Optical flow evaluation results |
Statistics:
Average
SD
R0.5
R1.0
R2.0
A50
A75
A95
Error type: endpoint angle interpolation normalized interpolation |
A75 endpoint error |
avg. |
Army (Hidden texture) GT im0 im1 |
Mequon (Hidden texture) GT im0 im1 |
Schefflera (Hidden texture) GT im0 im1 |
Wooden (Hidden texture) GT im0 im1 |
Grove (Synthetic) GT im0 im1 |
Urban (Synthetic) GT im0 im1 |
Yosemite (Synthetic) GT im0 im1 |
Teddy (Stereo) GT im0 im1 | ||||||||||||||||
rank | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | |
NNF-Local [87] | 9.4 | 0.06 2 | 0.15 4 | 0.05 2 | 0.13 3 | 0.55 15 | 0.13 8 | 0.16 1 | 0.28 9 | 0.16 2 | 0.07 10 | 0.33 15 | 0.07 14 | 0.36 6 | 0.61 4 | 0.22 6 | 0.16 23 | 0.47 5 | 0.14 31 | 0.13 20 | 0.15 2 | 0.23 20 | 0.23 1 | 0.69 21 | 0.20 2 |
MDP-Flow2 [68] | 11.9 | 0.08 36 | 0.17 13 | 0.07 15 | 0.13 3 | 0.42 2 | 0.12 2 | 0.17 6 | 0.22 3 | 0.17 10 | 0.08 18 | 0.35 18 | 0.07 14 | 0.42 12 | 0.84 16 | 0.24 9 | 0.16 23 | 0.43 2 | 0.13 17 | 0.13 20 | 0.16 12 | 0.22 19 | 0.27 4 | 0.53 4 | 0.29 7 |
NN-field [71] | 12.2 | 0.06 2 | 0.17 13 | 0.05 2 | 0.15 14 | 0.62 29 | 0.15 21 | 0.17 6 | 0.27 7 | 0.17 10 | 0.07 10 | 0.29 8 | 0.07 14 | 0.35 5 | 0.61 4 | 0.21 5 | 0.13 11 | 0.39 1 | 0.11 5 | 0.16 45 | 0.16 12 | 0.27 48 | 0.24 2 | 0.67 17 | 0.19 1 |
OFLAF [77] | 16.1 | 0.07 12 | 0.15 4 | 0.07 15 | 0.15 14 | 0.46 5 | 0.14 14 | 0.16 1 | 0.21 2 | 0.16 2 | 0.07 10 | 0.20 3 | 0.06 5 | 0.31 1 | 0.59 3 | 0.19 3 | 0.22 70 | 0.59 12 | 0.17 60 | 0.14 25 | 0.16 12 | 0.29 55 | 0.29 25 | 0.53 4 | 0.33 30 |
ComponentFusion [96] | 17.0 | 0.06 2 | 0.16 6 | 0.05 2 | 0.14 9 | 0.45 4 | 0.14 14 | 0.16 1 | 0.30 10 | 0.16 2 | 0.06 1 | 0.37 22 | 0.05 1 | 0.47 19 | 0.99 32 | 0.33 30 | 0.21 62 | 1.08 46 | 0.16 54 | 0.14 25 | 0.17 27 | 0.20 17 | 0.27 4 | 0.64 11 | 0.29 7 |
WLIF-Flow [93] | 17.3 | 0.07 12 | 0.16 6 | 0.07 15 | 0.17 26 | 0.58 20 | 0.16 29 | 0.17 6 | 0.30 10 | 0.18 16 | 0.08 18 | 0.31 12 | 0.07 14 | 0.41 10 | 0.77 11 | 0.27 17 | 0.17 32 | 0.66 14 | 0.14 31 | 0.17 55 | 0.15 2 | 0.27 48 | 0.25 3 | 0.56 6 | 0.27 3 |
TC/T-Flow [76] | 17.5 | 0.05 1 | 0.19 24 | 0.04 1 | 0.13 3 | 0.61 27 | 0.12 2 | 0.17 6 | 0.34 16 | 0.16 2 | 0.06 1 | 0.43 37 | 0.05 1 | 0.47 19 | 0.92 24 | 0.24 9 | 0.11 3 | 0.53 7 | 0.10 3 | 0.14 25 | 0.15 2 | 0.42 101 | 0.31 39 | 0.82 36 | 0.33 30 |
NNF-EAC [103] | 19.0 | 0.08 36 | 0.18 20 | 0.07 15 | 0.14 9 | 0.50 9 | 0.14 14 | 0.18 21 | 0.27 7 | 0.17 10 | 0.08 18 | 0.43 37 | 0.08 32 | 0.43 16 | 0.84 16 | 0.26 16 | 0.16 23 | 0.54 9 | 0.13 17 | 0.15 37 | 0.16 12 | 0.24 30 | 0.29 25 | 0.66 14 | 0.30 12 |
LME [70] | 19.2 | 0.07 12 | 0.16 6 | 0.06 7 | 0.14 9 | 0.43 3 | 0.13 8 | 0.18 21 | 0.30 10 | 0.18 16 | 0.08 18 | 0.44 42 | 0.07 14 | 0.49 25 | 0.96 28 | 0.31 23 | 0.15 20 | 0.74 19 | 0.14 31 | 0.15 37 | 0.17 27 | 0.25 37 | 0.28 12 | 0.68 20 | 0.31 15 |
ALD-Flow [66] | 20.7 | 0.06 2 | 0.16 6 | 0.06 7 | 0.13 3 | 0.49 7 | 0.12 2 | 0.17 6 | 0.35 21 | 0.17 10 | 0.06 1 | 0.42 34 | 0.06 5 | 0.54 34 | 1.24 46 | 0.25 13 | 0.11 3 | 0.98 38 | 0.10 3 | 0.15 37 | 0.15 2 | 0.36 78 | 0.32 43 | 1.10 53 | 0.37 42 |
TC-Flow [46] | 20.7 | 0.06 2 | 0.17 13 | 0.07 15 | 0.11 1 | 0.46 5 | 0.10 1 | 0.17 6 | 0.34 16 | 0.16 2 | 0.06 1 | 0.41 31 | 0.06 5 | 0.53 32 | 1.21 42 | 0.25 13 | 0.13 11 | 1.04 43 | 0.13 17 | 0.13 20 | 0.15 2 | 0.36 78 | 0.32 43 | 1.07 48 | 0.41 50 |
Layers++ [37] | 21.0 | 0.07 12 | 0.17 13 | 0.08 44 | 0.19 46 | 0.56 18 | 0.19 52 | 0.17 6 | 0.25 4 | 0.18 16 | 0.06 1 | 0.15 2 | 0.06 5 | 0.31 1 | 0.51 1 | 0.18 2 | 0.17 32 | 0.77 21 | 0.13 17 | 0.18 61 | 0.18 49 | 0.31 60 | 0.27 4 | 0.66 14 | 0.32 23 |
RNLOD-Flow [121] | 21.4 | 0.06 2 | 0.14 3 | 0.06 7 | 0.15 14 | 0.65 36 | 0.13 8 | 0.17 6 | 0.34 16 | 0.17 10 | 0.07 10 | 0.25 6 | 0.06 5 | 0.41 10 | 0.86 19 | 0.24 9 | 0.16 23 | 0.66 14 | 0.14 31 | 0.20 84 | 0.23 93 | 0.39 90 | 0.27 4 | 0.57 7 | 0.28 6 |
nLayers [57] | 22.6 | 0.06 2 | 0.12 1 | 0.06 7 | 0.23 78 | 0.63 31 | 0.23 88 | 0.17 6 | 0.32 13 | 0.20 52 | 0.06 1 | 0.14 1 | 0.06 5 | 0.32 4 | 0.53 2 | 0.19 3 | 0.16 23 | 0.53 7 | 0.13 17 | 0.15 37 | 0.17 27 | 0.26 43 | 0.29 25 | 0.70 22 | 0.39 47 |
HAST [109] | 23.5 | 0.06 2 | 0.12 1 | 0.05 2 | 0.13 3 | 0.50 9 | 0.12 2 | 0.16 1 | 0.17 1 | 0.15 1 | 0.06 1 | 0.21 4 | 0.05 1 | 0.31 1 | 0.63 6 | 0.17 1 | 0.22 70 | 0.95 35 | 0.17 60 | 0.24 100 | 0.25 107 | 0.70 121 | 0.28 12 | 0.44 1 | 0.32 23 |
PMMST [114] | 24.6 | 0.09 46 | 0.17 13 | 0.09 62 | 0.19 46 | 0.57 19 | 0.19 52 | 0.18 21 | 0.33 14 | 0.19 28 | 0.09 44 | 0.21 4 | 0.09 60 | 0.40 9 | 0.69 8 | 0.23 8 | 0.17 32 | 0.45 4 | 0.13 17 | 0.14 25 | 0.16 12 | 0.23 20 | 0.30 31 | 0.60 8 | 0.29 7 |
AGIF+OF [85] | 25.3 | 0.07 12 | 0.18 20 | 0.07 15 | 0.21 63 | 0.79 58 | 0.20 66 | 0.17 6 | 0.37 27 | 0.18 16 | 0.08 18 | 0.36 19 | 0.07 14 | 0.42 12 | 0.74 10 | 0.28 18 | 0.16 23 | 0.75 20 | 0.13 17 | 0.17 55 | 0.17 27 | 0.28 52 | 0.28 12 | 0.65 13 | 0.31 15 |
OAR-Flow [125] | 26.5 | 0.07 12 | 0.21 31 | 0.06 7 | 0.15 14 | 0.70 40 | 0.14 14 | 0.17 6 | 0.61 59 | 0.19 28 | 0.06 1 | 0.45 44 | 0.06 5 | 0.65 47 | 1.39 54 | 0.31 23 | 0.09 1 | 1.09 47 | 0.09 2 | 0.11 3 | 0.14 1 | 0.24 30 | 0.36 56 | 1.09 52 | 0.46 60 |
IROF++ [58] | 26.7 | 0.07 12 | 0.18 20 | 0.07 15 | 0.19 46 | 0.78 55 | 0.19 52 | 0.18 21 | 0.40 35 | 0.19 28 | 0.08 18 | 0.42 34 | 0.08 32 | 0.47 19 | 0.88 21 | 0.33 30 | 0.17 32 | 1.00 40 | 0.13 17 | 0.12 7 | 0.18 49 | 0.15 7 | 0.28 12 | 0.73 24 | 0.31 15 |
Classic+CPF [83] | 28.0 | 0.07 12 | 0.20 28 | 0.07 15 | 0.20 58 | 0.76 51 | 0.19 52 | 0.17 6 | 0.38 32 | 0.18 16 | 0.07 10 | 0.38 27 | 0.07 14 | 0.46 18 | 0.84 16 | 0.32 28 | 0.17 32 | 0.77 21 | 0.13 17 | 0.18 61 | 0.17 27 | 0.40 93 | 0.28 12 | 0.64 11 | 0.31 15 |
PH-Flow [101] | 28.1 | 0.07 12 | 0.21 31 | 0.07 15 | 0.18 34 | 0.73 47 | 0.18 41 | 0.18 21 | 0.34 16 | 0.19 28 | 0.08 18 | 0.39 30 | 0.08 32 | 0.45 17 | 0.79 13 | 0.33 30 | 0.18 45 | 0.69 16 | 0.14 31 | 0.18 61 | 0.17 27 | 0.38 85 | 0.27 4 | 0.66 14 | 0.29 7 |
FC-2Layers-FF [74] | 28.3 | 0.07 12 | 0.16 6 | 0.07 15 | 0.18 34 | 0.71 44 | 0.18 41 | 0.18 21 | 0.26 5 | 0.19 28 | 0.08 18 | 0.29 8 | 0.08 32 | 0.38 8 | 0.64 7 | 0.29 22 | 0.19 51 | 0.73 18 | 0.15 43 | 0.20 84 | 0.18 49 | 0.39 90 | 0.27 4 | 0.67 17 | 0.32 23 |
Efficient-NL [60] | 29.0 | 0.06 2 | 0.16 6 | 0.06 7 | 0.20 58 | 0.74 49 | 0.19 52 | 0.18 21 | 0.36 26 | 0.18 16 | 0.08 18 | 0.30 10 | 0.07 14 | 0.42 12 | 0.77 11 | 0.28 18 | 0.19 51 | 0.91 31 | 0.15 43 | 0.18 61 | 0.18 49 | 0.34 71 | 0.30 31 | 0.61 10 | 0.33 30 |
Sparse-NonSparse [56] | 30.3 | 0.07 12 | 0.21 31 | 0.07 15 | 0.19 46 | 0.76 51 | 0.19 52 | 0.17 6 | 0.37 27 | 0.19 28 | 0.08 18 | 0.44 42 | 0.07 14 | 0.54 34 | 1.03 35 | 0.37 45 | 0.17 32 | 0.93 34 | 0.13 17 | 0.18 61 | 0.15 2 | 0.37 80 | 0.27 4 | 0.75 27 | 0.31 15 |
ProbFlowFields [128] | 32.1 | 0.09 46 | 0.33 72 | 0.08 44 | 0.18 34 | 0.58 20 | 0.17 34 | 0.18 21 | 0.46 45 | 0.21 65 | 0.07 10 | 0.36 19 | 0.07 14 | 0.61 42 | 1.29 47 | 0.35 36 | 0.12 6 | 0.57 11 | 0.12 9 | 0.12 7 | 0.16 12 | 0.24 30 | 0.36 56 | 1.07 48 | 0.37 42 |
Correlation Flow [75] | 32.6 | 0.08 36 | 0.21 31 | 0.08 44 | 0.15 14 | 0.51 11 | 0.13 8 | 0.18 21 | 0.35 21 | 0.17 10 | 0.09 44 | 0.32 13 | 0.08 32 | 0.51 26 | 1.08 38 | 0.28 18 | 0.25 85 | 0.80 24 | 0.20 81 | 0.19 73 | 0.17 27 | 0.38 85 | 0.29 25 | 0.60 8 | 0.29 7 |
LSM [39] | 33.4 | 0.07 12 | 0.21 31 | 0.07 15 | 0.18 34 | 0.78 55 | 0.18 41 | 0.18 21 | 0.38 32 | 0.19 28 | 0.08 18 | 0.45 44 | 0.08 32 | 0.53 32 | 1.02 34 | 0.35 36 | 0.18 45 | 0.98 38 | 0.14 31 | 0.19 73 | 0.16 12 | 0.38 85 | 0.27 4 | 0.79 33 | 0.31 15 |
COFM [59] | 33.6 | 0.06 2 | 0.19 24 | 0.05 2 | 0.15 14 | 0.60 24 | 0.15 21 | 0.17 6 | 0.45 44 | 0.19 28 | 0.06 1 | 0.32 13 | 0.05 1 | 0.69 56 | 1.39 54 | 0.51 65 | 0.24 81 | 0.84 27 | 0.17 60 | 0.16 45 | 0.15 2 | 0.38 85 | 0.37 59 | 0.81 34 | 0.45 58 |
Ramp [62] | 33.9 | 0.07 12 | 0.21 31 | 0.07 15 | 0.18 34 | 0.78 55 | 0.19 52 | 0.18 21 | 0.35 21 | 0.19 28 | 0.09 44 | 0.41 31 | 0.08 32 | 0.52 31 | 0.96 28 | 0.36 41 | 0.18 45 | 0.88 30 | 0.14 31 | 0.18 61 | 0.16 12 | 0.41 97 | 0.28 12 | 0.75 27 | 0.32 23 |
FMOF [94] | 34.1 | 0.07 12 | 0.18 20 | 0.07 15 | 0.22 71 | 0.79 58 | 0.21 73 | 0.18 21 | 0.34 16 | 0.20 52 | 0.08 18 | 0.34 16 | 0.07 14 | 0.47 19 | 0.89 22 | 0.34 34 | 0.17 32 | 0.78 23 | 0.13 17 | 0.19 73 | 0.18 49 | 0.40 93 | 0.30 31 | 0.74 25 | 0.31 15 |
FESL [72] | 34.6 | 0.07 12 | 0.16 6 | 0.08 44 | 0.22 71 | 0.84 65 | 0.20 66 | 0.18 21 | 0.37 27 | 0.19 28 | 0.08 18 | 0.28 7 | 0.07 14 | 0.48 23 | 0.82 15 | 0.39 48 | 0.17 32 | 0.71 17 | 0.15 43 | 0.18 61 | 0.20 71 | 0.29 55 | 0.31 39 | 0.67 17 | 0.33 30 |
2DHMM-SAS [92] | 35.3 | 0.07 12 | 0.21 31 | 0.07 15 | 0.18 34 | 0.86 71 | 0.18 41 | 0.18 21 | 0.48 48 | 0.19 28 | 0.08 18 | 0.43 37 | 0.08 32 | 0.51 26 | 0.95 27 | 0.35 36 | 0.19 51 | 1.01 42 | 0.14 31 | 0.18 61 | 0.17 27 | 0.40 93 | 0.28 12 | 0.78 31 | 0.32 23 |
Classic+NL [31] | 36.5 | 0.07 12 | 0.20 28 | 0.07 15 | 0.19 46 | 0.79 58 | 0.19 52 | 0.18 21 | 0.37 27 | 0.19 28 | 0.09 44 | 0.43 37 | 0.08 32 | 0.51 26 | 0.94 26 | 0.36 41 | 0.19 51 | 0.92 33 | 0.15 43 | 0.19 73 | 0.17 27 | 0.39 90 | 0.28 12 | 0.78 31 | 0.32 23 |
S2D-Matching [84] | 36.7 | 0.07 12 | 0.21 31 | 0.07 15 | 0.18 34 | 0.79 58 | 0.18 41 | 0.18 21 | 0.47 47 | 0.19 28 | 0.09 44 | 0.37 22 | 0.08 32 | 0.51 26 | 0.97 30 | 0.35 36 | 0.20 57 | 0.91 31 | 0.15 43 | 0.20 84 | 0.17 27 | 0.41 97 | 0.28 12 | 0.71 23 | 0.33 30 |
TV-L1-MCT [64] | 40.4 | 0.07 12 | 0.19 24 | 0.07 15 | 0.22 71 | 0.87 72 | 0.21 73 | 0.18 21 | 0.40 35 | 0.20 52 | 0.09 44 | 0.37 22 | 0.08 32 | 0.61 42 | 1.14 39 | 0.52 68 | 0.22 70 | 1.00 40 | 0.16 54 | 0.13 20 | 0.17 27 | 0.20 17 | 0.29 25 | 0.86 40 | 0.43 54 |
SVFilterOh [111] | 40.5 | 0.09 46 | 0.17 13 | 0.09 62 | 0.20 58 | 0.59 23 | 0.18 41 | 0.19 41 | 0.26 5 | 0.20 52 | 0.10 64 | 0.34 16 | 0.09 60 | 0.37 7 | 0.71 9 | 0.22 6 | 0.19 51 | 0.80 24 | 0.15 43 | 0.26 110 | 0.26 113 | 0.58 111 | 0.28 12 | 0.51 2 | 0.27 3 |
MDP-Flow [26] | 40.6 | 0.08 36 | 0.27 53 | 0.08 44 | 0.18 34 | 0.54 14 | 0.19 52 | 0.18 21 | 0.42 39 | 0.18 16 | 0.08 18 | 0.70 71 | 0.08 32 | 0.63 44 | 1.21 42 | 0.46 59 | 0.17 32 | 1.13 50 | 0.15 43 | 0.14 25 | 0.17 27 | 0.23 20 | 0.35 51 | 1.58 79 | 0.54 73 |
SimpleFlow [49] | 40.6 | 0.07 12 | 0.23 42 | 0.07 15 | 0.21 63 | 0.85 67 | 0.21 73 | 0.19 41 | 0.53 52 | 0.20 52 | 0.09 44 | 0.51 49 | 0.09 60 | 0.56 37 | 1.07 37 | 0.41 50 | 0.18 45 | 0.83 26 | 0.14 31 | 0.16 45 | 0.15 2 | 0.28 52 | 0.28 12 | 0.85 38 | 0.33 30 |
PMF [73] | 41.1 | 0.08 36 | 0.19 24 | 0.07 15 | 0.17 26 | 0.65 36 | 0.15 21 | 0.19 41 | 0.42 39 | 0.20 52 | 0.09 44 | 0.30 10 | 0.09 60 | 0.48 23 | 0.90 23 | 0.24 9 | 0.21 62 | 1.14 52 | 0.16 54 | 0.27 113 | 0.30 123 | 0.50 107 | 0.28 12 | 0.51 2 | 0.27 3 |
IROF-TV [53] | 42.7 | 0.08 36 | 0.22 40 | 0.08 44 | 0.20 58 | 0.85 67 | 0.19 52 | 0.18 21 | 0.40 35 | 0.20 52 | 0.09 44 | 0.79 75 | 0.09 60 | 0.59 41 | 1.06 36 | 0.45 57 | 0.22 70 | 1.74 95 | 0.17 60 | 0.11 3 | 0.16 12 | 0.14 2 | 0.28 12 | 0.85 38 | 0.31 15 |
Adaptive [20] | 43.3 | 0.07 12 | 0.23 42 | 0.06 7 | 0.19 46 | 0.75 50 | 0.17 34 | 0.20 60 | 0.67 65 | 0.19 28 | 0.09 44 | 0.74 72 | 0.08 32 | 0.71 60 | 1.34 50 | 0.56 74 | 0.15 20 | 1.22 59 | 0.11 5 | 0.18 61 | 0.20 71 | 0.28 52 | 0.29 25 | 0.86 40 | 0.33 30 |
OFH [38] | 43.4 | 0.09 46 | 0.27 53 | 0.10 77 | 0.13 3 | 0.60 24 | 0.13 8 | 0.18 21 | 0.63 60 | 0.16 2 | 0.07 10 | 0.61 60 | 0.06 5 | 0.74 66 | 1.54 68 | 0.43 53 | 0.18 45 | 1.47 80 | 0.18 68 | 0.14 25 | 0.18 49 | 0.26 43 | 0.34 50 | 1.60 80 | 0.38 45 |
Occlusion-TV-L1 [63] | 44.0 | 0.08 36 | 0.24 47 | 0.07 15 | 0.17 26 | 0.70 40 | 0.17 34 | 0.19 41 | 0.66 64 | 0.19 28 | 0.09 44 | 0.62 64 | 0.08 32 | 0.75 68 | 1.54 68 | 0.56 74 | 0.14 19 | 1.27 65 | 0.16 54 | 0.12 7 | 0.17 27 | 0.14 2 | 0.37 59 | 1.80 92 | 0.41 50 |
DeepFlow2 [108] | 48.0 | 0.09 46 | 0.33 72 | 0.09 62 | 0.17 26 | 0.73 47 | 0.16 29 | 0.20 60 | 0.74 67 | 0.22 72 | 0.09 44 | 0.82 76 | 0.08 32 | 0.64 45 | 1.37 51 | 0.33 30 | 0.12 6 | 1.24 62 | 0.11 5 | 0.14 25 | 0.15 2 | 0.31 60 | 0.47 80 | 1.50 72 | 0.65 82 |
AggregFlow [97] | 48.7 | 0.09 46 | 0.28 56 | 0.09 62 | 0.23 78 | 1.02 89 | 0.20 66 | 0.22 76 | 0.84 77 | 0.25 85 | 0.10 64 | 0.37 22 | 0.09 60 | 0.65 47 | 1.39 54 | 0.31 23 | 0.12 6 | 0.43 2 | 0.11 5 | 0.12 7 | 0.18 49 | 0.15 7 | 0.43 73 | 1.03 46 | 0.51 68 |
Aniso-Texture [82] | 50.0 | 0.07 12 | 0.17 13 | 0.08 44 | 0.19 46 | 0.49 7 | 0.20 66 | 0.19 41 | 0.46 45 | 0.20 52 | 0.08 18 | 0.37 22 | 0.08 32 | 0.67 52 | 1.44 58 | 0.40 49 | 0.21 62 | 1.12 48 | 0.18 68 | 0.26 110 | 0.27 117 | 0.40 93 | 0.31 39 | 0.93 44 | 0.47 63 |
MLDP_OF [89] | 50.1 | 0.11 84 | 0.27 53 | 0.11 85 | 0.17 26 | 0.55 15 | 0.16 29 | 0.19 41 | 0.41 38 | 0.19 28 | 0.10 64 | 0.36 19 | 0.09 60 | 0.58 40 | 0.98 31 | 0.36 41 | 0.24 81 | 0.62 13 | 0.24 87 | 0.21 91 | 0.20 71 | 0.63 115 | 0.30 31 | 0.77 30 | 0.33 30 |
S2F-IF [123] | 51.3 | 0.09 46 | 0.48 97 | 0.07 15 | 0.22 71 | 0.90 75 | 0.20 66 | 0.21 67 | 0.76 69 | 0.23 74 | 0.08 18 | 0.59 56 | 0.07 14 | 0.82 79 | 1.52 66 | 0.55 72 | 0.13 11 | 1.06 44 | 0.12 9 | 0.12 7 | 0.17 27 | 0.25 37 | 0.47 80 | 1.37 60 | 0.53 71 |
Sparse Occlusion [54] | 51.5 | 0.09 46 | 0.21 31 | 0.08 44 | 0.22 71 | 0.62 29 | 0.22 81 | 0.19 41 | 0.44 43 | 0.19 28 | 0.09 44 | 0.43 37 | 0.08 32 | 0.56 37 | 1.14 39 | 0.31 23 | 0.22 70 | 0.96 36 | 0.17 60 | 0.28 119 | 0.29 120 | 0.38 85 | 0.32 43 | 0.84 37 | 0.35 40 |
RFlow [90] | 51.8 | 0.10 71 | 0.28 56 | 0.10 77 | 0.15 14 | 0.51 11 | 0.15 21 | 0.19 41 | 0.59 57 | 0.18 16 | 0.08 18 | 0.61 60 | 0.08 32 | 0.72 62 | 1.53 67 | 0.48 63 | 0.20 57 | 1.52 83 | 0.17 60 | 0.19 73 | 0.19 64 | 0.35 75 | 0.35 51 | 1.47 68 | 0.39 47 |
Classic++ [32] | 52.2 | 0.07 12 | 0.24 47 | 0.07 15 | 0.18 34 | 0.70 40 | 0.19 52 | 0.19 41 | 0.64 61 | 0.19 28 | 0.09 44 | 0.85 80 | 0.08 32 | 0.73 64 | 1.68 87 | 0.43 53 | 0.20 57 | 1.75 96 | 0.15 43 | 0.20 84 | 0.18 49 | 0.41 97 | 0.30 31 | 1.53 75 | 0.33 30 |
Steered-L1 [118] | 52.3 | 0.08 36 | 0.22 40 | 0.08 44 | 0.12 2 | 0.38 1 | 0.12 2 | 0.17 6 | 0.35 21 | 0.16 2 | 0.08 18 | 0.64 66 | 0.07 14 | 0.75 68 | 1.47 64 | 0.62 82 | 0.21 62 | 1.36 74 | 0.16 54 | 0.28 119 | 0.24 102 | 0.89 125 | 0.42 71 | 1.71 86 | 0.92 97 |
PGM-C [120] | 53.9 | 0.09 46 | 0.50 105 | 0.07 15 | 0.23 78 | 0.92 77 | 0.22 81 | 0.21 67 | 0.83 75 | 0.24 77 | 0.08 18 | 0.85 80 | 0.07 14 | 0.80 74 | 1.59 74 | 0.54 70 | 0.12 6 | 1.33 71 | 0.12 9 | 0.12 7 | 0.16 12 | 0.23 20 | 0.47 80 | 1.48 69 | 0.51 68 |
FlowFields [110] | 55.5 | 0.09 46 | 0.49 101 | 0.07 15 | 0.23 78 | 0.92 77 | 0.21 73 | 0.22 76 | 0.83 75 | 0.24 77 | 0.09 44 | 0.60 58 | 0.08 32 | 0.84 81 | 1.55 71 | 0.59 78 | 0.12 6 | 1.23 61 | 0.12 9 | 0.12 7 | 0.17 27 | 0.25 37 | 0.46 78 | 1.49 70 | 0.44 56 |
CPM-Flow [116] | 55.8 | 0.09 46 | 0.50 105 | 0.07 15 | 0.23 78 | 0.93 80 | 0.22 81 | 0.21 67 | 0.81 73 | 0.24 77 | 0.08 18 | 0.83 77 | 0.07 14 | 0.81 77 | 1.60 76 | 0.55 72 | 0.13 11 | 1.32 70 | 0.12 9 | 0.12 7 | 0.17 27 | 0.23 20 | 0.49 87 | 1.57 78 | 0.54 73 |
TV-L1-improved [17] | 55.8 | 0.07 12 | 0.25 51 | 0.06 7 | 0.16 21 | 0.64 32 | 0.15 21 | 0.19 41 | 0.64 61 | 0.19 28 | 0.08 18 | 0.61 60 | 0.08 32 | 0.73 64 | 1.58 72 | 0.44 55 | 0.33 99 | 1.76 98 | 0.32 103 | 0.24 100 | 0.25 107 | 0.44 103 | 0.32 43 | 1.49 70 | 0.37 42 |
EPPM w/o HM [88] | 56.0 | 0.10 71 | 0.35 78 | 0.09 62 | 0.17 26 | 0.68 38 | 0.15 21 | 0.20 60 | 0.53 52 | 0.20 52 | 0.11 74 | 0.56 53 | 0.10 70 | 0.55 36 | 0.93 25 | 0.31 23 | 0.32 96 | 1.30 68 | 0.25 92 | 0.21 91 | 0.19 64 | 0.64 116 | 0.31 39 | 0.74 25 | 0.30 12 |
EpicFlow [102] | 56.2 | 0.09 46 | 0.50 105 | 0.07 15 | 0.23 78 | 0.94 82 | 0.22 81 | 0.21 67 | 0.93 85 | 0.24 77 | 0.08 18 | 0.84 79 | 0.07 14 | 0.81 77 | 1.61 79 | 0.56 74 | 0.13 11 | 1.35 73 | 0.13 17 | 0.12 7 | 0.16 12 | 0.23 20 | 0.48 85 | 1.52 73 | 0.54 73 |
Kuang [131] | 56.2 | 0.09 46 | 0.48 97 | 0.07 15 | 0.21 63 | 1.03 91 | 0.19 52 | 0.21 67 | 0.93 85 | 0.21 65 | 0.08 18 | 0.65 68 | 0.08 32 | 0.91 90 | 1.67 85 | 0.64 83 | 0.17 32 | 1.40 79 | 0.15 43 | 0.12 7 | 0.16 12 | 0.25 37 | 0.36 56 | 1.37 60 | 0.48 65 |
FlowFields+ [130] | 56.2 | 0.09 46 | 0.49 101 | 0.07 15 | 0.23 78 | 0.92 77 | 0.21 73 | 0.22 76 | 0.84 77 | 0.24 77 | 0.09 44 | 0.59 56 | 0.08 32 | 0.83 80 | 1.51 65 | 0.58 77 | 0.13 11 | 1.18 55 | 0.12 9 | 0.12 7 | 0.18 49 | 0.24 30 | 0.47 80 | 1.45 66 | 0.51 68 |
CostFilter [40] | 56.6 | 0.09 46 | 0.24 47 | 0.09 62 | 0.18 34 | 0.64 32 | 0.17 34 | 0.20 60 | 0.48 48 | 0.21 65 | 0.12 80 | 0.54 50 | 0.12 81 | 0.51 26 | 1.00 33 | 0.25 13 | 0.24 81 | 1.19 57 | 0.19 76 | 0.27 113 | 0.34 126 | 0.54 110 | 0.30 31 | 0.89 42 | 0.30 12 |
NL-TV-NCC [25] | 57.6 | 0.10 71 | 0.23 42 | 0.09 62 | 0.21 63 | 0.72 45 | 0.18 41 | 0.19 41 | 0.39 34 | 0.18 16 | 0.11 74 | 0.41 31 | 0.10 70 | 0.67 52 | 1.29 47 | 0.34 34 | 0.33 99 | 1.50 82 | 0.25 92 | 0.21 91 | 0.20 71 | 0.32 64 | 0.39 65 | 1.07 48 | 0.39 47 |
CombBMOF [113] | 59.0 | 0.10 71 | 0.31 65 | 0.08 44 | 0.22 71 | 0.64 32 | 0.19 52 | 0.19 41 | 0.42 39 | 0.19 28 | 0.10 64 | 0.62 64 | 0.10 70 | 0.57 39 | 0.87 20 | 0.37 45 | 0.41 107 | 1.07 45 | 0.39 111 | 0.21 91 | 0.22 87 | 0.37 80 | 0.35 51 | 0.81 34 | 0.48 65 |
BriefMatch [124] | 59.1 | 0.08 36 | 0.24 47 | 0.08 44 | 0.16 21 | 0.69 39 | 0.14 14 | 0.16 1 | 0.37 27 | 0.16 2 | 0.08 18 | 0.42 34 | 0.08 32 | 0.77 71 | 1.61 79 | 0.60 80 | 0.46 109 | 1.68 92 | 0.39 111 | 0.22 97 | 0.22 87 | 0.66 118 | 0.37 59 | 2.06 99 | 1.09 101 |
Complementary OF [21] | 59.3 | 0.11 84 | 0.31 65 | 0.12 90 | 0.14 9 | 0.55 15 | 0.13 8 | 0.19 41 | 0.48 48 | 0.20 52 | 0.12 80 | 0.55 51 | 0.12 81 | 0.84 81 | 1.58 72 | 0.69 87 | 0.21 62 | 1.24 62 | 0.18 68 | 0.14 25 | 0.17 27 | 0.31 60 | 0.48 85 | 1.70 85 | 0.68 85 |
ACK-Prior [27] | 60.5 | 0.11 84 | 0.26 52 | 0.10 77 | 0.16 21 | 0.52 13 | 0.15 21 | 0.19 41 | 0.35 21 | 0.18 16 | 0.11 74 | 0.38 27 | 0.10 70 | 0.68 54 | 1.23 45 | 0.47 60 | 0.32 96 | 1.30 68 | 0.24 87 | 0.28 119 | 0.21 80 | 0.66 118 | 0.40 67 | 1.43 64 | 0.55 76 |
TF+OM [100] | 60.8 | 0.09 46 | 0.23 42 | 0.08 44 | 0.18 34 | 0.58 20 | 0.18 41 | 0.19 41 | 0.58 56 | 0.23 74 | 0.12 80 | 0.47 46 | 0.12 81 | 0.78 73 | 1.61 79 | 0.54 70 | 0.20 57 | 1.24 62 | 0.18 68 | 0.20 84 | 0.22 87 | 0.37 80 | 0.42 71 | 1.41 63 | 0.46 60 |
ComplOF-FED-GPU [35] | 61.0 | 0.10 71 | 0.32 70 | 0.11 85 | 0.14 9 | 0.80 63 | 0.12 2 | 0.19 41 | 0.57 55 | 0.18 16 | 0.10 64 | 0.68 70 | 0.10 70 | 0.80 74 | 1.60 76 | 0.47 60 | 0.24 81 | 1.75 96 | 0.20 81 | 0.17 55 | 0.17 27 | 0.41 97 | 0.37 59 | 1.73 87 | 0.43 54 |
DeepFlow [86] | 61.2 | 0.11 84 | 0.34 75 | 0.12 90 | 0.19 46 | 0.79 58 | 0.18 41 | 0.21 67 | 0.86 82 | 0.25 85 | 0.11 74 | 0.95 84 | 0.11 78 | 0.65 47 | 1.42 57 | 0.32 28 | 0.13 11 | 1.38 77 | 0.12 9 | 0.14 25 | 0.16 12 | 0.32 64 | 0.55 90 | 1.78 89 | 0.90 95 |
TCOF [69] | 61.5 | 0.10 71 | 0.30 60 | 0.11 85 | 0.23 78 | 0.84 65 | 0.22 81 | 0.24 86 | 0.85 81 | 0.25 85 | 0.21 100 | 0.64 66 | 0.25 101 | 0.80 74 | 1.46 62 | 0.44 55 | 0.13 11 | 0.56 10 | 0.13 17 | 0.19 73 | 0.21 80 | 0.23 20 | 0.32 43 | 0.91 43 | 0.33 30 |
HBM-GC [105] | 62.4 | 0.14 99 | 0.20 28 | 0.14 101 | 0.25 89 | 0.64 32 | 0.24 90 | 0.23 83 | 0.33 14 | 0.25 85 | 0.17 94 | 0.38 27 | 0.16 93 | 0.42 12 | 0.81 14 | 0.28 18 | 0.23 78 | 0.48 6 | 0.21 85 | 0.29 123 | 0.27 117 | 0.47 106 | 0.30 31 | 0.75 27 | 0.38 45 |
CRTflow [80] | 63.8 | 0.09 46 | 0.37 80 | 0.08 44 | 0.19 46 | 0.72 45 | 0.17 34 | 0.20 60 | 0.76 69 | 0.19 28 | 0.10 64 | 0.76 74 | 0.09 60 | 0.68 54 | 1.44 58 | 0.37 45 | 0.49 111 | 1.87 100 | 0.52 116 | 0.15 37 | 0.20 71 | 0.27 48 | 0.46 78 | 1.63 82 | 0.59 80 |
TriangleFlow [30] | 65.6 | 0.10 71 | 0.29 58 | 0.11 85 | 0.19 46 | 0.81 64 | 0.16 29 | 0.20 60 | 0.70 66 | 0.18 16 | 0.08 18 | 0.61 60 | 0.07 14 | 1.03 98 | 1.80 102 | 0.93 97 | 0.42 108 | 1.36 74 | 0.33 107 | 0.19 73 | 0.24 102 | 0.31 60 | 0.37 59 | 1.19 54 | 0.42 53 |
Aniso. Huber-L1 [22] | 66.7 | 0.08 36 | 0.29 58 | 0.08 44 | 0.31 94 | 1.02 89 | 0.32 94 | 0.24 86 | 0.75 68 | 0.28 91 | 0.13 86 | 0.75 73 | 0.12 81 | 0.66 51 | 1.31 49 | 0.42 51 | 0.20 57 | 1.16 54 | 0.17 60 | 0.21 91 | 0.21 80 | 0.32 64 | 0.33 49 | 0.94 45 | 0.41 50 |
SRR-TVOF-NL [91] | 67.2 | 0.11 84 | 0.31 65 | 0.09 62 | 0.19 46 | 0.91 76 | 0.17 34 | 0.21 67 | 0.84 77 | 0.21 65 | 0.10 64 | 0.55 51 | 0.09 60 | 0.74 66 | 1.22 44 | 0.59 78 | 0.22 70 | 1.13 50 | 0.18 68 | 0.24 100 | 0.23 93 | 0.43 102 | 0.45 76 | 1.04 47 | 0.50 67 |
F-TV-L1 [15] | 67.5 | 0.14 99 | 0.35 78 | 0.17 105 | 0.23 78 | 0.99 87 | 0.22 81 | 0.22 76 | 0.88 84 | 0.21 65 | 0.13 86 | 0.99 86 | 0.13 88 | 0.70 59 | 1.54 68 | 0.51 65 | 0.17 32 | 1.56 85 | 0.14 31 | 0.17 55 | 0.19 64 | 0.25 37 | 0.30 31 | 1.27 56 | 0.32 23 |
Rannacher [23] | 67.9 | 0.09 46 | 0.31 65 | 0.09 62 | 0.21 63 | 0.85 67 | 0.20 66 | 0.21 67 | 0.79 72 | 0.21 65 | 0.10 64 | 0.83 77 | 0.10 70 | 0.75 68 | 1.67 85 | 0.45 57 | 0.25 85 | 1.97 103 | 0.19 76 | 0.19 73 | 0.20 71 | 0.37 80 | 0.32 43 | 1.44 65 | 0.35 40 |
LocallyOriented [52] | 68.2 | 0.09 46 | 0.37 80 | 0.08 44 | 0.25 89 | 1.14 95 | 0.22 81 | 0.25 90 | 1.23 99 | 0.26 89 | 0.12 80 | 0.66 69 | 0.11 78 | 0.87 88 | 1.60 76 | 0.52 68 | 0.16 23 | 0.96 36 | 0.16 54 | 0.15 37 | 0.20 71 | 0.30 58 | 0.41 69 | 1.34 58 | 0.46 60 |
ROF-ND [107] | 68.5 | 0.11 84 | 0.32 70 | 0.10 77 | 0.21 63 | 0.70 40 | 0.18 41 | 0.19 41 | 0.42 39 | 0.19 28 | 0.16 92 | 0.48 48 | 0.13 88 | 0.71 60 | 1.37 51 | 0.47 60 | 0.37 104 | 1.12 48 | 0.24 87 | 0.27 113 | 0.24 102 | 0.50 107 | 0.43 73 | 1.52 73 | 0.44 56 |
LDOF [28] | 69.7 | 0.09 46 | 0.38 85 | 0.09 62 | 0.20 58 | 1.19 97 | 0.19 52 | 0.25 90 | 1.01 87 | 0.22 72 | 0.10 64 | 1.96 111 | 0.08 32 | 0.90 89 | 1.76 94 | 0.82 91 | 0.15 20 | 2.20 107 | 0.13 17 | 0.14 25 | 0.18 49 | 0.23 20 | 0.63 99 | 2.20 107 | 0.93 98 |
SIOF [67] | 69.9 | 0.10 71 | 0.23 42 | 0.10 77 | 0.17 26 | 1.08 92 | 0.16 29 | 0.23 83 | 1.12 91 | 0.27 90 | 0.12 80 | 1.29 99 | 0.12 81 | 0.86 87 | 1.70 88 | 0.87 95 | 0.21 62 | 1.34 72 | 0.19 76 | 0.15 37 | 0.17 27 | 0.24 30 | 0.40 67 | 1.66 83 | 0.88 93 |
Second-order prior [8] | 70.7 | 0.09 46 | 0.37 80 | 0.08 44 | 0.19 46 | 1.12 94 | 0.17 34 | 0.22 76 | 1.11 90 | 0.21 65 | 0.07 10 | 1.15 92 | 0.06 5 | 0.85 86 | 1.62 83 | 0.61 81 | 0.31 95 | 2.31 111 | 0.19 76 | 0.24 100 | 0.22 87 | 0.45 104 | 0.35 51 | 1.56 77 | 0.47 63 |
Fusion [6] | 71.4 | 0.09 46 | 0.40 89 | 0.11 85 | 0.17 26 | 0.60 24 | 0.18 41 | 0.19 41 | 0.51 51 | 0.20 52 | 0.09 44 | 1.20 95 | 0.08 32 | 0.93 91 | 1.74 93 | 1.09 101 | 0.29 90 | 1.22 59 | 0.29 98 | 0.25 107 | 0.26 113 | 0.33 70 | 0.50 88 | 1.99 97 | 0.60 81 |
Brox et al. [5] | 72.6 | 0.10 71 | 0.37 80 | 0.12 90 | 0.23 78 | 0.97 85 | 0.23 88 | 0.22 76 | 0.84 77 | 0.20 52 | 0.09 44 | 1.09 87 | 0.08 32 | 1.07 105 | 1.79 99 | 1.90 112 | 0.17 32 | 1.92 102 | 0.17 60 | 0.13 20 | 0.18 49 | 0.15 7 | 0.64 101 | 2.09 100 | 0.90 95 |
DPOF [18] | 73.8 | 0.12 93 | 0.44 93 | 0.09 62 | 0.24 88 | 0.85 67 | 0.21 73 | 0.22 76 | 0.56 54 | 0.23 74 | 0.15 89 | 0.58 55 | 0.13 88 | 0.69 56 | 1.17 41 | 0.42 51 | 0.23 78 | 1.15 53 | 0.18 68 | 0.29 123 | 0.19 64 | 0.73 122 | 0.45 76 | 1.07 48 | 0.58 79 |
DF-Auto [115] | 74.1 | 0.10 71 | 0.43 92 | 0.08 44 | 0.43 102 | 1.24 98 | 0.45 102 | 0.35 101 | 1.21 97 | 0.73 103 | 0.15 89 | 1.12 90 | 0.14 91 | 0.84 81 | 1.61 79 | 0.78 89 | 0.11 3 | 0.87 29 | 0.12 9 | 0.17 55 | 0.23 93 | 0.14 2 | 0.64 101 | 1.45 66 | 0.83 92 |
FlowNetS+ft+v [112] | 74.5 | 0.09 46 | 0.34 75 | 0.09 62 | 0.21 63 | 0.93 80 | 0.20 66 | 0.24 86 | 1.17 94 | 0.32 93 | 0.09 44 | 1.22 96 | 0.09 60 | 1.05 101 | 1.79 99 | 1.25 104 | 0.16 23 | 1.76 98 | 0.14 31 | 0.18 61 | 0.21 80 | 0.34 71 | 0.47 80 | 1.77 88 | 0.74 88 |
CBF [12] | 74.6 | 0.10 71 | 0.30 60 | 0.10 77 | 0.39 98 | 0.88 73 | 0.46 103 | 0.20 60 | 0.64 61 | 0.24 77 | 0.09 44 | 0.96 85 | 0.08 32 | 0.72 62 | 1.44 58 | 0.48 63 | 0.23 78 | 1.27 65 | 0.20 81 | 0.28 119 | 0.26 113 | 0.52 109 | 0.41 69 | 1.26 55 | 0.56 77 |
CLG-TV [48] | 76.1 | 0.09 46 | 0.30 60 | 0.09 62 | 0.34 96 | 0.94 82 | 0.36 95 | 0.25 90 | 0.77 71 | 0.32 93 | 0.19 97 | 0.89 82 | 0.18 97 | 0.77 71 | 1.59 74 | 0.51 65 | 0.22 70 | 1.65 90 | 0.20 81 | 0.20 84 | 0.22 87 | 0.26 43 | 0.38 64 | 1.27 56 | 0.53 71 |
Dynamic MRF [7] | 76.4 | 0.12 93 | 0.37 80 | 0.13 96 | 0.16 21 | 0.77 53 | 0.14 14 | 0.19 41 | 0.81 73 | 0.19 28 | 0.10 64 | 0.90 83 | 0.10 70 | 1.06 104 | 2.10 113 | 0.99 99 | 0.33 99 | 2.77 117 | 0.30 101 | 0.16 45 | 0.17 27 | 0.45 104 | 0.50 88 | 2.78 118 | 1.10 102 |
Local-TV-L1 [65] | 77.5 | 0.13 97 | 0.38 85 | 0.15 102 | 0.38 97 | 1.16 96 | 0.39 97 | 0.33 99 | 1.12 91 | 0.46 97 | 0.19 97 | 1.65 102 | 0.20 99 | 0.64 45 | 1.37 51 | 0.35 36 | 0.18 45 | 1.27 65 | 0.14 31 | 0.16 45 | 0.16 12 | 0.29 55 | 0.78 107 | 1.90 94 | 2.32 114 |
Bartels [41] | 78.7 | 0.11 84 | 0.30 60 | 0.13 96 | 0.21 63 | 0.61 27 | 0.21 73 | 0.21 67 | 0.59 57 | 0.24 77 | 0.16 92 | 0.56 53 | 0.16 93 | 0.84 81 | 1.78 97 | 0.65 84 | 0.25 85 | 1.70 93 | 0.30 101 | 0.22 97 | 0.23 93 | 0.61 113 | 0.35 51 | 1.85 93 | 0.45 58 |
CNN-flow-warp+ref [117] | 79.2 | 0.10 71 | 0.49 101 | 0.10 77 | 0.29 93 | 0.95 84 | 0.30 93 | 0.26 94 | 1.40 102 | 0.50 98 | 0.11 74 | 1.15 92 | 0.10 70 | 1.10 108 | 1.85 104 | 1.67 110 | 0.16 23 | 1.87 100 | 0.15 43 | 0.11 3 | 0.17 27 | 0.19 16 | 0.76 106 | 2.23 109 | 1.23 103 |
SuperFlow [81] | 82.0 | 0.09 46 | 0.31 65 | 0.09 62 | 0.33 95 | 0.99 87 | 0.36 95 | 0.26 94 | 1.13 93 | 0.62 101 | 0.18 96 | 1.33 101 | 0.18 97 | 0.97 95 | 1.65 84 | 1.33 106 | 0.17 32 | 1.57 86 | 0.15 43 | 0.19 73 | 0.23 93 | 0.25 37 | 0.62 98 | 1.96 96 | 0.88 93 |
p-harmonic [29] | 83.0 | 0.12 93 | 0.40 89 | 0.12 90 | 0.22 71 | 0.77 53 | 0.21 73 | 0.24 86 | 0.87 83 | 0.24 77 | 0.13 86 | 1.29 99 | 0.12 81 | 0.98 96 | 1.70 88 | 1.56 109 | 0.26 88 | 1.58 87 | 0.24 87 | 0.20 84 | 0.21 80 | 0.27 48 | 0.39 65 | 1.78 89 | 0.76 89 |
StereoOF-V1MT [119] | 85.7 | 0.11 84 | 0.47 94 | 0.12 90 | 0.18 34 | 1.56 105 | 0.14 14 | 0.27 97 | 1.22 98 | 0.19 28 | 0.11 74 | 1.81 107 | 0.11 78 | 1.40 114 | 2.31 118 | 1.48 107 | 0.40 106 | 2.77 117 | 0.29 98 | 0.14 25 | 0.19 64 | 0.34 71 | 0.92 109 | 2.94 119 | 1.45 106 |
TriFlow [95] | 88.0 | 0.11 84 | 0.34 75 | 0.10 77 | 0.28 92 | 0.89 74 | 0.29 92 | 0.26 94 | 1.31 100 | 0.51 99 | 0.17 94 | 0.60 58 | 0.17 96 | 0.98 96 | 1.78 97 | 1.06 100 | 0.21 62 | 0.84 27 | 0.19 76 | 0.76 128 | 0.31 125 | 1.79 127 | 0.60 96 | 1.36 59 | 0.67 83 |
Shiralkar [42] | 88.9 | 0.12 93 | 0.48 97 | 0.12 90 | 0.16 21 | 1.38 101 | 0.15 21 | 0.23 83 | 1.06 88 | 0.20 52 | 0.12 80 | 1.66 103 | 0.12 81 | 1.05 101 | 2.01 111 | 0.84 93 | 0.57 115 | 2.44 113 | 0.37 110 | 0.24 100 | 0.21 80 | 0.58 111 | 0.56 92 | 2.58 115 | 0.67 83 |
Ad-TV-NDC [36] | 89.1 | 0.23 111 | 0.41 91 | 0.33 120 | 0.82 115 | 2.22 115 | 0.89 117 | 0.64 111 | 1.71 107 | 0.84 106 | 0.38 108 | 1.67 104 | 0.47 111 | 0.65 47 | 1.46 62 | 0.36 41 | 0.21 62 | 1.38 77 | 0.18 68 | 0.16 45 | 0.17 27 | 0.26 43 | 1.26 116 | 2.20 107 | 5.65 127 |
Learning Flow [11] | 89.8 | 0.10 71 | 0.33 72 | 0.09 62 | 0.25 89 | 1.09 93 | 0.25 91 | 0.25 90 | 1.18 96 | 0.28 91 | 0.15 89 | 1.97 112 | 0.14 91 | 1.43 116 | 2.32 119 | 2.38 119 | 0.22 70 | 2.47 115 | 0.21 85 | 0.18 61 | 0.23 93 | 0.30 58 | 0.43 73 | 2.42 113 | 0.72 87 |
SegOF [10] | 90.0 | 0.15 101 | 0.50 105 | 0.08 44 | 0.67 111 | 1.75 109 | 0.70 113 | 0.54 106 | 1.51 105 | 0.92 109 | 0.32 106 | 1.10 88 | 0.28 103 | 1.56 118 | 2.23 117 | 2.37 117 | 0.30 93 | 2.13 106 | 0.26 95 | 0.10 2 | 0.18 49 | 0.14 2 | 0.64 101 | 1.53 75 | 0.70 86 |
StereoFlow [44] | 90.1 | 0.55 129 | 0.93 123 | 0.67 128 | 1.85 126 | 3.09 128 | 1.58 123 | 1.78 128 | 2.38 127 | 1.80 124 | 1.81 125 | 3.31 118 | 1.74 125 | 1.05 101 | 1.72 91 | 0.87 95 | 0.09 1 | 1.18 55 | 0.08 1 | 0.08 1 | 0.15 2 | 0.09 1 | 0.74 105 | 2.10 101 | 1.44 104 |
BlockOverlap [61] | 94.7 | 0.18 107 | 0.30 60 | 0.16 103 | 0.48 103 | 1.24 98 | 0.51 105 | 0.39 102 | 1.42 104 | 0.58 100 | 0.28 104 | 1.12 90 | 0.30 105 | 0.69 56 | 1.45 61 | 0.68 86 | 0.30 93 | 1.37 76 | 0.24 87 | 0.27 113 | 0.25 107 | 0.69 120 | 0.56 92 | 1.62 81 | 3.07 119 |
FlowNet2 [122] | 95.9 | 0.24 114 | 0.64 112 | 0.21 109 | 0.77 113 | 1.66 107 | 0.76 114 | 0.63 110 | 1.41 103 | 1.01 111 | 0.26 102 | 0.47 46 | 0.25 101 | 1.03 98 | 1.73 92 | 0.70 88 | 0.35 103 | 1.19 57 | 0.32 103 | 0.24 100 | 0.28 119 | 0.32 64 | 0.60 96 | 1.38 62 | 0.56 77 |
Modified CLG [34] | 96.4 | 0.19 108 | 0.65 113 | 0.20 106 | 0.60 110 | 1.43 102 | 0.66 111 | 0.65 112 | 1.85 109 | 1.15 115 | 0.40 110 | 1.70 105 | 0.43 108 | 1.19 109 | 2.00 110 | 2.00 114 | 0.19 51 | 2.26 110 | 0.18 68 | 0.14 25 | 0.19 64 | 0.23 20 | 0.99 111 | 2.47 114 | 1.77 109 |
SPSA-learn [13] | 97.4 | 0.17 103 | 0.50 105 | 0.20 106 | 0.50 106 | 1.62 106 | 0.52 106 | 0.50 105 | 1.67 106 | 0.90 107 | 0.34 107 | 1.83 108 | 0.40 107 | 1.07 105 | 1.79 99 | 1.70 111 | 0.29 90 | 1.99 104 | 0.33 107 | 0.16 45 | 0.18 49 | 0.23 20 | 1.03 114 | 2.32 112 | 1.78 110 |
IAOF2 [51] | 99.3 | 0.13 97 | 0.38 85 | 0.13 96 | 0.39 98 | 1.26 100 | 0.41 98 | 0.31 98 | 1.17 94 | 0.40 96 | 1.38 122 | 3.05 116 | 1.52 122 | 0.96 94 | 1.77 95 | 0.94 98 | 0.53 112 | 1.52 83 | 0.35 109 | 0.25 107 | 0.25 107 | 0.35 75 | 0.56 92 | 1.78 89 | 1.06 100 |
GroupFlow [9] | 100.9 | 0.24 114 | 0.71 116 | 0.28 117 | 0.93 119 | 2.46 119 | 0.89 117 | 0.78 113 | 1.98 113 | 0.91 108 | 0.28 104 | 1.16 94 | 0.30 105 | 1.54 117 | 2.43 120 | 0.83 92 | 0.79 119 | 2.82 119 | 1.03 124 | 0.11 3 | 0.18 49 | 0.16 11 | 1.01 113 | 2.26 110 | 1.45 106 |
Black & Anandan [4] | 101.0 | 0.19 108 | 0.50 105 | 0.23 113 | 0.49 104 | 1.82 112 | 0.49 104 | 0.58 108 | 1.73 108 | 0.79 104 | 0.39 109 | 1.88 110 | 0.44 110 | 1.09 107 | 1.92 107 | 1.51 108 | 0.33 99 | 2.22 109 | 0.32 103 | 0.19 73 | 0.23 93 | 0.17 13 | 0.82 108 | 2.19 105 | 1.44 104 |
2bit-BM-tele [98] | 101.2 | 0.23 111 | 0.47 94 | 0.29 118 | 0.40 100 | 0.98 86 | 0.42 99 | 0.34 100 | 1.09 89 | 0.37 95 | 0.27 103 | 1.23 97 | 0.29 104 | 0.84 81 | 1.77 95 | 0.65 84 | 0.48 110 | 1.71 94 | 0.46 114 | 0.31 125 | 0.30 123 | 0.92 126 | 0.56 92 | 2.00 98 | 0.78 90 |
IAOF [50] | 101.2 | 0.17 103 | 0.48 97 | 0.21 109 | 0.57 107 | 1.44 103 | 0.61 108 | 0.56 107 | 1.87 110 | 0.67 102 | 0.62 113 | 1.78 106 | 0.77 114 | 0.93 91 | 1.70 88 | 0.85 94 | 0.55 113 | 2.21 108 | 0.29 98 | 0.23 99 | 0.20 71 | 0.32 64 | 0.99 111 | 1.91 95 | 2.88 118 |
HBpMotionGpu [43] | 101.8 | 0.16 102 | 0.39 88 | 0.13 96 | 0.59 108 | 1.84 113 | 0.62 109 | 0.59 109 | 2.13 118 | 1.05 113 | 0.23 101 | 1.11 89 | 0.24 100 | 0.93 91 | 2.01 111 | 1.25 104 | 0.29 90 | 1.49 81 | 0.26 95 | 0.26 110 | 0.25 107 | 0.37 80 | 0.70 104 | 2.30 111 | 2.03 112 |
2D-CLG [1] | 102.2 | 0.25 116 | 0.96 125 | 0.21 109 | 0.87 116 | 1.80 110 | 0.93 119 | 1.14 119 | 2.15 119 | 1.69 122 | 1.49 124 | 3.45 120 | 1.68 124 | 1.42 115 | 2.11 114 | 2.71 122 | 0.27 89 | 2.44 113 | 0.28 97 | 0.12 7 | 0.16 12 | 0.17 13 | 1.33 117 | 2.76 117 | 2.16 113 |
Filter Flow [19] | 102.4 | 0.17 103 | 0.47 94 | 0.13 96 | 0.42 101 | 1.49 104 | 0.42 99 | 0.41 104 | 1.91 111 | 1.24 116 | 0.66 114 | 3.00 115 | 0.71 113 | 1.27 110 | 1.86 105 | 2.34 116 | 0.32 96 | 1.66 91 | 0.25 92 | 0.27 113 | 0.29 120 | 0.34 71 | 0.55 90 | 1.68 84 | 1.02 99 |
GraphCuts [14] | 102.4 | 0.17 103 | 0.49 101 | 0.16 103 | 0.49 104 | 1.80 110 | 0.44 101 | 0.39 102 | 1.36 101 | 0.82 105 | 0.19 97 | 1.23 97 | 0.16 93 | 1.03 98 | 1.89 106 | 0.78 89 | 0.91 121 | 1.64 88 | 0.32 103 | 0.25 107 | 0.22 87 | 0.64 116 | 0.94 110 | 2.19 105 | 1.81 111 |
UnFlow [129] | 104.9 | 0.46 125 | 0.99 128 | 0.27 115 | 0.92 118 | 2.11 114 | 0.87 116 | 1.08 118 | 2.15 119 | 1.00 110 | 0.53 112 | 1.85 109 | 0.49 112 | 1.81 122 | 2.48 123 | 2.11 115 | 0.58 117 | 2.61 116 | 0.54 117 | 0.16 45 | 0.20 71 | 0.14 2 | 0.63 99 | 2.18 103 | 0.82 91 |
Nguyen [33] | 106.8 | 0.23 111 | 0.61 111 | 0.22 112 | 1.17 121 | 1.72 108 | 1.26 121 | 0.96 116 | 2.15 119 | 1.37 120 | 1.32 120 | 3.25 117 | 1.54 123 | 1.28 111 | 1.96 108 | 1.90 112 | 0.38 105 | 2.38 112 | 0.40 113 | 0.18 61 | 0.19 64 | 0.24 30 | 1.42 118 | 2.67 116 | 2.35 115 |
SILK [79] | 108.5 | 0.29 117 | 0.73 118 | 0.38 122 | 0.76 112 | 2.24 116 | 0.80 115 | 0.88 115 | 2.12 117 | 1.09 114 | 0.41 111 | 2.08 113 | 0.43 108 | 1.76 121 | 2.45 121 | 3.06 123 | 0.57 115 | 3.06 123 | 0.46 114 | 0.15 37 | 0.18 49 | 0.32 64 | 1.53 120 | 3.07 120 | 3.18 120 |
Horn & Schunck [3] | 110.2 | 0.22 110 | 0.67 114 | 0.26 114 | 0.59 108 | 2.41 118 | 0.55 107 | 0.87 114 | 1.97 112 | 1.01 111 | 0.75 115 | 3.59 122 | 0.83 115 | 1.61 119 | 2.22 116 | 2.62 121 | 0.56 114 | 3.11 124 | 0.55 118 | 0.21 91 | 0.25 107 | 0.17 13 | 1.74 121 | 3.29 123 | 2.65 117 |
TI-DOFE [24] | 113.0 | 0.44 124 | 0.82 120 | 0.62 127 | 1.53 125 | 2.70 122 | 1.64 126 | 1.50 124 | 2.33 125 | 1.81 125 | 1.88 126 | 3.69 123 | 2.14 126 | 1.63 120 | 2.20 115 | 2.37 117 | 0.77 118 | 3.02 122 | 0.81 121 | 0.16 45 | 0.21 80 | 0.16 11 | 2.20 125 | 3.62 125 | 3.53 121 |
Periodicity [78] | 114.0 | 0.31 119 | 1.15 129 | 0.20 106 | 0.90 117 | 3.84 129 | 1.08 120 | 2.44 129 | 2.78 129 | 2.17 129 | 0.95 118 | 5.25 129 | 0.90 116 | 6.03 129 | 11.5 131 | 4.83 129 | 4.75 129 | 8.29 131 | 3.21 129 | 0.12 7 | 0.23 93 | 0.15 7 | 2.20 125 | 7.49 129 | 5.15 126 |
Heeger++ [104] | 114.1 | 0.49 127 | 0.76 119 | 0.30 119 | 0.95 120 | 2.91 126 | 0.68 112 | 1.27 120 | 1.98 113 | 1.24 116 | 1.27 119 | 2.82 114 | 1.26 120 | 2.71 126 | 3.10 127 | 3.78 126 | 1.64 127 | 4.13 126 | 1.39 127 | 0.16 45 | 0.24 102 | 0.26 43 | 1.90 122 | 3.25 121 | 3.63 122 |
SLK [47] | 116.7 | 0.33 120 | 0.98 127 | 0.43 124 | 1.50 123 | 2.71 123 | 1.60 125 | 1.34 122 | 2.30 124 | 1.77 123 | 1.98 128 | 3.91 125 | 2.22 127 | 2.14 125 | 2.64 124 | 3.49 125 | 0.92 123 | 3.28 125 | 0.96 123 | 0.17 55 | 0.24 102 | 0.24 30 | 2.89 127 | 4.18 126 | 4.97 125 |
FFV1MT [106] | 117.3 | 0.46 125 | 0.82 120 | 0.27 115 | 0.77 113 | 2.83 125 | 0.65 110 | 1.52 125 | 2.34 126 | 1.59 121 | 1.36 121 | 4.35 127 | 1.33 121 | 2.72 127 | 3.08 126 | 4.06 127 | 1.45 126 | 4.59 128 | 1.24 126 | 0.19 73 | 0.23 93 | 0.35 75 | 1.90 122 | 3.25 121 | 3.63 122 |
Adaptive flow [45] | 118.0 | 0.41 122 | 0.71 116 | 0.44 125 | 1.50 123 | 2.24 116 | 1.58 123 | 1.34 122 | 2.27 123 | 1.90 126 | 0.90 116 | 3.35 119 | 0.99 117 | 1.29 112 | 1.97 109 | 1.24 103 | 0.91 121 | 2.11 105 | 0.65 119 | 0.70 127 | 0.47 128 | 1.88 128 | 1.12 115 | 2.12 102 | 2.48 116 |
HCIC-L [99] | 119.3 | 0.53 128 | 0.84 122 | 0.34 121 | 2.27 128 | 2.56 120 | 3.42 129 | 1.66 127 | 2.05 115 | 2.02 128 | 2.04 129 | 4.10 126 | 2.31 128 | 1.32 113 | 1.80 102 | 1.20 102 | 0.87 120 | 1.64 88 | 0.81 121 | 1.05 129 | 0.76 129 | 1.94 129 | 1.50 119 | 2.18 103 | 1.70 108 |
PGAM+LK [55] | 122.3 | 0.36 121 | 0.93 123 | 0.54 126 | 1.23 122 | 2.78 124 | 1.35 122 | 1.04 117 | 2.09 116 | 1.26 118 | 1.38 122 | 4.86 128 | 1.25 119 | 1.81 122 | 2.47 122 | 2.51 120 | 1.00 124 | 2.91 121 | 0.79 120 | 0.47 126 | 0.36 127 | 0.86 124 | 2.11 124 | 3.52 124 | 4.01 124 |
FOLKI [16] | 122.5 | 0.29 117 | 0.96 125 | 0.39 123 | 1.93 127 | 3.01 127 | 2.64 127 | 1.30 121 | 2.51 128 | 1.34 119 | 0.92 117 | 3.46 121 | 1.21 118 | 2.13 124 | 2.77 125 | 3.35 124 | 1.03 125 | 4.15 127 | 1.21 125 | 0.24 100 | 0.26 113 | 0.74 123 | 3.53 128 | 4.69 127 | 6.56 128 |
Pyramid LK [2] | 124.6 | 0.41 122 | 0.69 115 | 0.77 129 | 2.66 129 | 2.61 121 | 3.17 128 | 1.58 126 | 2.26 122 | 1.94 127 | 1.95 127 | 3.70 124 | 2.60 129 | 4.10 128 | 5.31 128 | 4.43 128 | 3.35 128 | 2.88 120 | 2.91 128 | 0.27 113 | 0.29 120 | 0.61 113 | 6.29 129 | 7.34 128 | 7.54 129 |
AdaConv-v1 [126] | 129.9 | 1.01 130 | 1.20 130 | 1.04 130 | 4.85 130 | 5.45 130 | 4.31 130 | 4.84 130 | 3.27 130 | 4.55 130 | 5.24 130 | 6.19 130 | 5.44 130 | 9.07 130 | 10.1 129 | 8.92 130 | 7.23 130 | 6.02 129 | 6.73 130 | 2.68 130 | 1.63 130 | 3.81 130 | 10.2 130 | 10.8 130 | 9.31 130 |
SepConv-v1 [127] | 129.9 | 1.01 130 | 1.20 130 | 1.04 130 | 4.85 130 | 5.45 130 | 4.31 130 | 4.84 130 | 3.27 130 | 4.55 130 | 5.24 130 | 6.19 130 | 5.44 130 | 9.07 130 | 10.1 129 | 8.92 130 | 7.23 130 | 6.02 129 | 6.73 130 | 2.68 130 | 1.63 130 | 3.81 130 | 10.2 130 | 10.8 130 | 9.31 130 |
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. |