40 25 | 2.38 25 | 14.8 24 | 0.53 6 | 3.58 17 | 12.5 18 | 0.84 14 | 2.62 50 | 17.3 49 | 0.51 15 | 14.9 26 | 24.9 32 | 4.14 7 | 9.99 71 | 29.2 72 | 6.06 51 | 1.38 106 | 0.81 119 | 6.01 106 | 8.00 33 | 23.2 42 | 9.84 17 |
OAR-Flow [125] | 44.8 | 1.00 49 | 5.81 54 | 0.55 66 | 3.94 75 | 18.5 44 | 1.99 68 | 6.44 64 | 20.5 64 | 2.66 64 | 2.84 61 | 18.5 54 | 0.71 46 | 18.9 51 | 30.1 54 | 11.2 42 | 5.95 15 | 26.2 46 | 3.42 8 | 0.00 1 | 0.00 1 | 0.00 1 | 9.00 44 | 26.4 52 | 12.2 51 |
Complementary OF [21] | 45.4 | 0.91 31 | 5.39 33 | 0.43 39 | 2.42 27 | 15.2 27 | 0.74 17 | 4.36 39 | 15.5 45 | 1.16 24 | 2.63 51 | 19.5 59 | 0.76 48 | 22.5 83 | 33.0 74 | 20.1 82 | 9.92 68 | 28.5 66 | 4.80 22 | 0.00 1 | 0.00 1 | 0.00 1 | 12.6 77 | 35.6 95 | 16.9 80 |
S2D-Matching [84] | 46.3 | 1.09 60 | 6.39 67 | 0.51 55 | 3.35 55 | 20.9 60 | 1.52 45 | 5.55 53 | 17.8 53 | 2.21 53 | 1.91 16 | 13.6 16 | 0.56 19 | 15.2 30 | 24.5 30 | 9.72 33 | 10.1 73 | 23.6 31 | 6.34 62 | 0.52 83 | 0.02 30 | 2.09 90 | 7.62 26 | 20.0 25 | 11.6 46 |
COFM [59] | 47.1 | 1.15 66 | 6.80 75 | 0.58 74 | 2.62 31 | 15.8 34 | 1.25 35 | 5.68 55 | 18.2 54 | 2.12 51 | 2.20 31 | 13.5 15 | 0.58 24 | 19.6 56 | 31.0 61 | 15.7 70 | 9.91 67 | 23.3 28 | 6.03 50 | 0.81 96 | 0.00 1 | 1.43 76 | 7.76 28 | 20.7 27 | 10.6 25 |
SimpleFlow [49] | 47.5 | 0.94 40 | 5.57 45 | 0.44 43 | 3.52 59 | 21.7 68 | 1.79 60 | 5.82 56 | 17.6 52 | 2.36 58 | 2.55 46 | 16.5 42 | 0.81 52 | 16.3 37 | 26.0 36 | 11.8 46 | 10.3 74 | 23.1 27 | 6.33 61 | 0.24 62 | 0.00 1 | 0.81 57 | 8.33 37 | 22.7 38 | 11.5 43 |
IROF-TV [53] | 47.5 | 1.10 62 | 6.24 63 | 0.57 71 | 3.29 54 | 21.5 65 | 1.72 55 | 4.40 41 | 14.2 35 | 1.87 45 | 3.04 68 | 21.7 73 | 1.11 64 | 16.2 36 | 26.0 36 | 11.3 44 | 9.60 60 | 32.4 92 | 5.72 36 | 0.00 1 | 0.02 30 | 0.00 1 | 8.00 33 | 22.4 37 | 11.2 38 |
2DHMM-SAS [92] | 47.9 | 0.91 31 | 5.42 36 | 0.41 30 | 3.67 66 | 21.9 70 | 1.52 45 | 5.62 54 | 16.1 49 | 2.28 56 | 2.44 43 | 15.9 39 | 0.72 47 | 15.0 28 | 24.2 27 | 9.48 31 | 11.1 86 | 25.1 42 | 6.36 64 | 0.38 77 | 0.02 30 | 1.67 83 | 8.04 35 | 21.7 34 | 11.7 47 |
ACK-Prior [27] | 48.1 | 0.82 12 | 4.87 13 | 0.32 6 | 2.12 16 | 13.7 19 | 0.43 3 | 3.68 19 | 12.9 21 | 0.92 19 | 1.77 13 | 14.0 19 | 0.19 5 | 19.5 55 | 28.2 43 | 16.7 75 | 12.3 95 | 29.1 71 | 7.52 89 | 2.44 115 | 0.30 103 | 8.47 114 | 13.9 84 | 30.2 63 | 18.0 83 |
ROF-ND [107] | 49.2 | 1.27 80 | 6.15 61 | 0.38 18 | 4.71 84 | 18.9 46 | 1.07 27 | 4.89 46 | 15.6 46 | 1.21 25 | 0.65 1 | 6.22 1 | 0.29 7 | 19.7 57 | 30.5 56 | 14.5 60 | 11.5 91 | 26.5 49 | 6.25 58 | 0.39 78 | 0.02 30 | 0.84 58 | 12.3 73 | 31.5 66 | 13.8 62 |
TV-L1-MCT [64] | 50.5 | 0.90 27 | 5.30 28 | 0.41 30 | 3.73 67 | 22.1 71 | 1.79 60 | 4.61 44 | 15.3 43 | 1.63 39 | 2.16 28 | 14.7 31 | 0.67 37 | 17.6 42 | 27.1 40 | 15.2 66 | 11.0 83 | 25.0 40 | 6.58 70 | 0.36 75 | 0.02 30 | 2.46 94 | 9.73 50 | 23.0 40 | 16.2 76 |
RFlow [90] | 52.3 | 0.91 31 | 5.43 37 | 0.47 49 | 2.46 29 | 15.6 32 | 1.13 30 | 6.42 63 | 19.3 59 | 1.66 41 | 2.77 56 | 21.4 71 | 1.16 67 | 20.7 66 | 31.7 65 | 18.0 80 | 9.69 62 | 30.4 78 | 6.14 53 | 0.01 36 | 0.02 30 | 0.15 45 | 10.9 57 | 30.0 62 | 13.1 57 |
S2F-IF [123] | 53.8 | 1.28 82 | 7.44 83 | 0.84 89 | 3.48 58 | 22.4 74 | 1.86 64 | 5.52 51 | 19.0 56 | 3.05 66 | 2.96 66 | 16.9 46 | 1.21 69 | 21.3 73 | 34.1 83 | 14.5 60 | 5.45 6 | 25.6 43 | 4.62 14 | 0.00 1 | 0.00 1 | 0.00 1 | 12.0 69 | 32.3 72 | 14.7 65 |
DPOF [18] | 54.1 | 1.11 63 | 6.56 69 | 0.53 59 | 4.51 82 | 21.0 61 | 2.42 80 | 3.25 13 | 11.3 14 | 0.84 14 | 2.03 23 | 15.3 36 | 0.70 44 | 17.8 43 | 28.8 45 | 9.36 29 | 11.4 90 | 26.9 51 | 6.26 60 | 4.21 121 | 0.02 30 | 10.5 117 | 10.2 53 | 26.7 53 | 11.8 48 |
Occlusion-TV-L1 [63] | 54.1 | 0.98 47 | 5.50 42 | 0.48 51 | 3.25 51 | 19.5 51 | 1.82 63 | 7.36 77 | 21.2 72 | 2.44 61 | 2.73 54 | 20.4 64 | 0.93 58 | 20.5 63 | 32.1 66 | 15.5 68 | 8.22 35 | 28.1 60 | 6.69 73 | 0.00 1 | 0.00 1 | 0.00 1 | 13.1 81 | 33.5 83 | 15.9 75 |
DeepFlow2 [108] | 54.4 | 1.04 54 | 5.76 53 | 0.54 64 | 3.86 71 | 19.7 52 | 2.02 71 | 6.79 67 | 20.5 64 | 3.55 72 | 3.64 79 | 22.5 76 | 1.44 76 | 18.8 49 | 30.1 54 | 12.0 48 | 7.01 24 | 27.7 58 | 4.65 15 | 0.00 1 | 0.02 30 | 0.00 1 | 12.6 77 | 32.0 70 | 16.9 80 |
FlowFields+ [130] | 56.1 | 1.31 84 | 7.52 85 | 0.92 97 | 3.61 62 | 23.0 77 | 1.98 67 | 6.08 57 | 20.6 66 | 3.39 69 | 2.82 59 | 16.4 41 | 1.23 70 | 21.4 76 | 34.3 86 | 14.1 56 | 5.45 6 | 27.5 57 | 4.68 18 | 0.00 1 | 0.02 30 | 0.00 1 | 11.3 61 | 33.1 79 | 11.4 41 |
TF+OM [100] | 57.2 | 1.11 63 | 6.49 68 | 0.69 82 | 2.94 40 | 16.8 35 | 1.78 58 | 7.92 81 | 20.7 68 | 9.65 93 | 2.85 62 | 20.5 65 | 1.05 63 | 22.0 81 | 32.2 67 | 20.3 83 | 8.74 41 | 28.3 63 | 4.67 17 | 0.00 1 | 0.02 30 | 0.00 1 | 12.1 71 | 30.5 65 | 15.8 74 |
AggregFlow [97] | 58.2 | 1.68 99 | 9.22 104 | 0.86 92 | 4.76 85 | 25.3 89 | 2.56 83 | 7.33 76 | 22.6 80 | 5.07 87 | 2.64 52 | 16.5 42 | 0.69 42 | 19.1 53 | 30.7 58 | 11.2 42 | 5.11 4 | 17.3 5 | 3.29 7 | 0.14 53 | 0.02 30 | 0.96 63 | 9.35 48 | 25.7 46 | 13.1 57 |
Steered-L1 [118] | 58.2 | 0.63 1 | 3.72 1 | 0.42 34 | 1.53 3 | 10.4 3 | 0.75 18 | 3.84 20 | 13.2 26 | 1.32 26 | 2.80 57 | 21.1 69 | 0.98 60 | 21.2 72 | 31.2 62 | 20.3 83 | 10.7 80 | 29.3 74 | 7.45 88 | 4.27 122 | 0.34 105 | 19.6 124 | 16.5 92 | 33.3 80 | 24.6 98 |
FlowFields [110] | 58.4 | 1.32 85 | 7.63 87 | 0.93 99 | 3.61 62 | 22.9 76 | 2.00 70 | 6.11 58 | 20.6 66 | 3.58 73 | 2.85 62 | 16.6 44 | 1.24 71 | 21.9 80 | 35.0 95 | 15.1 65 | 5.70 11 | 28.0 59 | 4.72 19 | 0.00 1 | 0.02 30 | 0.00 1 | 11.7 64 | 33.4 81 | 11.5 43 |
CRTflow [80] | 58.6 | 1.02 53 | 5.69 50 | 0.58 74 | 3.12 44 | 18.1 42 | 1.46 42 | 6.89 69 | 20.9 70 | 2.40 59 | 3.38 75 | 22.2 74 | 1.42 75 | 19.7 57 | 31.3 63 | 12.3 49 | 11.0 83 | 35.9 104 | 10.1 103 | 0.00 1 | 0.00 1 | 0.00 1 | 12.0 69 | 33.6 84 | 14.7 65 |
ComplOF-FED-GPU [35] | 59.4 | 0.85 19 | 5.04 20 | 0.42 34 | 3.90 73 | 21.4 63 | 1.78 58 | 4.90 47 | 16.8 50 | 1.41 30 | 3.18 70 | 21.1 69 | 1.03 62 | 21.6 78 | 33.8 81 | 15.4 67 | 10.8 81 | 34.7 102 | 5.93 43 | 0.12 50 | 0.02 30 | 1.43 76 | 11.9 66 | 34.2 87 | 15.3 69 |
TCOF [69] | 59.6 | 1.00 49 | 5.63 47 | 0.59 77 | 3.53 60 | 21.5 65 | 1.69 54 | 7.64 79 | 22.0 76 | 3.79 76 | 2.80 57 | 19.9 62 | 0.77 49 | 21.1 70 | 32.7 71 | 13.9 55 | 7.79 27 | 20.7 14 | 5.77 37 | 0.92 100 | 0.03 85 | 3.23 98 | 8.40 39 | 23.2 42 | 11.4 41 |
Adaptive [20] | 61.0 | 1.05 55 | 6.01 58 | 0.48 51 | 3.27 52 | 20.1 56 | 1.79 60 | 7.11 73 | 20.1 61 | 1.62 38 | 3.29 72 | 22.4 75 | 1.15 66 | 18.8 49 | 29.8 50 | 12.6 51 | 10.6 77 | 28.4 65 | 6.72 75 | 0.57 89 | 0.71 117 | 0.96 63 | 8.64 41 | 23.1 41 | 10.9 29 |
SRR-TVOF-NL [91] | 62.2 | 1.15 66 | 6.13 60 | 0.60 78 | 5.04 87 | 23.3 79 | 2.68 85 | 8.17 82 | 22.9 81 | 4.22 84 | 2.76 55 | 16.9 46 | 0.68 38 | 19.8 60 | 29.0 46 | 17.7 79 | 8.07 32 | 27.0 53 | 6.17 54 | 0.16 55 | 0.02 30 | 0.86 60 | 11.8 65 | 25.9 48 | 15.3 69 |
TV-L1-improved [17] | 63.4 | 0.94 40 | 5.45 38 | 0.52 57 | 2.91 38 | 17.8 39 | 1.58 51 | 7.00 71 | 20.2 63 | 2.24 55 | 3.00 67 | 21.5 72 | 1.16 67 | 20.6 64 | 32.3 68 | 15.0 63 | 12.2 93 | 34.2 100 | 7.87 92 | 0.19 57 | 0.30 103 | 0.49 54 | 10.7 55 | 29.7 61 | 12.8 54 |
DeepFlow [86] | 63.6 | 1.19 71 | 6.04 59 | 0.57 71 | 4.41 79 | 21.3 62 | 2.41 79 | 8.35 85 | 22.9 81 | 6.63 92 | 4.03 88 | 24.5 84 | 1.69 84 | 19.0 52 | 30.9 59 | 11.8 46 | 7.29 25 | 29.5 76 | 4.88 23 | 0.00 1 | 0.02 30 | 0.00 1 | 15.7 91 | 35.1 93 | 23.4 95 |
PGM-C [120] | 64.8 | 1.52 92 | 8.68 98 | 0.99 104 | 3.66 64 | 23.0 77 | 2.03 72 | 6.30 61 | 21.2 72 | 3.90 78 | 3.82 84 | 22.9 80 | 1.68 83 | 21.3 73 | 34.3 86 | 14.1 56 | 6.89 23 | 28.8 68 | 5.60 33 | 0.00 1 | 0.02 30 | 0.00 1 | 11.9 66 | 34.3 88 | 14.2 64 |
Aniso. Huber-L1 [22] | 66.0 | 1.06 56 | 5.69 50 | 0.65 79 | 5.24 88 | 25.4 90 | 3.29 89 | 8.19 83 | 21.4 75 | 4.09 82 | 3.10 69 | 21.0 67 | 0.97 59 | 18.5 46 | 29.1 47 | 12.6 51 | 9.08 47 | 27.0 53 | 5.56 32 | 0.68 92 | 0.08 93 | 2.93 97 | 9.25 46 | 24.1 44 | 11.8 48 |
Classic++ [32] | 67.1 | 1.00 49 | 5.92 57 | 0.56 68 | 3.28 53 | 19.2 47 | 1.87 65 | 6.88 68 | 20.7 68 | 3.38 68 | 3.41 77 | 23.6 82 | 1.30 72 | 20.8 68 | 33.2 77 | 15.0 63 | 10.0 72 | 31.8 89 | 6.69 73 | 0.66 91 | 0.02 30 | 2.59 95 | 11.3 61 | 29.5 58 | 13.5 60 |
CPM-Flow [116] | 67.1 | 1.53 93 | 8.72 100 | 0.98 101 | 3.74 68 | 23.5 80 | 2.08 73 | 6.22 59 | 20.9 70 | 3.88 77 | 3.78 82 | 22.5 76 | 1.64 81 | 21.3 73 | 34.4 88 | 14.2 58 | 7.87 29 | 28.8 68 | 6.40 65 | 0.00 1 | 0.02 30 | 0.00 1 | 12.5 76 | 35.6 95 | 14.9 67 |
Bartels [41] | 67.9 | 1.28 82 | 7.59 86 | 0.50 53 | 2.39 26 | 15.4 28 | 1.04 26 | 5.52 51 | 19.2 57 | 2.54 63 | 2.83 60 | 19.9 62 | 1.30 72 | 22.7 86 | 34.1 83 | 20.4 85 | 9.92 68 | 30.5 79 | 6.93 78 | 1.88 111 | 0.02 30 | 12.3 119 | 12.7 79 | 31.9 68 | 16.4 78 |
CBF [12] | 68.5 | 0.85 19 | 4.89 15 | 0.43 39 | 4.99 86 | 22.3 73 | 4.63 96 | 6.60 65 | 19.2 57 | 4.08 81 | 3.61 78 | 24.5 84 | 1.49 77 | 20.0 61 | 30.9 59 | 16.2 73 | 9.67 61 | 27.4 56 | 5.64 34 | 2.65 116 | 0.37 106 | 6.97 109 | 11.5 63 | 28.4 57 | 16.9 80 |
Kuang [131] | 68.8 | 1.38 87 | 8.01 90 | 0.89 93 | 4.13 76 | 25.8 91 | 2.18 75 | 6.98 70 | 23.6 85 | 3.42 70 | 3.29 72 | 19.2 56 | 1.38 74 | 23.1 89 | 36.8 102 | 15.8 71 | 9.05 46 | 31.3 86 | 7.14 82 | 0.00 1 | 0.02 30 | 0.00 1 | 11.2 60 | 30.4 64 | 16.6 79 |
EpicFlow [102] | 69.4 | 1.51 90 | 8.63 97 | 0.98 101 | 3.76 69 | 23.5 80 | 2.11 74 | 7.14 74 | 23.6 85 | 3.97 80 | 3.79 83 | 22.6 78 | 1.64 81 | 21.5 77 | 34.4 88 | 14.8 62 | 8.97 44 | 29.2 72 | 6.53 67 | 0.00 1 | 0.02 30 | 0.00 1 | 12.4 74 | 34.5 89 | 15.2 68 |
LocallyOriented [52] | 70.5 | 1.78 105 | 9.64 107 | 0.77 86 | 6.11 96 | 28.2 97 | 3.79 93 | 10.9 93 | 28.0 98 | 5.52 89 | 3.28 71 | 19.5 59 | 1.55 78 | 22.8 87 | 33.9 82 | 17.6 78 | 9.84 65 | 24.6 36 | 6.63 71 | 0.00 1 | 0.00 1 | 0.00 1 | 11.9 66 | 29.6 60 | 15.7 73 |
CLG-TV [48] | 71.4 | 1.01 52 | 5.46 39 | 0.50 53 | 4.16 78 | 23.5 80 | 2.40 78 | 7.52 78 | 21.2 72 | 2.51 62 | 3.33 74 | 22.8 79 | 1.14 65 | 20.9 69 | 32.4 69 | 15.6 69 | 8.94 43 | 31.7 88 | 6.35 63 | 1.27 104 | 1.18 122 | 3.55 101 | 11.1 59 | 28.2 56 | 13.5 60 |
TriangleFlow [30] | 71.8 | 1.19 71 | 6.73 74 | 0.53 59 | 3.88 72 | 21.8 69 | 1.64 53 | 6.61 66 | 20.1 61 | 1.59 36 | 2.35 39 | 19.3 57 | 0.89 54 | 25.6 99 | 37.3 104 | 23.5 97 | 13.4 103 | 30.5 79 | 8.48 98 | 0.81 96 | 0.17 95 | 1.33 74 | 10.7 55 | 28.1 55 | 13.1 57 |
Rannacher [23] | 72.5 | 1.09 60 | 6.27 64 | 0.54 64 | 3.77 70 | 22.1 71 | 2.27 76 | 7.89 80 | 22.4 78 | 3.34 67 | 3.67 80 | 23.5 81 | 1.61 79 | 21.1 70 | 33.1 75 | 15.8 71 | 12.9 99 | 35.4 103 | 7.98 93 | 0.43 82 | 0.02 30 | 1.63 81 | 10.5 54 | 29.5 58 | 12.8 54 |
Fusion [6] | 72.5 | 1.15 66 | 6.83 76 | 0.71 84 | 2.10 15 | 14.5 22 | 1.23 34 | 4.58 43 | 15.4 44 | 3.60 74 | 3.73 81 | 27.2 93 | 2.38 90 | 23.9 92 | 33.7 79 | 26.4 102 | 8.36 37 | 27.3 55 | 7.11 81 | 1.00 101 | 0.64 116 | 2.66 96 | 14.5 89 | 34.0 86 | 18.7 85 |
SIOF [67] | 73.2 | 1.24 75 | 6.63 71 | 0.51 55 | 5.26 89 | 26.2 92 | 3.22 88 | 11.5 94 | 26.1 89 | 12.3 96 | 4.49 90 | 27.4 96 | 2.29 89 | 22.8 87 | 33.3 78 | 23.4 96 | 8.56 39 | 28.8 68 | 7.15 83 | 0.00 1 | 0.02 30 | 0.00 1 | 13.6 83 | 32.1 71 | 23.6 97 |
F-TV-L1 [15] | 74.0 | 1.22 73 | 6.63 71 | 0.53 59 | 5.86 94 | 24.8 87 | 3.51 92 | 9.25 88 | 23.4 84 | 3.44 71 | 3.91 85 | 24.9 86 | 1.61 79 | 20.3 62 | 31.5 64 | 16.2 73 | 11.3 88 | 30.9 83 | 7.40 87 | 0.15 54 | 0.47 111 | 0.17 46 | 9.88 51 | 27.6 54 | 11.1 34 |
Local-TV-L1 [65] | 75.8 | 1.57 94 | 7.45 84 | 0.67 81 | 7.93 100 | 28.0 96 | 5.97 101 | 12.9 100 | 26.6 91 | 12.2 95 | 6.04 104 | 31.1 103 | 3.55 103 | 18.7 48 | 29.9 53 | 12.9 53 | 9.33 54 | 28.1 60 | 5.97 46 | 0.00 1 | 0.02 30 | 0.00 1 | 21.0 107 | 37.7 100 | 37.5 113 |
p-harmonic [29] | 76.1 | 1.08 58 | 6.28 66 | 0.55 66 | 3.66 64 | 20.5 59 | 2.48 82 | 8.22 84 | 23.0 83 | 3.92 79 | 5.04 93 | 28.4 99 | 3.51 101 | 24.8 96 | 34.1 83 | 30.1 105 | 7.78 26 | 32.3 91 | 6.54 68 | 0.19 57 | 0.44 110 | 0.00 1 | 14.2 88 | 33.0 76 | 21.8 91 |
BriefMatch [124] | 76.6 | 0.96 44 | 5.52 44 | 0.53 59 | 3.55 61 | 18.6 45 | 1.97 66 | 4.95 48 | 16.9 51 | 1.82 44 | 2.57 48 | 19.7 61 | 0.92 57 | 21.8 79 | 32.7 71 | 20.8 88 | 16.2 112 | 33.7 98 | 13.5 114 | 3.95 120 | 0.97 121 | 15.8 120 | 17.3 97 | 34.7 91 | 25.4 100 |
TriFlow [95] | 79.0 | 1.51 90 | 8.71 99 | 0.78 87 | 4.56 83 | 22.8 75 | 3.37 91 | 12.5 98 | 28.2 99 | 17.8 100 | 2.41 42 | 18.3 52 | 0.91 55 | 24.8 96 | 34.4 88 | 25.7 100 | 5.97 16 | 23.5 30 | 4.74 20 | 19.3 128 | 0.27 101 | 59.0 128 | 13.2 82 | 32.5 74 | 14.1 63 |
Dynamic MRF [7] | 79.1 | 1.26 78 | 7.42 81 | 0.57 71 | 3.39 57 | 21.4 63 | 1.58 51 | 7.00 71 | 22.5 79 | 2.22 54 | 3.40 76 | 24.1 83 | 1.69 84 | 25.9 103 | 37.6 105 | 24.8 99 | 14.4 107 | 41.5 113 | 9.85 102 | 0.09 48 | 0.00 1 | 0.96 63 | 19.2 103 | 39.4 105 | 25.5 101 |
DF-Auto [115] | 79.8 | 1.84 106 | 8.87 102 | 0.90 94 | 8.40 101 | 30.1 100 | 6.82 102 | 13.3 101 | 27.9 97 | 19.6 101 | 5.29 95 | 26.6 89 | 3.01 94 | 22.2 82 | 32.9 73 | 21.0 89 | 5.80 13 | 23.6 31 | 5.02 25 | 0.18 56 | 0.61 114 | 0.00 1 | 14.8 90 | 32.4 73 | 19.9 86 |
Brox et al. [5] | 82.0 | 1.22 73 | 6.66 73 | 0.70 83 | 4.15 77 | 24.3 85 | 2.39 77 | 7.21 75 | 22.1 77 | 4.18 83 | 4.91 91 | 26.3 88 | 2.65 91 | 26.2 106 | 35.7 99 | 31.4 107 | 10.5 75 | 33.4 97 | 7.34 86 | 0.01 36 | 0.13 94 | 0.00 1 | 17.1 96 | 39.1 104 | 23.0 94 |
FlowNet2 [122] | 83.2 | 2.63 111 | 12.9 114 | 1.14 108 | 17.9 112 | 43.1 114 | 16.1 114 | 17.0 104 | 32.9 104 | 25.3 112 | 3.92 86 | 16.7 45 | 2.16 87 | 25.8 102 | 40.2 112 | 17.0 77 | 10.6 77 | 28.2 62 | 8.09 94 | 0.02 39 | 0.00 1 | 0.20 48 | 12.2 72 | 34.5 89 | 9.71 12 |
SuperFlow [81] | 84.4 | 1.26 78 | 5.91 56 | 0.66 80 | 6.58 98 | 24.8 87 | 5.70 100 | 12.7 99 | 26.7 92 | 20.1 102 | 5.60 99 | 28.6 100 | 3.33 98 | 24.6 95 | 33.7 79 | 31.7 110 | 8.09 33 | 30.8 82 | 7.19 84 | 0.02 39 | 0.07 91 | 0.02 41 | 16.6 93 | 37.3 98 | 22.5 92 |
Second-order prior [8] | 88.2 | 1.24 75 | 6.93 78 | 0.58 74 | 5.26 89 | 27.0 94 | 3.34 90 | 9.68 90 | 26.8 93 | 5.39 88 | 4.25 89 | 26.1 87 | 2.25 88 | 22.5 83 | 34.4 88 | 19.0 81 | 12.9 99 | 41.2 112 | 8.26 97 | 1.14 103 | 0.07 91 | 2.41 92 | 12.7 79 | 32.5 74 | 18.2 84 |
SegOF [10] | 90.1 | 1.62 97 | 9.24 105 | 1.14 108 | 14.8 111 | 38.8 111 | 14.3 112 | 17.8 106 | 33.2 105 | 22.3 106 | 6.57 105 | 27.5 97 | 4.43 106 | 32.5 116 | 41.8 115 | 43.4 120 | 14.1 106 | 38.0 107 | 10.5 104 | 0.00 1 | 0.00 1 | 0.00 1 | 14.0 85 | 33.7 85 | 12.6 53 |
StereoOF-V1MT [119] | 91.1 | 1.42 89 | 8.08 93 | 0.56 68 | 6.56 97 | 34.1 106 | 2.73 86 | 10.0 92 | 29.8 101 | 2.19 52 | 4.91 91 | 34.6 108 | 2.66 92 | 31.4 115 | 44.5 117 | 31.4 107 | 15.8 111 | 48.0 119 | 11.6 109 | 0.05 46 | 0.00 1 | 0.52 55 | 24.0 110 | 48.7 117 | 28.5 104 |
Shiralkar [42] | 91.7 | 1.27 80 | 7.43 82 | 0.53 59 | 5.83 93 | 30.1 100 | 2.93 87 | 9.62 89 | 26.2 90 | 3.70 75 | 5.11 94 | 30.7 102 | 3.08 95 | 25.7 100 | 39.1 109 | 22.5 94 | 17.9 113 | 45.5 115 | 9.73 101 | 1.80 109 | 0.00 1 | 8.23 113 | 18.4 101 | 44.9 113 | 19.9 86 |
FlowNetS+ft+v [112] | 92.8 | 1.40 88 | 7.39 80 | 0.80 88 | 5.75 92 | 23.6 83 | 4.35 95 | 11.7 96 | 27.3 95 | 12.4 97 | 5.33 96 | 27.2 93 | 3.18 97 | 25.7 100 | 35.4 98 | 26.9 103 | 8.52 38 | 32.0 90 | 6.85 76 | 2.34 114 | 1.61 125 | 10.1 116 | 14.0 85 | 34.9 92 | 20.1 89 |
CNN-flow-warp+ref [117] | 93.2 | 1.63 98 | 9.14 103 | 0.91 96 | 5.41 91 | 24.0 84 | 4.66 97 | 11.6 95 | 29.8 101 | 10.7 94 | 5.59 98 | 27.1 92 | 3.43 99 | 26.6 108 | 36.0 100 | 31.5 109 | 11.3 88 | 33.1 95 | 7.62 90 | 0.03 42 | 0.25 99 | 0.07 42 | 20.4 105 | 40.5 108 | 28.3 103 |
Ad-TV-NDC [36] | 93.5 | 3.59 116 | 8.26 94 | 6.67 125 | 21.3 116 | 38.0 109 | 22.4 119 | 19.7 109 | 33.6 106 | 21.8 105 | 13.5 111 | 33.9 107 | 15.0 112 | 19.4 54 | 30.6 57 | 12.5 50 | 9.58 58 | 28.6 67 | 6.54 68 | 0.21 59 | 0.37 106 | 0.17 46 | 28.1 116 | 43.2 111 | 47.4 122 |
Learning Flow [11] | 93.8 | 1.35 86 | 7.83 89 | 0.56 68 | 4.48 81 | 26.8 93 | 2.43 81 | 9.85 91 | 27.1 94 | 5.06 86 | 6.65 106 | 33.5 106 | 4.13 104 | 29.9 112 | 40.0 111 | 34.1 114 | 12.8 97 | 38.5 109 | 8.86 100 | 0.28 67 | 0.29 102 | 1.13 66 | 17.0 95 | 37.6 99 | 22.8 93 |
StereoFlow [44] | 93.8 | 7.67 126 | 21.8 123 | 3.86 122 | 51.5 128 | 74.0 129 | 46.2 125 | 43.7 129 | 63.5 129 | 36.8 122 | 51.6 128 | 79.4 129 | 47.5 127 | 26.1 104 | 38.0 106 | 21.1 90 | 5.83 14 | 26.9 51 | 4.93 24 | 0.00 1 | 0.02 30 | 0.00 1 | 20.7 106 | 38.1 102 | 29.7 105 |
2bit-BM-tele [98] | 94.2 | 1.75 101 | 9.59 106 | 0.85 90 | 4.44 80 | 24.7 86 | 2.62 84 | 8.64 86 | 25.8 88 | 4.56 85 | 3.99 87 | 27.8 98 | 1.98 86 | 22.6 85 | 33.1 75 | 20.5 86 | 14.6 108 | 32.7 94 | 10.7 105 | 5.96 125 | 1.68 126 | 21.9 126 | 14.1 87 | 33.4 81 | 19.9 86 |
SPSA-learn [13] | 94.9 | 1.77 103 | 7.72 88 | 0.90 94 | 11.0 105 | 33.2 103 | 9.40 107 | 17.3 105 | 34.2 107 | 22.7 108 | 11.0 108 | 32.2 104 | 10.9 108 | 26.1 104 | 34.9 94 | 31.8 112 | 12.8 97 | 34.2 100 | 11.9 110 | 0.00 1 | 0.03 85 | 0.00 1 | 25.5 114 | 39.4 105 | 39.6 115 |
LDOF [28] | 95.0 | 1.59 96 | 8.06 91 | 0.97 100 | 6.08 95 | 27.9 95 | 3.79 93 | 8.98 87 | 25.2 87 | 6.05 90 | 5.90 102 | 33.2 105 | 3.14 96 | 23.5 90 | 34.5 92 | 22.7 95 | 9.83 63 | 34.0 99 | 7.30 85 | 0.86 98 | 1.28 123 | 3.67 102 | 16.7 94 | 39.7 107 | 23.5 96 |
BlockOverlap [61] | 96.2 | 1.73 100 | 8.32 95 | 1.07 105 | 8.43 102 | 28.3 98 | 7.39 103 | 14.3 102 | 29.4 100 | 16.3 99 | 6.01 103 | 26.7 91 | 4.24 105 | 20.6 64 | 29.8 50 | 20.6 87 | 12.4 96 | 29.4 75 | 8.20 96 | 3.91 119 | 0.92 120 | 16.5 122 | 19.1 102 | 31.7 67 | 36.0 109 |
Filter Flow [19] | 96.7 | 1.97 108 | 10.2 109 | 1.14 108 | 8.79 103 | 33.9 105 | 5.66 98 | 18.8 107 | 35.7 108 | 26.2 113 | 21.9 115 | 42.4 112 | 22.0 115 | 27.9 110 | 36.8 102 | 35.0 115 | 13.2 102 | 32.6 93 | 8.11 95 | 0.05 46 | 0.02 30 | 0.37 53 | 17.5 98 | 33.0 76 | 25.2 99 |
HBpMotionGpu [43] | 97.2 | 2.47 110 | 11.8 111 | 1.09 106 | 11.4 107 | 35.3 107 | 10.0 109 | 20.3 112 | 38.5 112 | 26.3 114 | 5.67 100 | 26.6 89 | 3.51 101 | 24.1 93 | 34.8 93 | 26.1 101 | 10.9 82 | 30.7 81 | 7.04 79 | 0.27 66 | 0.05 89 | 0.89 62 | 19.3 104 | 37.1 97 | 32.8 107 |
GraphCuts [14] | 100.7 | 1.57 94 | 8.32 95 | 0.92 97 | 12.3 109 | 39.3 112 | 8.40 104 | 15.2 103 | 31.3 103 | 23.1 109 | 5.40 97 | 28.8 101 | 2.88 93 | 25.4 98 | 38.0 106 | 21.1 90 | 24.5 123 | 31.1 84 | 14.4 116 | 1.86 110 | 0.02 30 | 7.91 112 | 23.9 109 | 41.6 110 | 37.4 112 |
UnFlow [129] | 100.9 | 7.34 123 | 24.6 128 | 3.32 119 | 21.7 117 | 50.1 118 | 19.1 115 | 26.8 119 | 53.1 125 | 25.0 110 | 13.7 112 | 42.5 113 | 12.5 111 | 42.2 124 | 53.7 125 | 45.6 123 | 15.1 110 | 46.2 116 | 12.1 111 | 0.00 1 | 0.00 1 | 0.00 1 | 17.5 98 | 43.7 112 | 21.5 90 |
IAOF [50] | 101.0 | 1.77 103 | 8.80 101 | 0.98 101 | 11.2 106 | 32.5 102 | 9.32 106 | 19.8 110 | 35.7 108 | 20.2 103 | 17.5 113 | 37.6 109 | 19.8 113 | 23.7 91 | 35.0 95 | 22.3 93 | 18.1 116 | 40.2 110 | 10.9 106 | 0.56 86 | 0.02 30 | 2.17 91 | 24.8 112 | 37.8 101 | 43.9 117 |
IAOF2 [51] | 103.2 | 1.85 107 | 9.64 107 | 1.13 107 | 7.56 99 | 29.4 99 | 5.66 98 | 12.2 97 | 27.5 96 | 15.7 98 | 32.6 121 | 43.3 115 | 38.7 124 | 24.3 94 | 35.0 95 | 23.9 98 | 17.9 113 | 33.1 95 | 13.0 112 | 1.11 102 | 0.25 99 | 4.83 104 | 17.8 100 | 35.5 94 | 25.9 102 |
Black & Anandan [4] | 103.7 | 1.75 101 | 8.07 92 | 0.73 85 | 11.6 108 | 36.6 108 | 8.94 105 | 18.9 108 | 36.4 110 | 20.3 104 | 12.4 110 | 40.5 111 | 12.0 110 | 26.3 107 | 36.2 101 | 30.5 106 | 13.4 103 | 37.3 105 | 11.0 107 | 0.75 94 | 0.42 109 | 1.90 88 | 21.4 108 | 38.6 103 | 32.5 106 |
Nguyen [33] | 105.5 | 2.73 112 | 11.0 110 | 1.16 112 | 33.4 122 | 38.0 109 | 43.1 124 | 24.6 116 | 41.9 113 | 32.1 119 | 28.7 119 | 46.5 116 | 32.2 120 | 29.8 111 | 39.8 110 | 35.5 116 | 13.9 105 | 40.4 111 | 13.0 112 | 0.03 42 | 0.02 30 | 0.20 48 | 31.6 117 | 46.3 115 | 50.5 124 |
Modified CLG [34] | 106.7 | 2.46 109 | 12.2 112 | 1.37 113 | 10.5 104 | 33.6 104 | 9.99 108 | 20.2 111 | 37.9 111 | 27.9 116 | 9.52 107 | 38.0 110 | 7.95 107 | 27.6 109 | 38.6 108 | 31.7 110 | 11.2 87 | 37.6 106 | 8.53 99 | 0.70 93 | 0.24 98 | 3.33 100 | 24.7 111 | 45.8 114 | 38.5 114 |
2D-CLG [1] | 107.6 | 6.98 122 | 23.0 125 | 3.54 120 | 20.1 115 | 40.7 113 | 21.4 117 | 26.6 118 | 44.0 114 | 36.7 120 | 34.7 122 | 55.1 120 | 39.7 125 | 31.1 114 | 41.5 114 | 38.2 117 | 15.0 109 | 42.0 114 | 13.6 115 | 0.02 39 | 0.02 30 | 0.12 44 | 31.7 118 | 51.0 119 | 44.9 118 |
GroupFlow [9] | 110.2 | 3.39 114 | 16.8 119 | 1.37 113 | 23.0 118 | 51.6 120 | 21.5 118 | 20.7 113 | 45.1 117 | 22.3 106 | 5.67 100 | 27.3 95 | 3.50 100 | 34.6 118 | 51.5 123 | 22.0 92 | 22.4 120 | 47.9 118 | 25.4 124 | 0.55 85 | 0.47 111 | 1.70 84 | 25.2 113 | 47.9 116 | 33.5 108 |
SILK [79] | 110.6 | 3.45 115 | 15.8 117 | 2.61 117 | 19.0 113 | 44.9 115 | 19.5 116 | 23.5 115 | 44.1 115 | 26.6 115 | 12.0 109 | 42.7 114 | 11.1 109 | 35.3 119 | 46.3 120 | 44.8 121 | 18.0 115 | 49.4 120 | 14.5 117 | 1.53 107 | 0.00 1 | 5.00 105 | 32.1 121 | 50.8 118 | 47.1 121 |
Horn & Schunck [3] | 112.8 | 3.02 113 | 12.7 113 | 1.15 111 | 14.5 110 | 45.9 116 | 11.1 110 | 22.6 114 | 44.4 116 | 25.2 111 | 21.6 114 | 47.3 117 | 22.5 116 | 34.0 117 | 43.8 116 | 43.1 119 | 19.6 117 | 51.5 122 | 18.6 120 | 0.56 86 | 0.22 97 | 1.77 85 | 34.9 123 | 55.9 123 | 46.4 120 |
Heeger++ [104] | 113.9 | 3.74 117 | 16.1 118 | 1.49 115 | 23.6 119 | 64.4 128 | 14.1 111 | 36.0 125 | 49.4 123 | 37.3 123 | 38.6 126 | 67.3 126 | 38.6 123 | 46.7 127 | 58.2 127 | 50.9 126 | 36.6 127 | 68.1 129 | 34.5 128 | 0.41 80 | 0.00 1 | 1.87 86 | 31.8 119 | 51.3 120 | 37.0 110 |
TI-DOFE [24] | 114.2 | 7.50 124 | 18.0 120 | 10.6 126 | 41.8 126 | 54.1 123 | 49.7 126 | 31.9 123 | 54.7 128 | 39.7 125 | 41.8 127 | 61.8 123 | 48.6 128 | 35.5 121 | 45.7 119 | 45.0 122 | 21.9 119 | 52.6 123 | 21.7 122 | 0.25 63 | 0.00 1 | 1.31 73 | 43.7 126 | 61.4 126 | 58.6 126 |
FFV1MT [106] | 116.2 | 4.51 119 | 19.1 121 | 2.74 118 | 19.4 114 | 58.6 127 | 14.7 113 | 40.8 127 | 53.4 127 | 50.0 128 | 38.5 125 | 73.8 128 | 37.7 121 | 46.4 126 | 56.0 126 | 56.7 129 | 33.1 126 | 66.2 127 | 31.1 126 | 0.75 94 | 0.02 30 | 2.04 89 | 31.8 119 | 51.3 120 | 37.0 110 |
Periodicity [78] | 117.5 | 6.73 121 | 29.6 131 | 3.88 123 | 24.0 120 | 52.2 121 | 25.5 120 | 36.6 126 | 47.1 120 | 40.1 126 | 23.0 116 | 60.3 122 | 20.8 114 | 53.1 129 | 69.7 129 | 49.1 125 | 36.9 128 | 67.0 128 | 33.4 127 | 0.54 84 | 0.02 30 | 7.78 111 | 34.7 122 | 64.9 128 | 46.1 119 |
Adaptive flow [45] | 118.1 | 4.48 118 | 15.3 116 | 1.90 116 | 37.1 124 | 47.9 117 | 40.5 122 | 28.1 120 | 45.1 117 | 37.9 124 | 23.3 117 | 53.8 119 | 24.8 117 | 30.1 113 | 41.4 113 | 28.5 104 | 22.6 121 | 46.3 117 | 15.6 118 | 17.3 127 | 5.51 128 | 58.1 127 | 26.0 115 | 41.5 109 | 40.5 116 |
SLK [47] | 119.4 | 8.22 129 | 24.0 127 | 12.3 127 | 41.4 125 | 57.7 126 | 50.8 127 | 29.7 122 | 53.3 126 | 36.7 120 | 52.4 129 | 57.7 121 | 61.8 129 | 42.6 125 | 52.1 124 | 54.9 127 | 23.9 122 | 54.4 125 | 24.4 123 | 3.11 118 | 0.00 1 | 7.07 110 | 45.8 128 | 61.9 127 | 61.9 127 |
PGAM+LK [55] | 122.4 | 7.83 127 | 22.3 124 | 13.7 128 | 29.1 121 | 54.2 124 | 31.3 121 | 25.6 117 | 48.1 122 | 29.9 118 | 29.7 120 | 68.4 127 | 28.3 119 | 38.2 122 | 50.8 122 | 43.0 118 | 25.1 124 | 56.4 126 | 21.4 121 | 6.54 126 | 0.57 113 | 19.1 123 | 38.5 124 | 60.8 125 | 51.7 125 |
FOLKI [16] | 122.8 | 5.65 120 | 23.4 126 | 4.60 124 | 35.1 123 | 52.9 122 | 42.6 123 | 28.3 121 | 52.7 124 | 29.6 117 | 24.1 118 | 53.7 118 | 27.7 118 | 38.9 123 | 49.3 121 | 47.8 124 | 25.3 125 | 54.3 124 | 27.7 125 | 5.73 124 | 1.38 124 | 20.1 125 | 43.9 127 | 60.5 124 | 62.2 128 |
HCIC-L [99] | 123.0 | 8.04 128 | 19.9 122 | 3.64 121 | 56.4 129 | 56.0 125 | 70.0 129 | 40.9 128 | 45.5 119 | 62.3 129 | 38.3 124 | 62.5 124 | 38.2 122 | 35.3 119 | 45.3 118 | 32.0 113 | 20.7 118 | 38.1 108 | 18.5 119 | 26.5 129 | 13.0 129 | 59.2 129 | 40.6 125 | 51.8 122 | 48.7 123 |
Pyramid LK [2] | 125.4 | 7.59 125 | 14.5 115 | 15.4 129 | 47.0 127 | 50.5 119 | 58.8 128 | 32.1 124 | 47.7 121 | 42.4 127 | 36.1 123 | 62.9 125 | 41.1 126 | 48.9 128 | 61.1 128 | 55.0 128 | 41.7 129 | 50.3 121 | 40.3 129 | 4.64 123 | 2.07 127 | 16.3 121 | 56.9 129 | 71.9 129 | 77.2 129 |
AdaConv-v1 [126] | 130.0 | 25.9 130 | 27.4 129 | 29.8 130 | 96.8 130 | 97.6 130 | 95.4 130 | 93.0 130 | 90.8 130 | 99.0 130 | 88.2 130 | 85.6 130 | 91.5 130 | 97.0 130 | 98.5 130 | 88.6 130 | 86.2 130 | 81.3 130 | 83.9 130 | 64.9 130 | 56.4 130 | 97.3 130 | 100.0 130 | 99.9 130 | 99.9 130 |
SepConv-v1 [127] | 130.0 | 25.9 130 | 27.4 129 | 29.8 130 | 96.8 130 | 97.6 130 | 95.4 130 | 93.0 130 | 90.8 130 | 99.0 130 | 88.2 130 | 85.6 130 | 91.5 130 | 97.0 130 | 98.5 130 | 88.6 130 | 86.2 130 | 81.3 130 | 83.9 130 | 64.9 130 | 56.4 130 | 97.3 130 | 100.0 130 | 99.9 130 | 99.9 130 |
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. |