Optical flow evaluation results |
Statistics:
Average
SD
R2.5
R5.0
R10.0
A50
A75
A95
Error type: endpoint angle interpolation normalized interpolation |
A75 angle error |
avg. |
Army (Hidden texture) GT im0 im1 |
Mequon (Hidden texture) GT im0 im1 |
Schefflera (Hidden texture) GT im0 im1 |
Wooden (Hidden texture) GT im0 im1 |
Grove (Synthetic) GT im0 im1 |
Urban (Synthetic) GT im0 im1 |
Yosemite (Synthetic) GT im0 im1 |
Teddy (Stereo) GT im0 im1 | ||||||||||||||||
rank | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | |
NNF-Local [87] | 14.9 | 2.17 4 | 5.35 5 | 1.92 4 | 1.56 4 | 6.39 19 | 1.67 9 | 1.51 7 | 3.67 8 | 1.61 7 | 1.20 34 | 4.36 16 | 1.02 36 | 2.30 4 | 3.18 3 | 1.73 6 | 2.16 37 | 6.35 8 | 2.39 61 | 2.54 20 | 4.18 12 | 2.17 25 | 0.81 9 | 1.50 16 | 0.67 4 |
NN-field [71] | 15.2 | 2.31 8 | 5.94 13 | 1.98 5 | 1.83 21 | 7.21 35 | 1.97 27 | 1.54 8 | 3.55 7 | 1.68 11 | 0.96 9 | 3.04 4 | 0.75 5 | 2.30 4 | 3.25 4 | 1.69 5 | 1.72 17 | 3.81 1 | 1.65 6 | 3.12 47 | 4.76 49 | 2.60 46 | 0.79 6 | 1.58 24 | 0.59 2 |
TC/T-Flow [76] | 18.0 | 2.06 1 | 7.42 29 | 1.55 1 | 1.61 10 | 6.77 28 | 1.52 3 | 1.46 5 | 4.33 16 | 1.56 4 | 0.88 2 | 6.96 39 | 0.69 1 | 2.91 19 | 4.35 25 | 1.88 10 | 1.47 4 | 6.12 7 | 1.61 5 | 2.22 5 | 3.98 3 | 4.00 86 | 1.06 38 | 1.99 42 | 1.19 48 |
ComponentFusion [96] | 18.4 | 2.12 3 | 6.18 15 | 1.80 3 | 1.67 13 | 4.82 4 | 1.91 22 | 1.34 2 | 4.01 10 | 1.47 2 | 0.88 2 | 4.74 17 | 0.72 3 | 2.99 23 | 4.35 25 | 1.96 13 | 2.03 29 | 11.2 68 | 1.96 26 | 2.90 40 | 4.62 44 | 1.90 14 | 0.93 25 | 1.56 23 | 0.89 15 |
ALD-Flow [66] | 20.9 | 2.26 6 | 5.81 10 | 2.07 6 | 1.56 4 | 5.71 13 | 1.66 7 | 1.46 5 | 4.64 22 | 1.66 10 | 0.95 8 | 7.15 43 | 0.80 10 | 3.18 32 | 4.57 34 | 1.76 7 | 1.51 5 | 7.63 20 | 1.60 4 | 2.61 23 | 4.27 17 | 3.90 82 | 1.04 36 | 2.22 53 | 1.18 45 |
WLIF-Flow [93] | 21.3 | 2.63 18 | 5.51 6 | 2.41 16 | 2.14 42 | 7.08 32 | 2.32 41 | 1.63 11 | 4.19 13 | 1.84 17 | 1.05 14 | 4.21 13 | 0.88 22 | 2.92 21 | 4.37 27 | 2.30 28 | 2.02 28 | 7.21 16 | 1.89 22 | 2.73 30 | 4.27 17 | 2.78 52 | 0.84 10 | 1.37 5 | 0.83 10 |
nLayers [57] | 22.7 | 2.33 9 | 5.14 3 | 2.17 9 | 2.75 87 | 7.22 37 | 3.07 88 | 1.69 19 | 4.04 11 | 2.21 60 | 0.88 2 | 2.83 2 | 0.70 2 | 2.08 2 | 3.25 4 | 1.30 1 | 1.88 20 | 6.06 6 | 1.85 13 | 2.89 39 | 4.59 39 | 2.28 30 | 0.92 22 | 1.48 14 | 0.94 25 |
OFLAF [77] | 23.1 | 2.77 34 | 5.70 8 | 2.49 18 | 1.76 16 | 5.35 9 | 1.84 20 | 1.54 8 | 2.69 2 | 1.72 13 | 1.30 40 | 3.55 9 | 1.12 49 | 2.30 4 | 3.62 7 | 1.64 4 | 2.23 42 | 5.93 3 | 2.06 36 | 2.81 34 | 4.29 20 | 2.99 57 | 1.15 47 | 1.69 29 | 1.18 45 |
RNLOD-Flow [121] | 23.7 | 2.19 5 | 4.96 2 | 2.19 10 | 1.79 18 | 7.20 34 | 1.76 15 | 1.42 3 | 4.31 15 | 1.56 4 | 0.97 10 | 3.42 8 | 0.84 15 | 2.75 14 | 4.16 16 | 2.01 15 | 1.86 19 | 7.24 17 | 1.95 25 | 4.02 81 | 6.41 96 | 4.48 101 | 0.90 17 | 1.51 17 | 0.85 11 |
MDP-Flow2 [68] | 23.8 | 3.08 39 | 6.23 16 | 2.73 42 | 1.55 3 | 4.80 3 | 1.64 6 | 1.63 11 | 3.27 4 | 1.61 7 | 1.37 51 | 5.15 21 | 1.15 54 | 2.91 19 | 4.19 19 | 2.20 20 | 2.24 44 | 6.43 9 | 2.17 45 | 2.62 24 | 4.35 24 | 1.88 13 | 1.08 40 | 1.66 28 | 0.95 28 |
OAR-Flow [125] | 24.8 | 2.55 12 | 7.57 30 | 2.36 14 | 1.81 19 | 7.94 43 | 1.94 26 | 1.72 26 | 8.40 57 | 1.95 24 | 0.94 7 | 5.90 31 | 0.79 8 | 3.49 40 | 4.83 42 | 1.84 9 | 1.16 2 | 7.84 25 | 1.22 2 | 1.93 2 | 3.63 1 | 2.26 28 | 1.10 43 | 2.16 49 | 1.34 56 |
AGIF+OF [85] | 26.1 | 2.60 13 | 6.24 17 | 2.45 17 | 2.49 68 | 9.30 62 | 2.63 68 | 1.68 17 | 4.79 28 | 2.08 45 | 1.05 14 | 4.34 15 | 0.81 11 | 2.77 16 | 4.05 13 | 2.10 18 | 1.92 23 | 7.48 19 | 1.74 11 | 2.69 27 | 4.48 31 | 2.69 49 | 0.87 11 | 1.40 6 | 0.95 28 |
Layers++ [37] | 26.4 | 2.70 29 | 6.40 19 | 2.83 49 | 2.33 59 | 6.62 25 | 2.54 65 | 1.65 14 | 3.24 3 | 2.02 34 | 0.92 5 | 2.48 1 | 0.75 5 | 2.12 3 | 3.11 1 | 1.50 3 | 2.06 32 | 8.25 27 | 1.94 24 | 3.59 69 | 5.41 71 | 3.21 60 | 0.89 14 | 1.32 4 | 0.90 18 |
LME [70] | 27.6 | 2.90 36 | 5.83 11 | 2.30 13 | 1.60 7 | 4.45 2 | 1.74 14 | 1.71 23 | 4.14 12 | 2.00 30 | 1.35 46 | 6.61 34 | 1.13 50 | 3.07 28 | 4.32 24 | 2.57 40 | 2.08 34 | 8.09 26 | 2.00 32 | 2.78 33 | 4.53 36 | 2.29 31 | 1.06 38 | 1.74 31 | 0.96 32 |
HAST [109] | 27.9 | 2.09 2 | 4.28 1 | 1.69 2 | 1.60 7 | 5.31 8 | 1.55 4 | 1.28 1 | 2.09 1 | 1.40 1 | 0.92 5 | 3.39 7 | 0.78 7 | 2.01 1 | 3.14 2 | 1.48 2 | 2.40 56 | 8.62 34 | 2.41 63 | 4.08 83 | 7.33 115 | 7.69 117 | 1.23 54 | 1.59 26 | 1.73 71 |
TC-Flow [46] | 28.8 | 2.45 11 | 6.60 21 | 2.39 15 | 1.25 1 | 5.24 6 | 1.32 1 | 1.45 4 | 4.40 19 | 1.50 3 | 1.15 31 | 8.13 50 | 1.04 38 | 3.22 33 | 4.77 40 | 2.06 17 | 1.94 24 | 8.65 35 | 2.09 40 | 2.33 15 | 4.51 33 | 3.82 80 | 1.24 55 | 2.18 51 | 1.50 68 |
PH-Flow [101] | 29.9 | 2.62 15 | 7.58 32 | 2.53 23 | 2.13 39 | 8.78 52 | 2.37 44 | 1.70 20 | 4.39 18 | 2.06 41 | 1.08 17 | 7.06 42 | 0.85 16 | 2.72 11 | 3.91 11 | 2.04 16 | 2.06 32 | 8.33 28 | 1.96 26 | 3.48 65 | 4.62 44 | 4.03 87 | 0.90 17 | 1.41 9 | 0.88 13 |
Classic+CPF [83] | 30.2 | 2.65 24 | 7.22 28 | 2.53 23 | 2.37 61 | 9.14 59 | 2.51 61 | 1.67 16 | 5.05 32 | 2.03 36 | 1.01 11 | 5.38 24 | 0.79 8 | 2.90 18 | 4.17 17 | 2.33 30 | 1.88 20 | 8.44 30 | 1.70 8 | 3.19 50 | 4.60 41 | 3.72 73 | 0.92 22 | 1.40 6 | 0.95 28 |
Sparse-NonSparse [56] | 30.4 | 2.62 15 | 7.58 32 | 2.60 31 | 2.18 48 | 8.74 50 | 2.46 57 | 1.68 17 | 4.86 29 | 2.00 30 | 1.04 12 | 7.97 48 | 0.81 11 | 3.13 30 | 4.45 31 | 2.42 35 | 1.98 27 | 8.53 32 | 1.87 17 | 3.13 48 | 4.32 23 | 3.51 67 | 0.88 12 | 1.41 9 | 0.91 19 |
NNF-EAC [103] | 30.5 | 3.07 38 | 6.73 23 | 2.70 41 | 1.62 12 | 5.30 7 | 1.72 12 | 1.71 23 | 3.89 9 | 1.79 15 | 1.36 49 | 6.36 33 | 1.15 54 | 3.02 24 | 4.37 27 | 2.28 26 | 2.44 59 | 6.90 12 | 2.28 56 | 2.90 40 | 4.51 33 | 2.19 26 | 1.12 44 | 1.83 36 | 0.98 34 |
FC-2Layers-FF [74] | 30.7 | 2.63 18 | 5.87 12 | 2.68 39 | 2.19 52 | 8.07 45 | 2.39 49 | 1.65 14 | 3.42 5 | 2.05 38 | 1.10 23 | 3.11 5 | 0.88 22 | 2.57 7 | 3.49 6 | 2.30 28 | 2.26 46 | 7.68 22 | 2.17 45 | 3.70 71 | 5.18 66 | 3.73 74 | 0.90 17 | 1.41 9 | 0.93 23 |
IROF++ [58] | 31.2 | 2.66 25 | 6.82 24 | 2.58 29 | 2.17 46 | 9.07 57 | 2.41 50 | 1.75 32 | 5.03 31 | 2.06 41 | 1.11 25 | 7.65 45 | 0.90 28 | 2.92 21 | 4.18 18 | 2.25 24 | 2.13 35 | 9.85 50 | 1.98 29 | 2.53 19 | 4.53 36 | 1.43 3 | 0.97 33 | 1.58 24 | 0.94 25 |
COFM [59] | 33.2 | 2.28 7 | 7.19 26 | 2.08 8 | 1.78 17 | 6.57 22 | 1.93 24 | 1.56 10 | 5.33 37 | 2.19 58 | 0.86 1 | 4.90 18 | 0.73 4 | 3.76 49 | 4.80 41 | 3.87 86 | 2.03 29 | 7.67 21 | 1.72 9 | 2.76 31 | 4.21 13 | 4.04 89 | 1.62 77 | 1.87 38 | 2.04 82 |
FESL [72] | 33.4 | 2.63 18 | 5.15 4 | 2.77 46 | 2.62 80 | 9.27 61 | 2.73 74 | 1.72 26 | 4.77 26 | 2.07 44 | 1.15 31 | 3.36 6 | 0.98 34 | 2.75 14 | 3.93 12 | 2.20 20 | 1.95 25 | 7.12 14 | 1.98 29 | 3.40 60 | 5.71 80 | 2.89 54 | 0.88 12 | 1.55 20 | 0.87 12 |
Efficient-NL [60] | 33.9 | 2.38 10 | 5.67 7 | 2.07 6 | 2.45 67 | 8.54 47 | 2.55 66 | 1.64 13 | 4.63 21 | 1.95 24 | 1.04 12 | 5.76 29 | 0.81 11 | 2.74 12 | 4.14 15 | 1.94 12 | 2.87 72 | 8.58 33 | 2.23 49 | 3.30 55 | 5.12 63 | 3.05 58 | 1.15 47 | 1.83 36 | 1.19 48 |
LSM [39] | 35.6 | 2.60 13 | 7.70 36 | 2.61 33 | 2.19 52 | 8.77 51 | 2.44 55 | 1.70 20 | 4.78 27 | 2.06 41 | 1.08 17 | 8.13 50 | 0.86 18 | 3.05 26 | 4.30 22 | 2.47 36 | 2.18 39 | 8.66 38 | 2.08 38 | 3.56 68 | 4.68 48 | 3.74 75 | 0.90 17 | 1.43 12 | 0.93 23 |
PMMST [114] | 35.7 | 3.55 60 | 6.48 20 | 3.33 69 | 2.18 48 | 6.49 21 | 2.47 60 | 1.93 44 | 4.28 14 | 2.09 46 | 1.60 68 | 2.84 3 | 1.43 72 | 2.60 8 | 3.72 9 | 1.91 11 | 2.22 41 | 6.05 5 | 2.13 43 | 2.70 29 | 4.48 31 | 2.08 20 | 1.18 53 | 1.91 40 | 1.09 42 |
Classic+NL [31] | 36.5 | 2.63 18 | 7.57 30 | 2.64 37 | 2.18 48 | 9.04 55 | 2.41 50 | 1.71 23 | 4.70 23 | 2.09 46 | 1.08 17 | 7.69 46 | 0.88 22 | 3.04 25 | 4.31 23 | 2.41 34 | 2.27 47 | 8.65 35 | 2.09 40 | 3.79 74 | 5.12 63 | 3.81 78 | 0.89 14 | 1.44 13 | 0.89 15 |
FMOF [94] | 37.1 | 2.71 31 | 6.71 22 | 2.62 35 | 2.64 81 | 9.19 60 | 2.77 75 | 1.73 29 | 4.37 17 | 2.22 61 | 1.06 16 | 5.18 22 | 0.81 11 | 2.89 17 | 4.25 21 | 2.40 33 | 2.31 50 | 7.76 24 | 1.96 26 | 3.35 58 | 4.88 58 | 3.81 78 | 0.93 25 | 1.55 20 | 0.92 21 |
Ramp [62] | 37.5 | 2.64 23 | 7.64 35 | 2.56 27 | 2.20 55 | 8.90 54 | 2.46 57 | 1.73 29 | 4.74 25 | 2.09 46 | 1.11 25 | 6.81 38 | 0.88 22 | 3.07 28 | 4.46 32 | 2.38 32 | 2.28 48 | 8.52 31 | 2.15 44 | 3.40 60 | 4.25 14 | 4.16 96 | 0.94 27 | 1.53 19 | 0.97 33 |
ProbFlowFields [128] | 38.4 | 3.31 48 | 13.8 82 | 2.85 51 | 2.03 31 | 6.67 27 | 2.23 36 | 1.89 42 | 6.06 42 | 2.29 65 | 1.23 37 | 5.14 20 | 0.95 31 | 3.70 46 | 5.08 45 | 2.29 27 | 1.60 9 | 7.44 18 | 1.85 13 | 2.43 18 | 4.30 21 | 2.45 41 | 1.29 57 | 2.32 55 | 1.38 60 |
2DHMM-SAS [92] | 40.0 | 2.62 15 | 7.61 34 | 2.53 23 | 2.17 46 | 10.1 70 | 2.38 47 | 1.83 37 | 6.16 45 | 2.10 50 | 1.09 21 | 8.07 49 | 0.86 18 | 3.06 27 | 4.42 30 | 2.37 31 | 2.16 37 | 9.30 44 | 2.01 33 | 3.50 66 | 4.64 46 | 4.14 95 | 0.96 31 | 1.64 27 | 0.99 37 |
Adaptive [20] | 40.0 | 2.63 18 | 8.08 40 | 2.23 11 | 2.18 48 | 8.66 49 | 2.27 39 | 2.04 51 | 9.57 64 | 2.09 46 | 1.12 28 | 10.6 68 | 0.87 20 | 4.47 91 | 5.30 60 | 4.44 99 | 1.55 6 | 8.65 35 | 1.43 3 | 3.32 56 | 5.51 73 | 2.14 24 | 0.77 5 | 1.52 18 | 0.75 8 |
PMF [73] | 41.0 | 3.22 45 | 6.34 18 | 2.60 31 | 1.95 27 | 6.66 26 | 1.92 23 | 1.85 38 | 4.58 20 | 1.83 16 | 1.62 69 | 4.27 14 | 1.37 67 | 2.61 9 | 3.74 10 | 1.83 8 | 3.18 80 | 9.94 53 | 3.34 84 | 5.29 107 | 8.26 121 | 5.58 108 | 0.68 3 | 1.21 3 | 0.67 4 |
SVFilterOh [111] | 41.0 | 3.48 57 | 5.72 9 | 3.45 73 | 2.38 63 | 6.06 16 | 2.41 50 | 1.93 44 | 3.46 6 | 2.05 38 | 1.53 64 | 3.71 10 | 1.27 63 | 2.65 10 | 4.10 14 | 2.00 14 | 2.47 61 | 7.08 13 | 2.28 56 | 4.60 93 | 7.10 108 | 5.92 109 | 0.73 4 | 1.16 2 | 0.70 7 |
S2D-Matching [84] | 42.1 | 2.69 28 | 7.86 38 | 2.68 39 | 2.19 52 | 9.11 58 | 2.44 55 | 1.77 33 | 6.11 44 | 2.04 37 | 1.13 29 | 5.86 30 | 0.91 29 | 3.17 31 | 4.41 29 | 2.50 37 | 2.41 57 | 9.09 41 | 2.25 53 | 4.00 80 | 5.16 65 | 4.07 90 | 0.91 21 | 1.40 6 | 0.95 28 |
TV-L1-MCT [64] | 42.3 | 2.68 27 | 6.83 25 | 2.61 33 | 2.68 82 | 10.3 73 | 2.80 78 | 1.78 34 | 5.24 34 | 2.24 62 | 1.13 29 | 5.44 26 | 0.87 20 | 3.39 38 | 4.69 39 | 2.96 57 | 2.45 60 | 9.14 42 | 2.31 59 | 2.64 25 | 4.37 26 | 2.04 18 | 1.08 40 | 1.74 31 | 1.36 58 |
SimpleFlow [49] | 43.9 | 2.74 32 | 8.28 44 | 2.73 42 | 2.50 69 | 9.73 68 | 2.83 83 | 1.89 42 | 6.81 50 | 2.35 67 | 1.11 25 | 10.4 65 | 0.89 27 | 3.27 34 | 4.47 33 | 2.63 42 | 3.03 75 | 8.91 40 | 2.39 61 | 3.10 46 | 4.25 14 | 2.76 51 | 0.89 14 | 1.49 15 | 0.89 15 |
Occlusion-TV-L1 [63] | 44.5 | 3.15 41 | 8.42 46 | 2.50 19 | 2.03 31 | 7.42 39 | 2.14 33 | 2.24 62 | 9.79 66 | 2.16 56 | 1.35 46 | 9.59 58 | 1.11 45 | 4.10 71 | 5.77 88 | 3.22 67 | 1.68 16 | 9.21 43 | 2.08 38 | 2.69 27 | 4.59 39 | 1.70 9 | 1.05 37 | 2.36 58 | 0.98 34 |
IROF-TV [53] | 44.8 | 2.89 35 | 8.67 49 | 2.81 48 | 2.25 56 | 9.54 66 | 2.51 61 | 1.79 36 | 5.50 39 | 2.15 55 | 1.53 64 | 11.5 73 | 1.27 63 | 3.33 35 | 4.62 35 | 2.85 50 | 2.78 67 | 13.5 88 | 2.57 67 | 2.15 4 | 4.14 9 | 1.37 2 | 0.94 27 | 1.55 20 | 0.94 25 |
MDP-Flow [26] | 45.0 | 3.14 40 | 9.81 55 | 2.83 49 | 2.06 34 | 6.10 17 | 2.43 53 | 1.87 40 | 6.10 43 | 2.10 50 | 1.44 55 | 8.90 54 | 1.15 54 | 3.37 36 | 4.62 35 | 2.54 39 | 2.35 53 | 10.4 58 | 2.23 49 | 2.88 38 | 4.83 53 | 1.94 16 | 1.27 56 | 2.62 62 | 1.09 42 |
Correlation Flow [75] | 45.2 | 3.18 43 | 7.85 37 | 2.85 51 | 1.74 14 | 5.77 14 | 1.69 11 | 1.94 46 | 5.25 35 | 1.70 12 | 1.47 59 | 5.67 28 | 1.26 62 | 3.66 43 | 5.20 54 | 2.53 38 | 3.06 76 | 9.57 46 | 3.10 80 | 3.42 63 | 4.94 59 | 4.03 87 | 1.17 51 | 1.80 33 | 1.16 44 |
AggregFlow [97] | 47.7 | 3.29 47 | 8.49 47 | 3.14 61 | 2.70 84 | 12.2 90 | 2.68 71 | 2.32 66 | 9.03 59 | 2.88 83 | 1.44 55 | 4.19 12 | 1.25 61 | 3.48 39 | 5.09 46 | 2.19 19 | 1.55 6 | 5.36 2 | 1.68 7 | 2.56 22 | 4.78 51 | 1.77 11 | 1.54 73 | 2.15 47 | 2.16 85 |
Aniso-Texture [82] | 49.1 | 2.75 33 | 6.00 14 | 3.09 58 | 2.13 39 | 5.64 11 | 2.52 64 | 1.78 34 | 6.80 49 | 2.20 59 | 1.08 17 | 4.01 11 | 0.92 30 | 4.08 68 | 5.44 70 | 3.26 71 | 2.31 50 | 11.8 73 | 2.26 54 | 5.87 114 | 8.09 120 | 4.24 97 | 0.80 8 | 1.70 30 | 0.67 4 |
OFH [38] | 49.6 | 3.60 68 | 10.3 58 | 3.80 83 | 1.58 6 | 7.05 30 | 1.66 7 | 1.70 20 | 9.23 60 | 1.58 6 | 1.19 33 | 10.1 60 | 1.08 42 | 3.98 54 | 5.22 56 | 3.57 76 | 2.80 68 | 12.6 80 | 3.12 81 | 2.30 13 | 4.60 41 | 2.35 33 | 1.41 69 | 2.90 74 | 1.75 72 |
CostFilter [40] | 50.5 | 3.59 65 | 8.35 45 | 3.26 65 | 2.12 37 | 6.60 23 | 2.16 34 | 2.00 50 | 5.56 40 | 2.01 32 | 2.04 84 | 7.05 41 | 1.89 85 | 2.74 12 | 3.70 8 | 2.27 25 | 3.29 81 | 10.3 57 | 3.33 83 | 5.22 105 | 9.79 126 | 6.16 111 | 0.38 1 | 1.08 1 | 0.35 1 |
Classic++ [32] | 50.9 | 2.66 25 | 8.18 43 | 2.65 38 | 2.13 39 | 7.96 44 | 2.43 53 | 1.85 38 | 9.39 63 | 2.10 50 | 1.09 21 | 10.4 65 | 0.88 22 | 3.97 52 | 5.60 81 | 2.89 53 | 2.36 54 | 13.6 91 | 2.10 42 | 4.03 82 | 5.20 67 | 4.33 98 | 0.99 35 | 2.05 44 | 0.92 21 |
DeepFlow2 [108] | 51.1 | 3.44 53 | 12.6 72 | 3.36 71 | 1.98 28 | 8.60 48 | 2.10 31 | 2.38 69 | 11.1 70 | 2.60 75 | 1.34 43 | 14.6 83 | 1.11 45 | 3.53 41 | 5.09 46 | 2.23 23 | 1.66 15 | 9.90 52 | 1.77 12 | 2.76 31 | 4.06 6 | 3.40 64 | 1.72 81 | 3.21 84 | 2.13 83 |
S2F-IF [123] | 52.3 | 3.54 59 | 19.2 108 | 2.58 29 | 2.41 65 | 10.4 74 | 2.59 67 | 2.57 75 | 10.6 69 | 2.59 73 | 1.29 39 | 10.5 67 | 0.98 34 | 4.07 66 | 5.38 65 | 2.90 54 | 1.65 14 | 9.94 53 | 1.85 13 | 2.27 8 | 4.26 16 | 2.35 33 | 1.33 61 | 2.49 61 | 1.32 53 |
FlowFields+ [130] | 53.5 | 3.58 63 | 19.0 104 | 2.56 27 | 2.57 74 | 10.8 77 | 2.78 76 | 2.72 81 | 11.9 76 | 2.78 81 | 1.32 41 | 10.9 70 | 1.03 37 | 3.97 52 | 5.34 62 | 2.74 43 | 1.64 13 | 9.79 49 | 1.87 17 | 2.26 7 | 4.30 21 | 2.35 33 | 1.35 62 | 2.62 62 | 1.34 56 |
RFlow [90] | 53.8 | 3.62 70 | 9.91 56 | 3.53 76 | 1.83 21 | 5.50 10 | 1.93 24 | 2.14 57 | 9.57 64 | 1.86 19 | 1.32 41 | 6.75 35 | 1.14 52 | 3.98 54 | 5.35 63 | 3.24 68 | 2.39 55 | 11.7 72 | 2.24 51 | 3.45 64 | 4.60 41 | 3.63 68 | 1.64 79 | 2.90 74 | 1.91 78 |
CPM-Flow [116] | 54.2 | 3.47 55 | 19.0 104 | 2.52 20 | 2.59 76 | 11.0 82 | 2.82 80 | 2.56 73 | 11.3 72 | 2.75 77 | 1.34 43 | 15.7 87 | 1.06 39 | 4.02 59 | 5.42 67 | 2.78 45 | 1.59 8 | 9.39 45 | 1.87 17 | 2.30 13 | 4.17 11 | 2.36 36 | 1.35 62 | 2.69 68 | 1.39 62 |
PGM-C [120] | 54.7 | 3.48 57 | 19.0 104 | 2.52 20 | 2.59 76 | 10.8 77 | 2.82 80 | 2.59 76 | 11.8 75 | 2.75 77 | 1.34 43 | 16.5 91 | 1.06 39 | 4.03 61 | 5.46 72 | 2.78 45 | 1.60 9 | 9.63 47 | 1.88 20 | 2.28 9 | 3.98 3 | 2.36 36 | 1.37 64 | 2.69 68 | 1.45 63 |
EpicFlow [102] | 56.4 | 3.47 55 | 18.9 102 | 2.52 20 | 2.59 76 | 10.9 80 | 2.82 80 | 2.64 78 | 14.2 85 | 2.75 77 | 1.35 46 | 15.5 85 | 1.06 39 | 4.04 63 | 5.48 73 | 2.88 52 | 1.62 12 | 9.70 48 | 1.91 23 | 2.28 9 | 4.08 8 | 2.36 36 | 1.39 68 | 2.71 70 | 1.51 69 |
MLDP_OF [89] | 56.9 | 4.16 86 | 10.4 59 | 4.04 85 | 2.04 33 | 6.61 24 | 2.04 29 | 2.36 68 | 6.60 48 | 2.05 38 | 1.43 53 | 5.65 27 | 1.18 58 | 3.75 48 | 4.86 43 | 2.96 57 | 2.96 73 | 8.71 39 | 3.47 86 | 4.20 86 | 5.51 73 | 7.24 114 | 1.16 50 | 1.87 38 | 1.21 50 |
FlowFields [110] | 56.9 | 3.56 62 | 19.0 104 | 2.54 26 | 2.57 74 | 10.6 76 | 2.79 77 | 2.72 81 | 11.7 74 | 2.76 80 | 1.42 52 | 10.9 70 | 1.13 50 | 4.08 68 | 5.43 69 | 2.92 55 | 1.60 9 | 10.5 60 | 1.86 16 | 2.28 9 | 4.35 24 | 2.46 43 | 1.38 67 | 2.62 62 | 1.36 58 |
TV-L1-improved [17] | 57.2 | 2.70 29 | 9.05 51 | 2.29 12 | 1.85 23 | 7.06 31 | 1.97 27 | 1.94 46 | 9.28 62 | 1.90 23 | 1.10 23 | 8.96 55 | 0.85 16 | 4.07 66 | 5.60 81 | 2.75 44 | 5.44 110 | 17.3 102 | 6.29 113 | 4.75 100 | 6.82 101 | 4.73 104 | 1.13 45 | 2.68 67 | 1.06 41 |
BriefMatch [124] | 57.6 | 3.03 37 | 7.96 39 | 2.73 42 | 1.75 15 | 6.88 29 | 1.73 13 | 1.72 26 | 4.70 23 | 1.73 14 | 1.51 63 | 5.38 24 | 1.39 70 | 4.01 57 | 5.27 58 | 3.72 81 | 5.57 111 | 15.9 96 | 6.02 112 | 4.65 94 | 6.85 102 | 8.98 121 | 0.95 29 | 2.30 54 | 1.77 73 |
Kuang [131] | 57.8 | 3.27 46 | 18.0 97 | 2.62 35 | 2.32 57 | 12.0 88 | 2.37 44 | 2.45 72 | 13.3 84 | 2.42 68 | 1.21 36 | 10.3 63 | 0.97 33 | 4.22 77 | 5.64 83 | 3.06 62 | 2.30 49 | 12.5 78 | 2.50 66 | 2.23 6 | 4.15 10 | 2.36 36 | 1.30 60 | 2.79 71 | 1.47 65 |
Steered-L1 [118] | 58.2 | 3.21 44 | 8.15 42 | 3.11 60 | 1.39 2 | 4.13 1 | 1.51 2 | 1.73 29 | 5.20 33 | 1.65 9 | 1.28 38 | 10.3 63 | 1.11 45 | 4.05 65 | 5.35 63 | 3.55 74 | 3.10 77 | 12.6 80 | 2.59 68 | 6.15 119 | 6.90 103 | 13.1 126 | 1.73 82 | 3.04 82 | 2.39 89 |
CombBMOF [113] | 59.0 | 3.55 60 | 11.6 67 | 2.79 47 | 2.52 70 | 7.21 35 | 2.51 61 | 1.88 41 | 5.63 41 | 1.84 17 | 1.67 74 | 11.2 72 | 1.51 77 | 3.68 44 | 4.62 35 | 3.07 63 | 4.08 93 | 11.4 70 | 4.79 103 | 4.72 98 | 6.58 99 | 3.82 80 | 0.92 22 | 1.82 35 | 0.88 13 |
Sparse Occlusion [54] | 60.5 | 3.36 49 | 8.08 40 | 2.90 53 | 2.61 79 | 7.68 40 | 3.01 87 | 2.10 53 | 6.40 47 | 2.13 53 | 1.45 57 | 6.80 36 | 1.14 52 | 4.01 57 | 5.31 61 | 2.81 48 | 2.55 63 | 10.4 58 | 2.21 48 | 6.70 122 | 8.26 121 | 4.34 99 | 1.15 47 | 2.08 45 | 0.99 37 |
DeepFlow [86] | 61.0 | 3.94 78 | 12.7 73 | 4.14 89 | 2.12 37 | 9.06 56 | 2.28 40 | 2.84 85 | 12.5 80 | 3.16 88 | 1.68 75 | 15.6 86 | 1.44 74 | 3.58 42 | 5.10 48 | 2.20 20 | 1.78 18 | 11.1 66 | 1.88 20 | 2.65 26 | 4.07 7 | 3.40 64 | 2.08 96 | 3.59 97 | 3.09 99 |
EPPM w/o HM [88] | 61.8 | 4.03 83 | 13.7 81 | 3.25 64 | 1.91 25 | 7.71 41 | 1.83 18 | 2.14 57 | 7.85 55 | 1.96 26 | 1.80 77 | 10.2 61 | 1.63 81 | 3.72 47 | 4.62 35 | 3.24 68 | 3.93 91 | 13.2 86 | 3.79 94 | 4.35 90 | 5.68 79 | 7.45 115 | 0.97 33 | 1.93 41 | 0.98 34 |
Complementary OF [21] | 64.0 | 4.47 93 | 12.4 71 | 4.63 96 | 1.60 7 | 6.16 18 | 1.67 9 | 2.10 53 | 6.85 51 | 2.16 56 | 2.27 89 | 9.76 59 | 2.19 93 | 4.00 56 | 5.10 48 | 3.71 79 | 3.96 92 | 12.9 83 | 3.32 82 | 2.83 35 | 4.46 30 | 3.08 59 | 2.04 94 | 3.33 88 | 2.86 94 |
HBM-GC [105] | 64.5 | 5.52 99 | 7.21 27 | 5.03 101 | 2.96 90 | 7.15 33 | 3.23 90 | 2.79 84 | 4.90 30 | 2.88 83 | 3.12 100 | 4.92 19 | 2.97 100 | 3.37 36 | 4.23 20 | 3.46 72 | 3.80 90 | 6.63 10 | 3.52 87 | 5.86 113 | 7.23 113 | 4.53 102 | 0.64 2 | 2.02 43 | 0.64 3 |
TF+OM [100] | 65.6 | 3.58 63 | 9.07 52 | 2.75 45 | 2.07 35 | 6.43 20 | 2.37 44 | 1.99 49 | 7.56 54 | 2.78 81 | 2.07 85 | 7.02 40 | 2.07 91 | 4.19 76 | 5.12 51 | 4.32 95 | 3.15 79 | 10.1 56 | 3.00 76 | 4.10 84 | 6.00 85 | 3.92 83 | 1.54 73 | 2.98 79 | 1.94 79 |
Rannacher [23] | 66.3 | 3.60 68 | 11.3 64 | 3.27 66 | 2.41 65 | 9.53 65 | 2.63 68 | 2.60 77 | 11.9 76 | 2.58 72 | 1.36 49 | 12.1 75 | 1.09 43 | 4.22 77 | 5.90 94 | 3.14 65 | 3.63 86 | 16.1 98 | 2.75 72 | 3.72 72 | 5.24 68 | 3.70 72 | 0.96 31 | 2.16 49 | 0.91 19 |
Aniso. Huber-L1 [22] | 66.6 | 3.17 42 | 9.57 54 | 3.05 56 | 3.72 94 | 11.5 86 | 4.38 94 | 2.86 87 | 10.5 68 | 3.80 92 | 1.70 76 | 11.6 74 | 1.42 71 | 4.04 63 | 5.58 78 | 2.98 59 | 2.34 52 | 9.88 51 | 2.05 35 | 4.49 92 | 5.91 84 | 3.42 66 | 1.08 40 | 2.10 46 | 1.02 39 |
ComplOF-FED-GPU [35] | 67.1 | 4.09 85 | 12.8 74 | 4.09 88 | 1.61 10 | 9.86 69 | 1.62 5 | 2.12 55 | 8.39 56 | 1.87 21 | 1.85 79 | 12.4 77 | 1.70 84 | 3.95 51 | 5.25 57 | 3.25 70 | 3.54 84 | 15.1 95 | 3.60 91 | 3.93 77 | 4.83 53 | 4.60 103 | 1.55 75 | 2.93 78 | 1.80 74 |
F-TV-L1 [15] | 67.7 | 5.69 101 | 13.3 76 | 6.62 105 | 2.71 85 | 12.0 88 | 2.86 85 | 2.76 83 | 12.6 81 | 2.43 70 | 2.41 93 | 16.3 90 | 2.02 90 | 4.17 74 | 5.27 58 | 3.74 82 | 2.41 57 | 10.8 64 | 2.49 65 | 3.04 44 | 4.84 57 | 2.26 28 | 0.79 6 | 1.81 34 | 0.76 9 |
TCOF [69] | 67.8 | 3.95 79 | 11.0 62 | 4.20 90 | 2.56 73 | 9.31 63 | 2.68 71 | 2.71 80 | 12.8 83 | 3.34 90 | 2.30 90 | 6.80 36 | 2.33 95 | 4.50 93 | 6.28 107 | 2.58 41 | 1.89 22 | 6.02 4 | 2.06 36 | 4.72 98 | 6.30 90 | 2.58 44 | 1.37 64 | 2.63 65 | 1.22 52 |
NL-TV-NCC [25] | 67.9 | 3.89 76 | 8.49 47 | 3.34 70 | 2.52 70 | 8.44 46 | 2.38 47 | 2.25 63 | 5.49 38 | 1.99 28 | 1.87 80 | 7.61 44 | 1.53 78 | 4.36 85 | 5.91 95 | 2.78 45 | 4.12 96 | 13.0 84 | 3.58 90 | 3.85 75 | 5.74 81 | 3.79 77 | 1.63 78 | 2.81 72 | 1.46 64 |
ACK-Prior [27] | 68.2 | 4.28 88 | 9.53 53 | 3.85 84 | 1.87 24 | 5.68 12 | 1.83 18 | 1.97 48 | 5.25 35 | 1.96 26 | 1.98 83 | 5.26 23 | 1.65 82 | 4.08 68 | 5.12 51 | 3.79 85 | 4.53 102 | 13.0 84 | 3.61 92 | 5.63 109 | 6.40 94 | 8.50 119 | 1.92 91 | 2.90 74 | 2.64 91 |
ROF-ND [107] | 68.7 | 4.03 83 | 11.1 63 | 3.48 75 | 2.33 59 | 5.09 5 | 2.22 35 | 2.19 60 | 6.22 46 | 2.02 34 | 2.30 90 | 5.92 32 | 1.68 83 | 4.23 79 | 6.03 101 | 2.94 56 | 3.70 88 | 12.2 75 | 3.05 78 | 6.21 120 | 6.93 105 | 5.53 107 | 1.43 70 | 2.21 52 | 1.32 53 |
CRTflow [80] | 72.1 | 3.65 71 | 14.0 84 | 3.10 59 | 2.16 44 | 7.88 42 | 2.23 36 | 2.25 63 | 11.2 71 | 1.99 28 | 1.56 66 | 12.7 78 | 1.35 65 | 4.02 59 | 5.53 77 | 3.01 60 | 6.86 116 | 19.6 112 | 8.64 118 | 3.29 53 | 5.53 75 | 3.23 61 | 2.05 95 | 3.95 106 | 2.71 92 |
LDOF [28] | 73.1 | 3.72 72 | 14.9 88 | 3.59 79 | 2.38 63 | 14.0 97 | 2.46 57 | 2.69 79 | 14.4 86 | 2.55 71 | 1.48 61 | 33.9 111 | 1.10 44 | 4.24 81 | 5.59 80 | 3.75 83 | 2.04 31 | 16.4 100 | 1.99 31 | 2.83 35 | 4.83 53 | 2.43 40 | 2.28 103 | 4.02 107 | 3.38 102 |
SRR-TVOF-NL [91] | 73.7 | 4.62 96 | 12.2 70 | 3.55 77 | 2.32 57 | 10.8 77 | 2.34 42 | 2.56 73 | 12.4 79 | 2.59 73 | 1.49 62 | 8.56 53 | 1.17 57 | 4.12 72 | 5.10 48 | 3.51 73 | 2.63 64 | 10.9 65 | 2.26 54 | 5.64 110 | 6.92 104 | 4.13 94 | 2.19 101 | 2.87 73 | 2.86 94 |
DPOF [18] | 73.9 | 4.32 89 | 16.2 91 | 3.30 67 | 2.69 83 | 10.2 71 | 2.69 73 | 2.44 71 | 7.17 52 | 2.61 76 | 1.95 82 | 10.2 61 | 1.55 79 | 3.85 50 | 5.20 54 | 3.03 61 | 2.84 70 | 11.1 66 | 2.67 69 | 4.71 97 | 4.83 53 | 8.84 120 | 1.73 82 | 3.03 81 | 1.86 75 |
LocallyOriented [52] | 74.2 | 3.46 54 | 14.6 86 | 3.01 55 | 2.84 89 | 13.3 93 | 2.85 84 | 2.92 89 | 17.6 92 | 3.05 87 | 1.63 71 | 10.6 68 | 1.43 72 | 4.23 79 | 5.79 89 | 3.19 66 | 2.48 62 | 7.69 23 | 2.87 74 | 3.41 62 | 5.63 77 | 3.25 62 | 1.65 80 | 3.48 92 | 1.86 75 |
SIOF [67] | 74.2 | 4.00 81 | 8.74 50 | 3.46 74 | 2.00 30 | 13.6 94 | 2.13 32 | 3.02 93 | 15.7 88 | 3.38 91 | 2.55 96 | 13.5 82 | 2.50 97 | 4.27 82 | 5.70 84 | 3.70 78 | 3.55 85 | 11.5 71 | 4.01 95 | 3.17 49 | 4.64 46 | 2.12 23 | 1.85 90 | 3.29 87 | 2.15 84 |
Second-order prior [8] | 74.9 | 3.40 50 | 13.6 78 | 3.19 62 | 2.16 44 | 13.8 96 | 2.34 42 | 2.43 70 | 17.1 90 | 2.26 63 | 1.20 34 | 15.7 87 | 0.96 32 | 4.44 89 | 6.10 103 | 3.08 64 | 3.41 83 | 19.7 113 | 2.67 69 | 5.42 108 | 6.02 87 | 5.40 106 | 1.44 71 | 3.44 91 | 1.48 66 |
Brox et al. [5] | 75.5 | 4.01 82 | 14.7 87 | 4.49 94 | 2.75 87 | 11.5 86 | 3.21 89 | 2.33 67 | 12.2 78 | 2.34 66 | 1.46 58 | 19.9 95 | 1.19 59 | 4.62 97 | 5.71 85 | 4.89 105 | 2.13 35 | 13.3 87 | 2.28 56 | 2.87 37 | 4.78 51 | 1.55 6 | 2.30 105 | 3.68 100 | 3.31 100 |
Bartels [41] | 76.0 | 4.23 87 | 10.7 61 | 4.70 98 | 2.37 61 | 5.83 15 | 2.66 70 | 2.21 61 | 7.42 53 | 2.42 68 | 2.59 97 | 8.46 52 | 2.53 98 | 4.33 84 | 5.50 75 | 4.37 97 | 3.69 87 | 14.6 94 | 4.80 104 | 4.75 100 | 6.30 90 | 7.59 116 | 1.13 45 | 2.33 57 | 1.32 53 |
Dynamic MRF [7] | 76.1 | 4.55 95 | 13.6 78 | 5.02 100 | 1.81 19 | 8.86 53 | 1.82 17 | 2.13 56 | 12.6 81 | 1.87 21 | 1.62 69 | 13.2 80 | 1.45 75 | 4.61 95 | 5.80 92 | 4.32 95 | 4.14 97 | 21.3 115 | 4.42 98 | 3.22 52 | 4.41 29 | 5.01 105 | 2.11 97 | 3.92 105 | 3.51 103 |
TriangleFlow [30] | 76.9 | 3.96 80 | 11.5 65 | 4.08 87 | 2.14 42 | 10.2 71 | 2.07 30 | 2.16 59 | 9.80 67 | 1.86 19 | 1.47 59 | 9.22 56 | 1.11 45 | 5.37 112 | 7.25 118 | 4.72 102 | 4.49 101 | 13.7 92 | 4.62 101 | 3.78 73 | 7.33 115 | 4.11 92 | 1.73 82 | 3.48 92 | 2.30 86 |
CLG-TV [48] | 77.5 | 3.59 65 | 9.91 56 | 3.24 63 | 4.16 96 | 11.1 83 | 4.96 95 | 3.12 94 | 11.5 73 | 3.97 93 | 2.31 92 | 13.0 79 | 1.99 89 | 4.56 94 | 6.11 104 | 3.95 88 | 2.85 71 | 12.2 75 | 2.78 73 | 4.23 89 | 5.87 83 | 2.86 53 | 1.17 51 | 2.45 60 | 1.04 40 |
Local-TV-L1 [65] | 78.0 | 4.95 97 | 13.2 75 | 5.40 102 | 4.37 97 | 14.6 99 | 5.04 97 | 4.59 100 | 17.8 94 | 5.96 97 | 2.42 94 | 16.9 93 | 2.25 94 | 3.68 44 | 5.03 44 | 2.82 49 | 2.25 45 | 10.6 61 | 2.24 51 | 2.55 21 | 4.37 26 | 2.91 56 | 2.73 111 | 4.10 109 | 7.77 117 |
p-harmonic [29] | 78.4 | 4.47 93 | 14.4 85 | 4.52 95 | 2.71 85 | 9.33 64 | 2.89 86 | 3.40 95 | 15.0 87 | 3.02 86 | 1.93 81 | 24.1 102 | 1.59 80 | 4.15 73 | 5.18 53 | 3.66 77 | 3.37 82 | 16.0 97 | 3.54 88 | 3.90 76 | 5.36 70 | 2.71 50 | 1.29 57 | 2.41 59 | 1.38 60 |
FlowNetS+ft+v [112] | 78.6 | 3.42 52 | 13.4 77 | 3.39 72 | 2.54 72 | 11.2 84 | 2.80 78 | 2.94 90 | 18.6 97 | 4.76 95 | 1.43 53 | 27.4 105 | 1.20 60 | 4.67 99 | 6.35 109 | 3.71 79 | 1.96 26 | 12.3 77 | 2.01 33 | 4.10 84 | 6.00 85 | 4.11 92 | 1.76 85 | 3.49 95 | 2.37 88 |
DF-Auto [115] | 78.7 | 3.88 75 | 17.6 95 | 2.93 54 | 5.44 103 | 14.7 100 | 6.44 103 | 4.54 99 | 16.8 89 | 9.38 103 | 2.22 88 | 15.0 84 | 1.95 87 | 4.32 83 | 6.00 99 | 3.88 87 | 1.44 3 | 7.14 15 | 1.73 10 | 4.20 86 | 6.78 100 | 1.70 9 | 2.42 107 | 4.04 108 | 3.34 101 |
CNN-flow-warp+ref [117] | 79.2 | 3.90 77 | 19.8 109 | 3.73 82 | 3.40 93 | 10.9 80 | 4.21 93 | 3.85 98 | 23.8 105 | 6.07 98 | 1.65 73 | 22.4 100 | 1.37 67 | 4.38 88 | 5.58 78 | 4.08 90 | 2.23 42 | 13.8 93 | 2.33 60 | 2.41 17 | 4.27 17 | 2.24 27 | 2.43 108 | 3.66 99 | 3.64 107 |
CBF [12] | 79.5 | 3.59 65 | 10.5 60 | 3.68 81 | 4.72 99 | 10.4 74 | 6.02 102 | 2.28 65 | 9.24 61 | 2.96 85 | 1.63 71 | 12.2 76 | 1.36 66 | 4.48 92 | 5.75 87 | 3.99 89 | 2.70 66 | 10.7 63 | 2.48 64 | 6.13 118 | 7.02 106 | 5.92 109 | 1.45 72 | 2.65 66 | 1.66 70 |
SuperFlow [81] | 81.0 | 3.40 50 | 11.5 65 | 3.31 68 | 3.97 95 | 12.2 90 | 5.00 96 | 3.01 92 | 17.9 95 | 7.70 101 | 2.66 98 | 18.1 94 | 2.64 99 | 4.17 74 | 5.42 67 | 4.16 93 | 2.19 40 | 10.6 61 | 2.20 47 | 4.22 88 | 6.11 89 | 2.45 41 | 2.12 99 | 3.64 98 | 3.59 105 |
StereoFlow [44] | 83.5 | 21.8 128 | 37.8 124 | 27.4 128 | 24.3 126 | 37.6 128 | 22.4 124 | 28.3 128 | 39.2 124 | 28.8 124 | 24.0 127 | 47.7 121 | 21.6 126 | 5.15 110 | 5.49 74 | 6.01 117 | 0.95 1 | 6.87 11 | 1.06 1 | 1.68 1 | 3.70 2 | 0.92 1 | 1.29 57 | 2.32 55 | 1.49 67 |
Fusion [6] | 85.8 | 3.76 73 | 16.9 92 | 4.07 86 | 1.99 29 | 7.37 38 | 2.26 38 | 2.07 52 | 8.51 58 | 2.28 64 | 1.59 67 | 24.8 103 | 1.37 67 | 5.00 108 | 6.36 110 | 4.98 109 | 4.70 104 | 16.2 99 | 5.01 107 | 6.00 117 | 7.50 117 | 4.38 100 | 2.97 114 | 3.74 103 | 3.55 104 |
Learning Flow [11] | 86.1 | 3.80 74 | 11.9 68 | 3.58 78 | 3.02 91 | 13.0 92 | 3.34 91 | 2.84 85 | 17.9 95 | 3.18 89 | 1.82 78 | 34.6 114 | 1.50 76 | 5.44 114 | 7.32 119 | 4.61 101 | 3.10 77 | 18.8 108 | 3.04 77 | 3.94 78 | 6.38 93 | 3.65 70 | 1.37 64 | 3.38 89 | 1.18 45 |
TriFlow [95] | 86.4 | 4.44 90 | 13.8 82 | 3.62 80 | 3.16 92 | 9.65 67 | 3.81 92 | 2.89 88 | 19.6 99 | 6.36 99 | 2.48 95 | 7.88 47 | 2.33 95 | 4.37 87 | 5.44 70 | 4.28 94 | 2.98 74 | 8.42 29 | 3.07 79 | 11.7 127 | 7.70 118 | 21.5 127 | 1.76 85 | 2.98 79 | 1.94 79 |
Shiralkar [42] | 86.8 | 4.46 91 | 18.3 99 | 4.36 91 | 1.93 26 | 16.4 101 | 1.87 21 | 2.99 91 | 17.6 92 | 2.01 32 | 2.10 86 | 21.0 97 | 1.96 88 | 4.36 85 | 5.72 86 | 3.55 74 | 5.65 112 | 19.4 110 | 5.11 109 | 4.90 103 | 5.57 76 | 7.14 113 | 2.11 97 | 4.71 114 | 2.53 90 |
StereoOF-V1MT [119] | 88.8 | 4.46 91 | 18.0 97 | 4.46 93 | 2.09 36 | 18.6 105 | 1.79 16 | 3.70 97 | 20.6 102 | 2.13 53 | 2.18 87 | 25.0 104 | 1.91 86 | 5.52 116 | 6.98 116 | 4.82 103 | 5.01 108 | 25.8 120 | 4.73 102 | 3.21 51 | 5.27 69 | 3.64 69 | 2.32 106 | 4.64 112 | 2.83 93 |
SegOF [10] | 89.8 | 5.62 100 | 17.1 93 | 3.08 57 | 8.33 112 | 20.9 107 | 10.1 114 | 7.44 106 | 21.7 103 | 13.3 110 | 5.42 111 | 21.0 97 | 4.47 107 | 4.81 104 | 5.51 76 | 5.74 116 | 4.97 107 | 17.1 101 | 4.83 105 | 2.12 3 | 4.38 28 | 1.46 4 | 2.17 100 | 3.23 85 | 3.74 108 |
Ad-TV-NDC [36] | 92.1 | 8.75 112 | 15.3 90 | 12.3 120 | 10.5 117 | 24.2 115 | 12.3 117 | 8.96 111 | 28.2 108 | 11.5 105 | 5.31 110 | 22.8 101 | 5.55 111 | 4.03 61 | 5.79 89 | 2.86 51 | 2.80 68 | 10.0 55 | 2.87 74 | 3.04 44 | 4.52 35 | 2.66 48 | 4.62 121 | 5.79 121 | 30.9 127 |
Modified CLG [34] | 96.5 | 6.79 106 | 24.7 112 | 6.63 106 | 7.09 108 | 17.4 102 | 9.40 112 | 10.1 112 | 29.2 110 | 16.6 115 | 4.48 108 | 27.5 106 | 3.86 103 | 4.80 102 | 6.31 108 | 4.48 100 | 2.65 65 | 17.6 104 | 2.69 71 | 2.92 42 | 4.94 59 | 2.07 19 | 3.19 116 | 5.17 117 | 5.78 113 |
Filter Flow [19] | 96.7 | 6.76 105 | 17.6 95 | 4.37 92 | 5.01 101 | 17.6 103 | 5.49 100 | 5.98 103 | 26.3 106 | 18.4 118 | 7.23 113 | 29.9 109 | 6.91 114 | 5.12 109 | 6.23 106 | 5.36 111 | 5.23 109 | 11.9 74 | 4.95 106 | 6.64 121 | 8.75 124 | 3.75 76 | 0.95 29 | 2.15 47 | 1.21 50 |
IAOF2 [51] | 98.5 | 5.05 98 | 13.6 78 | 4.64 97 | 4.90 100 | 14.5 98 | 5.78 101 | 3.68 96 | 18.6 97 | 5.15 96 | 12.3 121 | 34.1 113 | 13.8 121 | 4.65 98 | 6.21 105 | 3.78 84 | 4.47 100 | 13.5 88 | 3.70 93 | 5.73 111 | 7.13 110 | 3.98 85 | 1.96 92 | 3.53 96 | 2.35 87 |
FlowNet2 [122] | 98.8 | 8.99 113 | 25.8 113 | 7.01 107 | 9.84 115 | 19.0 106 | 10.7 115 | 7.98 109 | 20.1 100 | 13.5 111 | 4.47 107 | 9.41 57 | 4.21 106 | 5.17 111 | 6.08 102 | 4.92 107 | 4.10 95 | 11.2 68 | 4.43 99 | 5.98 116 | 7.71 119 | 2.90 55 | 1.78 87 | 2.92 77 | 1.89 77 |
BlockOverlap [61] | 99.5 | 6.80 107 | 12.1 69 | 5.94 104 | 5.51 104 | 13.6 94 | 6.58 104 | 5.32 102 | 22.2 104 | 7.30 100 | 4.20 103 | 16.7 92 | 4.06 105 | 4.45 90 | 5.39 66 | 5.11 110 | 4.91 105 | 12.5 78 | 4.34 97 | 6.77 123 | 7.13 110 | 9.52 122 | 2.02 93 | 3.24 86 | 9.49 120 |
2D-CLG [1] | 100.5 | 9.69 116 | 37.7 123 | 7.18 109 | 11.1 118 | 21.9 111 | 13.9 120 | 19.0 124 | 34.8 116 | 28.7 123 | 13.0 122 | 46.7 119 | 12.8 120 | 4.97 106 | 5.79 89 | 5.47 114 | 4.08 93 | 21.2 114 | 4.32 96 | 2.29 12 | 4.00 5 | 1.64 7 | 4.47 120 | 5.51 120 | 6.55 115 |
HBpMotionGpu [43] | 100.5 | 5.92 102 | 15.0 89 | 4.79 99 | 7.78 110 | 22.4 112 | 9.04 111 | 7.17 105 | 39.2 124 | 17.3 117 | 3.31 101 | 13.4 81 | 3.14 101 | 4.71 100 | 5.88 93 | 4.84 104 | 3.74 89 | 13.5 88 | 3.54 88 | 5.96 115 | 7.17 112 | 3.68 71 | 2.24 102 | 3.43 90 | 4.65 109 |
SPSA-learn [13] | 100.7 | 6.87 108 | 21.3 110 | 7.92 112 | 6.02 106 | 21.1 109 | 6.96 106 | 7.55 107 | 27.5 107 | 12.7 109 | 4.44 105 | 29.2 107 | 4.59 109 | 4.80 102 | 5.92 96 | 4.93 108 | 4.94 106 | 17.3 102 | 5.02 108 | 3.37 59 | 5.01 61 | 2.29 31 | 4.14 119 | 4.97 116 | 6.49 114 |
GroupFlow [9] | 101.0 | 9.15 115 | 25.8 113 | 10.5 118 | 11.6 120 | 30.0 121 | 12.3 117 | 10.2 113 | 35.4 118 | 11.9 106 | 3.50 102 | 15.8 89 | 3.39 102 | 5.48 115 | 6.56 112 | 4.42 98 | 9.25 121 | 24.8 117 | 10.8 123 | 2.35 16 | 4.58 38 | 1.67 8 | 2.93 113 | 5.22 118 | 4.99 110 |
GraphCuts [14] | 101.9 | 6.34 103 | 17.1 93 | 5.55 103 | 5.30 102 | 20.9 107 | 5.26 99 | 6.05 104 | 20.4 101 | 12.4 108 | 2.85 99 | 20.9 96 | 2.15 92 | 4.74 101 | 5.95 97 | 4.90 106 | 8.69 120 | 12.6 80 | 5.19 110 | 5.79 112 | 6.40 94 | 6.80 112 | 2.45 109 | 3.48 92 | 3.59 105 |
Black & Anandan [4] | 102.2 | 7.19 109 | 18.9 102 | 8.40 113 | 5.96 105 | 22.6 113 | 6.69 105 | 8.73 110 | 28.7 109 | 12.1 107 | 4.46 106 | 29.4 108 | 4.52 108 | 4.91 105 | 6.59 113 | 4.09 91 | 4.18 98 | 19.4 110 | 4.44 100 | 4.69 95 | 6.36 92 | 2.01 17 | 3.13 115 | 4.46 110 | 5.06 111 |
IAOF [50] | 103.3 | 6.54 104 | 18.3 99 | 7.13 108 | 6.99 107 | 18.4 104 | 7.90 109 | 7.71 108 | 32.3 113 | 8.44 102 | 8.21 116 | 31.8 110 | 9.78 118 | 4.61 95 | 6.01 100 | 4.14 92 | 4.35 99 | 18.9 109 | 3.43 85 | 4.69 95 | 6.08 88 | 3.31 63 | 3.23 117 | 4.69 113 | 15.9 125 |
2bit-BM-tele [98] | 106.1 | 8.99 113 | 18.6 101 | 10.2 116 | 4.45 98 | 11.3 85 | 5.04 97 | 4.66 101 | 17.3 91 | 4.41 94 | 5.23 109 | 21.7 99 | 5.04 110 | 4.99 107 | 5.96 98 | 5.46 113 | 6.47 115 | 18.3 105 | 7.49 116 | 7.77 125 | 8.57 123 | 12.5 125 | 2.29 104 | 3.89 104 | 2.99 98 |
Nguyen [33] | 107.3 | 8.16 110 | 23.0 111 | 7.57 111 | 16.5 122 | 22.7 114 | 19.3 122 | 16.8 119 | 36.0 119 | 20.7 121 | 13.8 123 | 39.5 115 | 14.7 123 | 5.40 113 | 6.44 111 | 6.70 118 | 4.54 103 | 18.5 107 | 5.42 111 | 3.50 66 | 5.02 62 | 2.08 20 | 4.00 118 | 5.50 119 | 8.53 118 |
UnFlow [129] | 107.3 | 19.4 127 | 44.0 128 | 10.2 116 | 11.3 119 | 21.2 110 | 12.4 119 | 18.1 121 | 36.0 119 | 15.5 113 | 7.47 115 | 34.0 112 | 6.40 113 | 7.10 123 | 7.11 117 | 8.76 123 | 8.31 119 | 24.8 117 | 9.26 119 | 5.13 104 | 6.50 97 | 1.52 5 | 1.57 76 | 3.14 83 | 2.01 81 |
Heeger++ [104] | 109.5 | 18.3 126 | 32.1 120 | 10.5 118 | 9.98 116 | 34.3 127 | 7.90 109 | 16.0 117 | 32.7 114 | 11.0 104 | 9.23 117 | 47.6 120 | 7.86 116 | 5.95 118 | 6.74 114 | 5.71 115 | 23.4 127 | 49.5 128 | 24.5 127 | 3.61 70 | 6.53 98 | 2.58 44 | 1.78 87 | 3.68 100 | 2.96 96 |
SILK [79] | 110.2 | 10.7 117 | 31.4 119 | 13.1 122 | 8.77 113 | 26.6 117 | 9.80 113 | 13.6 115 | 34.9 117 | 16.7 116 | 6.53 112 | 45.4 117 | 6.08 112 | 6.11 119 | 7.36 121 | 6.71 119 | 6.96 117 | 29.4 123 | 7.39 115 | 2.97 43 | 4.77 50 | 3.96 84 | 5.00 122 | 6.99 122 | 10.7 121 |
Horn & Schunck [3] | 111.0 | 8.40 111 | 27.2 115 | 9.62 114 | 7.28 109 | 28.3 119 | 7.55 108 | 13.3 114 | 31.9 112 | 15.8 114 | 7.35 114 | 48.5 122 | 7.69 115 | 5.84 117 | 7.34 120 | 5.45 112 | 5.80 113 | 25.8 120 | 6.79 114 | 5.25 106 | 7.11 109 | 2.11 22 | 5.21 123 | 8.30 124 | 6.66 116 |
Periodicity [78] | 114.1 | 12.3 119 | 51.4 131 | 7.39 110 | 9.23 114 | 38.3 129 | 11.1 116 | 34.7 129 | 48.1 129 | 36.0 128 | 4.27 104 | 57.7 128 | 3.99 104 | 24.4 129 | 73.2 129 | 16.2 128 | 29.6 129 | 74.3 129 | 29.4 129 | 3.29 53 | 5.67 78 | 1.90 14 | 6.64 125 | 44.8 129 | 21.5 126 |
FFV1MT [106] | 114.8 | 17.0 125 | 35.1 122 | 10.1 115 | 8.30 111 | 33.0 125 | 7.40 107 | 17.2 120 | 40.0 126 | 15.3 112 | 9.57 118 | 55.8 127 | 8.75 117 | 7.99 127 | 8.46 127 | 10.1 126 | 21.8 126 | 36.1 126 | 23.3 126 | 4.43 91 | 7.02 106 | 4.09 91 | 1.78 87 | 3.68 100 | 2.96 96 |
TI-DOFE [24] | 114.8 | 16.4 123 | 34.0 121 | 21.2 127 | 21.5 125 | 31.9 122 | 25.3 126 | 24.4 127 | 41.0 127 | 33.0 127 | 22.8 126 | 46.3 118 | 25.2 127 | 6.25 120 | 7.67 123 | 6.82 120 | 6.37 114 | 25.6 119 | 7.87 117 | 3.94 78 | 5.78 82 | 1.78 12 | 8.49 127 | 9.86 126 | 12.5 122 |
SLK [47] | 115.7 | 12.3 119 | 43.0 127 | 16.5 124 | 19.8 124 | 32.9 124 | 22.4 124 | 21.4 125 | 38.4 122 | 29.3 126 | 41.6 129 | 51.6 124 | 44.5 129 | 6.87 122 | 7.63 122 | 8.94 124 | 8.09 118 | 31.2 125 | 9.56 120 | 3.34 57 | 5.41 71 | 2.60 46 | 8.13 126 | 9.23 125 | 13.9 124 |
Adaptive flow [45] | 121.2 | 16.4 123 | 28.7 117 | 16.7 125 | 17.2 123 | 25.5 116 | 19.6 123 | 18.8 123 | 37.6 121 | 36.5 129 | 11.2 120 | 43.6 116 | 11.9 119 | 7.11 124 | 7.79 124 | 7.88 121 | 10.1 124 | 24.1 116 | 10.1 122 | 16.1 128 | 14.2 128 | 22.1 128 | 2.92 112 | 4.89 115 | 5.22 112 |
HCIC-L [99] | 121.9 | 24.1 129 | 31.3 118 | 12.9 121 | 27.6 128 | 28.7 120 | 69.9 129 | 16.0 117 | 30.6 111 | 23.1 122 | 18.1 125 | 55.4 126 | 17.8 125 | 7.39 125 | 8.35 126 | 8.32 122 | 12.6 125 | 18.3 105 | 14.4 125 | 25.6 129 | 23.8 129 | 23.6 129 | 2.62 110 | 4.51 111 | 9.36 119 |
PGAM+LK [55] | 122.6 | 14.0 121 | 40.6 125 | 18.8 126 | 14.9 121 | 33.1 126 | 17.8 121 | 14.4 116 | 32.9 115 | 19.3 119 | 15.7 124 | 63.6 129 | 14.9 124 | 6.36 121 | 6.81 115 | 9.14 125 | 9.83 123 | 30.7 124 | 9.83 121 | 10.4 126 | 12.2 127 | 10.3 123 | 5.30 124 | 7.26 123 | 13.0 123 |
FOLKI [16] | 123.0 | 11.0 118 | 41.0 126 | 14.5 123 | 24.9 127 | 32.3 123 | 36.7 127 | 18.7 122 | 43.8 128 | 20.5 120 | 10.9 119 | 50.5 123 | 13.8 121 | 7.42 126 | 8.28 125 | 10.6 127 | 9.75 122 | 36.9 127 | 12.1 124 | 4.77 102 | 7.29 114 | 11.0 124 | 12.2 128 | 11.4 127 | 36.4 128 |
Pyramid LK [2] | 125.5 | 15.8 122 | 28.2 116 | 30.4 129 | 35.8 129 | 28.0 118 | 49.6 128 | 22.3 126 | 38.6 123 | 29.1 125 | 31.8 128 | 51.7 125 | 39.0 128 | 18.3 128 | 24.8 128 | 24.1 129 | 26.7 128 | 28.6 122 | 26.7 128 | 7.19 124 | 8.98 125 | 7.70 118 | 32.7 129 | 40.6 128 | 57.0 129 |
AdaConv-v1 [126] | 130.0 | 45.2 130 | 50.1 129 | 46.0 130 | 78.3 130 | 79.6 130 | 76.9 130 | 78.3 130 | 73.0 130 | 77.6 130 | 79.2 130 | 80.8 130 | 79.6 130 | 83.7 130 | 84.3 130 | 83.6 130 | 82.1 130 | 80.6 130 | 81.5 130 | 69.6 130 | 58.5 130 | 75.3 130 | 84.4 130 | 84.7 130 | 83.9 130 |
SepConv-v1 [127] | 130.0 | 45.2 130 | 50.1 129 | 46.0 130 | 78.3 130 | 79.6 130 | 76.9 130 | 78.3 130 | 73.0 130 | 77.6 130 | 79.2 130 | 80.8 130 | 79.6 130 | 83.7 130 | 84.3 130 | 83.6 130 | 82.1 130 | 80.6 130 | 81.5 130 | 69.6 130 | 58.5 130 | 75.3 130 | 84.4 130 | 84.7 130 | 83.9 130 |
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. |