Optical flow evaluation results |
Statistics:
Average
SD
R0.5
R1.0
R2.0
A90
A95
A99
Error type: endpoint angle interpolation normalized interpolation |
A99 normalized interpolation error |
avg. |
Mequon (Hidden texture) im0 GT im1 |
Schefflera (Hidden texture) im0 GT im1 |
Urban (Synthetic) im0 GT im1 |
Teddy (Stereo) im0 GT im1 |
Backyard (High-speed camera) im0 GT im1 |
Basketball (High-speed camera) im0 GT im1 |
Dumptruck (High-speed camera) im0 GT im1 |
Evergreen (High-speed camera) im0 GT 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 | |
PMMST [114] | 9.2 | 1.66 5 | 2.52 2 | 1.60 7 | 2.26 18 | 3.32 10 | 1.61 13 | 2.48 3 | 3.71 5 | 1.62 1 | 2.80 5 | 3.17 11 | 3.22 31 | 2.50 4 | 2.56 4 | 2.63 7 | 2.45 3 | 4.29 5 | 2.36 17 | 1.71 6 | 3.02 6 | 1.71 15 | 2.32 7 | 3.55 11 | 1.45 24 |
MDP-Flow2 [68] | 11.3 | 1.63 2 | 2.49 1 | 1.59 2 | 2.18 9 | 3.35 14 | 1.59 7 | 2.48 3 | 3.83 6 | 1.63 2 | 2.80 5 | 3.11 7 | 3.21 27 | 2.49 1 | 2.55 3 | 2.63 7 | 2.72 32 | 6.03 48 | 2.38 28 | 1.70 4 | 3.01 5 | 1.70 7 | 2.35 14 | 3.57 14 | 1.45 24 |
NNF-Local [87] | 15.9 | 1.63 2 | 2.52 2 | 1.59 2 | 2.02 1 | 2.94 1 | 1.58 3 | 2.46 2 | 3.62 1 | 1.63 2 | 2.90 39 | 3.45 50 | 3.20 19 | 2.49 1 | 2.54 2 | 2.62 4 | 2.96 58 | 6.49 60 | 2.43 51 | 1.71 6 | 3.24 23 | 1.72 31 | 2.32 7 | 3.56 12 | 1.43 2 |
CombBMOF [113] | 16.4 | 1.70 14 | 2.61 6 | 1.62 22 | 2.18 9 | 3.34 12 | 1.57 2 | 2.68 31 | 5.20 60 | 1.79 59 | 2.86 24 | 3.28 26 | 3.23 39 | 2.55 13 | 2.67 18 | 2.63 7 | 2.48 6 | 4.54 8 | 2.32 4 | 1.72 11 | 3.06 10 | 1.70 7 | 2.24 2 | 3.39 2 | 1.41 1 |
NN-field [71] | 19.1 | 1.66 5 | 2.68 15 | 1.60 7 | 2.05 3 | 3.00 4 | 1.56 1 | 2.74 49 | 3.66 2 | 1.66 6 | 2.96 61 | 3.72 84 | 3.24 48 | 2.49 1 | 2.53 1 | 2.61 2 | 2.77 43 | 5.87 42 | 2.39 33 | 1.69 1 | 3.08 13 | 1.71 15 | 2.31 6 | 3.56 12 | 1.44 4 |
PH-Flow [101] | 20.0 | 1.70 14 | 2.66 14 | 1.63 26 | 2.10 6 | 3.14 6 | 1.60 8 | 2.52 6 | 4.27 14 | 1.67 11 | 2.79 3 | 3.07 3 | 3.14 3 | 2.51 5 | 2.59 7 | 2.62 4 | 3.09 77 | 6.87 83 | 2.47 66 | 1.72 11 | 3.37 34 | 1.72 31 | 2.38 19 | 3.68 26 | 1.44 4 |
nLayers [57] | 24.5 | 1.71 18 | 2.70 19 | 1.63 26 | 2.19 11 | 3.30 9 | 1.61 13 | 2.53 8 | 3.90 7 | 1.66 6 | 2.83 11 | 3.27 22 | 3.18 13 | 2.64 51 | 2.84 57 | 2.72 54 | 2.93 56 | 6.57 65 | 2.48 73 | 1.71 6 | 2.99 4 | 1.72 31 | 2.32 7 | 3.59 17 | 1.44 4 |
IROF++ [58] | 26.7 | 1.70 14 | 2.63 9 | 1.61 16 | 2.33 26 | 3.53 24 | 1.60 8 | 2.73 45 | 4.75 37 | 1.73 36 | 2.79 3 | 3.07 3 | 3.21 27 | 2.55 13 | 2.70 23 | 2.69 44 | 2.60 13 | 5.43 28 | 2.34 11 | 1.76 40 | 3.48 47 | 1.72 31 | 2.47 38 | 3.95 60 | 1.46 44 |
FMOF [94] | 27.6 | 1.78 62 | 2.84 42 | 1.66 55 | 2.24 15 | 3.35 14 | 1.60 8 | 2.80 57 | 5.55 76 | 1.79 59 | 2.93 53 | 3.50 54 | 3.23 39 | 2.52 6 | 2.61 9 | 2.65 12 | 2.64 22 | 5.11 16 | 2.36 17 | 1.69 1 | 2.89 2 | 1.68 1 | 2.35 14 | 3.66 24 | 1.44 4 |
NNF-EAC [103] | 29.0 | 1.84 83 | 2.68 15 | 1.72 80 | 2.36 30 | 3.60 31 | 1.61 13 | 2.60 13 | 4.20 11 | 1.68 13 | 2.92 46 | 3.31 33 | 3.30 82 | 2.55 13 | 2.67 18 | 2.63 7 | 2.46 5 | 4.34 6 | 2.34 11 | 1.73 18 | 3.41 39 | 1.73 63 | 2.38 19 | 3.65 22 | 1.45 24 |
Layers++ [37] | 30.5 | 1.72 22 | 2.64 10 | 1.65 48 | 2.04 2 | 2.95 2 | 1.61 13 | 2.60 13 | 5.19 57 | 1.68 13 | 2.85 18 | 3.29 29 | 3.24 48 | 2.64 51 | 2.86 74 | 2.71 50 | 3.16 84 | 7.50 104 | 2.41 39 | 1.69 1 | 2.86 1 | 1.68 1 | 2.32 7 | 3.60 20 | 1.45 24 |
WLIF-Flow [93] | 33.8 | 1.70 14 | 2.64 10 | 1.62 22 | 2.40 35 | 3.73 40 | 1.65 39 | 2.60 13 | 4.43 19 | 1.66 6 | 2.87 31 | 3.14 8 | 3.31 85 | 2.57 21 | 2.70 23 | 2.69 44 | 3.18 87 | 6.96 88 | 2.53 88 | 1.71 6 | 3.09 14 | 1.71 15 | 2.41 28 | 3.76 32 | 1.46 44 |
COFM [59] | 33.9 | 1.72 22 | 2.68 15 | 1.65 48 | 2.31 25 | 3.53 24 | 1.62 23 | 2.58 11 | 4.59 31 | 1.68 13 | 2.81 8 | 3.24 18 | 3.14 3 | 2.54 10 | 2.66 14 | 2.63 7 | 3.29 95 | 7.42 100 | 2.46 63 | 1.74 25 | 3.43 43 | 1.75 80 | 2.37 18 | 3.70 27 | 1.49 90 |
HAST [109] | 35.0 | 1.69 13 | 2.61 6 | 1.63 26 | 2.17 8 | 3.24 8 | 1.61 13 | 2.83 61 | 5.94 94 | 1.73 36 | 2.77 1 | 3.00 1 | 3.13 1 | 2.60 29 | 2.76 36 | 2.67 23 | 3.25 92 | 7.71 115 | 2.43 51 | 1.74 25 | 3.39 35 | 1.71 15 | 2.54 54 | 4.02 73 | 1.45 24 |
2DHMM-SAS [92] | 35.2 | 1.76 45 | 2.76 27 | 1.63 26 | 2.95 75 | 4.41 78 | 1.67 46 | 2.68 31 | 5.20 60 | 1.68 13 | 2.86 24 | 3.30 31 | 3.16 6 | 2.60 29 | 2.78 42 | 2.73 61 | 2.61 15 | 5.10 15 | 2.33 7 | 1.74 25 | 3.42 41 | 1.72 31 | 2.54 54 | 3.94 58 | 1.44 4 |
LME [70] | 35.3 | 1.67 9 | 2.55 4 | 1.60 7 | 2.34 28 | 3.54 27 | 1.80 74 | 2.68 31 | 5.37 69 | 1.68 13 | 2.85 18 | 3.35 42 | 3.23 39 | 2.69 99 | 2.92 99 | 2.88 124 | 2.72 32 | 5.65 36 | 2.38 28 | 1.70 4 | 2.93 3 | 1.70 7 | 2.39 25 | 3.67 25 | 1.44 4 |
DeepFlow2 [108] | 35.8 | 1.77 54 | 2.84 42 | 1.68 63 | 2.72 61 | 4.23 66 | 1.74 63 | 2.68 31 | 4.82 41 | 1.73 36 | 2.95 58 | 3.32 35 | 3.23 39 | 2.57 21 | 2.67 18 | 2.68 32 | 2.45 3 | 4.04 2 | 2.34 11 | 1.73 18 | 3.22 20 | 1.72 31 | 2.39 25 | 3.65 22 | 1.47 66 |
DeepFlow [86] | 36.4 | 1.76 45 | 2.90 51 | 1.68 63 | 2.78 64 | 4.31 71 | 1.87 81 | 2.71 39 | 4.80 40 | 1.73 36 | 2.99 66 | 3.29 29 | 3.25 53 | 2.58 23 | 2.70 23 | 2.69 44 | 2.50 8 | 4.10 3 | 2.39 33 | 1.71 6 | 3.05 9 | 1.70 7 | 2.38 19 | 3.59 17 | 1.46 44 |
SepConv-v1 [127] | 37.0 | 1.49 1 | 2.64 10 | 1.41 1 | 2.52 43 | 3.59 30 | 1.75 66 | 2.25 1 | 4.60 32 | 2.02 96 | 3.21 94 | 3.50 54 | 3.37 103 | 2.52 6 | 2.57 5 | 2.61 2 | 2.30 1 | 3.53 1 | 2.25 1 | 1.80 67 | 3.57 53 | 1.82 106 | 1.87 1 | 2.69 1 | 1.52 113 |
DPOF [18] | 37.1 | 1.80 67 | 3.31 110 | 1.68 63 | 2.11 7 | 3.15 7 | 1.61 13 | 3.18 92 | 4.34 17 | 1.92 88 | 2.91 43 | 3.63 73 | 3.20 19 | 2.55 13 | 2.66 14 | 2.65 12 | 2.66 25 | 5.32 23 | 2.33 7 | 1.74 25 | 3.17 16 | 1.72 31 | 2.47 38 | 3.83 43 | 1.46 44 |
FlowFields [110] | 37.5 | 1.72 22 | 2.96 65 | 1.61 16 | 2.26 18 | 3.41 17 | 1.62 23 | 2.66 24 | 4.89 43 | 1.68 13 | 2.92 46 | 3.60 68 | 3.23 39 | 2.58 23 | 2.73 28 | 2.70 48 | 2.98 62 | 6.29 53 | 2.47 66 | 1.72 11 | 3.28 25 | 1.72 31 | 2.57 66 | 4.09 88 | 1.44 4 |
TV-L1-MCT [64] | 37.9 | 1.77 54 | 2.83 39 | 1.65 48 | 2.60 50 | 4.03 54 | 1.63 30 | 2.71 39 | 5.86 91 | 1.70 30 | 2.83 11 | 3.22 17 | 3.20 19 | 2.66 70 | 2.90 90 | 2.71 50 | 2.65 24 | 5.24 19 | 2.38 28 | 1.74 25 | 3.22 20 | 1.72 31 | 2.34 12 | 3.57 14 | 1.46 44 |
ALD-Flow [66] | 38.3 | 1.82 79 | 2.93 59 | 1.70 74 | 2.56 46 | 3.95 49 | 1.71 55 | 2.69 38 | 4.46 20 | 1.73 36 | 2.85 18 | 3.24 18 | 3.22 31 | 2.58 23 | 2.70 23 | 2.74 70 | 2.55 10 | 4.16 4 | 2.42 46 | 1.72 11 | 3.02 6 | 1.71 15 | 2.57 66 | 4.04 78 | 1.46 44 |
ComponentFusion [96] | 38.8 | 1.68 10 | 2.73 22 | 1.60 7 | 2.24 15 | 3.48 21 | 1.58 3 | 2.67 27 | 4.50 25 | 1.73 36 | 2.83 11 | 3.28 26 | 3.16 6 | 2.64 51 | 2.81 48 | 2.67 23 | 2.74 37 | 6.02 46 | 2.38 28 | 1.84 84 | 4.82 102 | 1.74 76 | 2.65 90 | 4.51 113 | 1.45 24 |
Aniso. Huber-L1 [22] | 39.3 | 1.81 75 | 2.96 65 | 1.70 74 | 3.33 95 | 4.67 94 | 1.84 77 | 2.83 61 | 4.22 12 | 1.79 59 | 2.89 36 | 3.27 22 | 3.23 39 | 2.58 23 | 2.73 28 | 2.67 23 | 2.58 11 | 4.86 13 | 2.32 4 | 1.73 18 | 3.06 10 | 1.71 15 | 2.36 17 | 3.51 7 | 1.47 66 |
Sparse-NonSparse [56] | 39.3 | 1.72 22 | 2.75 26 | 1.63 26 | 2.33 26 | 3.56 29 | 1.61 13 | 2.68 31 | 5.43 72 | 1.68 13 | 2.85 18 | 3.27 22 | 3.16 6 | 2.64 51 | 2.84 57 | 2.72 54 | 3.06 71 | 6.56 62 | 2.45 57 | 1.78 56 | 4.33 88 | 1.71 15 | 2.55 59 | 3.99 66 | 1.44 4 |
PMF [73] | 40.0 | 1.65 4 | 2.62 8 | 1.59 2 | 2.30 24 | 3.47 20 | 1.58 3 | 2.86 69 | 6.90 109 | 1.78 56 | 2.85 18 | 3.27 22 | 3.22 31 | 2.61 35 | 2.77 40 | 2.65 12 | 2.74 37 | 5.28 22 | 2.52 83 | 1.76 40 | 3.87 68 | 1.73 63 | 2.63 87 | 4.29 103 | 1.44 4 |
SuperFlow [81] | 42.7 | 1.78 62 | 2.87 47 | 1.72 80 | 2.90 69 | 4.24 67 | 2.09 100 | 2.80 57 | 4.33 16 | 1.80 70 | 3.01 73 | 3.40 45 | 3.25 53 | 2.54 10 | 2.62 10 | 2.72 54 | 2.43 2 | 4.36 7 | 2.30 2 | 1.76 40 | 3.46 44 | 1.73 63 | 2.30 5 | 3.44 4 | 1.46 44 |
AggregFlow [97] | 43.2 | 1.84 83 | 3.22 104 | 1.70 74 | 2.56 46 | 3.91 46 | 1.74 63 | 2.52 6 | 3.95 8 | 1.63 2 | 2.92 46 | 3.51 57 | 3.25 53 | 2.56 20 | 2.66 14 | 2.68 32 | 2.61 15 | 4.82 11 | 2.42 46 | 1.79 60 | 4.27 82 | 1.73 63 | 2.42 29 | 3.89 54 | 1.45 24 |
SRR-TVOF-NL [91] | 44.2 | 1.82 79 | 3.04 81 | 1.68 63 | 2.69 57 | 4.07 58 | 1.77 71 | 2.60 13 | 4.16 9 | 1.73 36 | 2.83 11 | 3.18 13 | 3.16 6 | 2.66 70 | 2.89 88 | 2.75 80 | 2.70 30 | 5.75 38 | 2.33 7 | 1.73 18 | 3.12 15 | 1.71 15 | 2.58 73 | 4.07 85 | 1.46 44 |
FlowFields+ [130] | 44.2 | 1.71 18 | 2.97 68 | 1.60 7 | 2.22 13 | 3.34 12 | 1.63 30 | 2.66 24 | 5.01 50 | 1.68 13 | 2.91 43 | 3.61 70 | 3.24 48 | 2.65 59 | 2.85 66 | 2.74 70 | 3.12 79 | 7.10 90 | 2.48 73 | 1.72 11 | 3.22 20 | 1.72 31 | 2.57 66 | 4.15 97 | 1.44 4 |
Brox et al. [5] | 45.0 | 1.77 54 | 3.05 84 | 1.63 26 | 2.69 57 | 4.02 53 | 1.73 60 | 2.86 69 | 5.19 57 | 1.81 74 | 2.92 46 | 3.19 14 | 3.24 48 | 2.58 23 | 2.68 21 | 2.68 32 | 2.96 58 | 6.85 82 | 2.41 39 | 1.78 56 | 3.73 60 | 1.72 31 | 2.32 7 | 3.48 6 | 1.45 24 |
TC/T-Flow [76] | 45.2 | 1.80 67 | 2.89 49 | 1.64 45 | 2.58 48 | 4.01 52 | 1.65 39 | 2.60 13 | 4.17 10 | 1.70 30 | 2.86 24 | 3.21 16 | 3.20 19 | 2.65 59 | 2.84 57 | 2.79 105 | 2.61 15 | 4.93 14 | 2.37 20 | 1.97 106 | 6.08 113 | 1.75 80 | 2.52 48 | 3.86 51 | 1.44 4 |
EPPM w/o HM [88] | 45.2 | 1.66 5 | 2.77 29 | 1.59 2 | 2.42 36 | 3.79 41 | 1.61 13 | 3.37 97 | 10.6 128 | 1.89 81 | 2.90 39 | 3.57 64 | 3.22 31 | 2.55 13 | 2.65 13 | 2.65 12 | 2.77 43 | 5.61 32 | 2.41 39 | 1.80 67 | 4.56 95 | 1.75 80 | 2.54 54 | 3.99 66 | 1.44 4 |
AGIF+OF [85] | 45.5 | 1.73 30 | 2.81 35 | 1.61 16 | 2.39 33 | 3.69 36 | 1.63 30 | 2.71 39 | 5.15 55 | 1.73 36 | 2.84 17 | 3.17 11 | 3.18 13 | 2.69 99 | 2.93 103 | 2.77 92 | 3.18 87 | 6.82 80 | 2.43 51 | 1.75 33 | 3.20 19 | 1.68 1 | 2.60 81 | 4.12 92 | 1.43 2 |
IROF-TV [53] | 46.0 | 1.79 66 | 2.91 53 | 1.68 63 | 2.44 39 | 3.69 36 | 1.63 30 | 2.81 60 | 5.61 79 | 1.76 54 | 2.85 18 | 3.30 31 | 3.21 27 | 2.65 59 | 2.85 66 | 2.80 112 | 2.80 45 | 6.02 46 | 2.35 14 | 1.76 40 | 3.34 32 | 1.72 31 | 2.38 19 | 3.59 17 | 1.47 66 |
DF-Auto [115] | 46.0 | 1.80 67 | 2.81 35 | 1.71 78 | 2.91 70 | 4.40 77 | 2.04 96 | 2.64 21 | 4.85 42 | 1.68 13 | 2.96 61 | 3.51 57 | 3.21 27 | 2.55 13 | 2.64 11 | 2.68 32 | 2.61 15 | 5.24 19 | 2.39 33 | 1.77 51 | 3.67 57 | 1.73 63 | 2.50 46 | 3.88 53 | 1.47 66 |
S2F-IF [123] | 46.0 | 1.71 18 | 2.94 61 | 1.60 7 | 2.20 12 | 3.32 10 | 1.62 23 | 2.65 23 | 5.27 63 | 1.68 13 | 2.89 36 | 3.57 64 | 3.16 6 | 2.65 59 | 2.85 66 | 2.76 91 | 3.04 68 | 6.81 79 | 2.48 73 | 1.75 33 | 3.46 44 | 1.73 63 | 2.58 73 | 4.14 96 | 1.45 24 |
ProbFlowFields [128] | 46.2 | 1.73 30 | 3.11 93 | 1.63 26 | 2.24 15 | 3.43 19 | 1.60 8 | 2.57 10 | 4.52 28 | 1.66 6 | 2.92 46 | 3.61 70 | 3.26 64 | 2.66 70 | 2.87 78 | 2.78 99 | 3.37 102 | 7.50 104 | 2.55 96 | 1.72 11 | 3.31 28 | 1.72 31 | 2.38 19 | 3.76 32 | 1.45 24 |
OFLAF [77] | 46.9 | 1.66 5 | 2.60 5 | 1.60 7 | 2.09 5 | 3.08 5 | 1.58 3 | 2.58 11 | 4.23 13 | 1.63 2 | 2.78 2 | 3.09 5 | 3.16 6 | 2.66 70 | 2.88 82 | 2.74 70 | 3.41 108 | 7.61 109 | 2.54 90 | 2.02 110 | 6.45 115 | 1.75 80 | 2.67 98 | 4.21 101 | 1.45 24 |
CPM-Flow [116] | 47.1 | 1.75 41 | 2.97 68 | 1.63 26 | 2.29 23 | 3.49 23 | 1.65 39 | 2.75 50 | 4.76 38 | 1.68 13 | 3.00 70 | 3.88 94 | 3.28 69 | 2.63 44 | 2.81 48 | 2.75 80 | 2.64 22 | 5.22 18 | 2.41 39 | 1.77 51 | 3.96 71 | 1.72 31 | 2.51 47 | 3.95 60 | 1.47 66 |
Ramp [62] | 47.2 | 1.76 45 | 2.82 38 | 1.66 55 | 2.38 32 | 3.66 34 | 1.63 30 | 2.66 24 | 5.19 57 | 1.67 11 | 2.81 8 | 3.15 10 | 3.15 5 | 2.63 44 | 2.84 57 | 2.73 61 | 3.38 104 | 7.56 108 | 2.53 88 | 1.78 56 | 4.02 74 | 1.70 7 | 2.61 82 | 4.05 80 | 1.45 24 |
LSM [39] | 47.5 | 1.74 39 | 2.81 35 | 1.63 26 | 2.35 29 | 3.55 28 | 1.61 13 | 2.72 43 | 5.64 83 | 1.68 13 | 2.86 24 | 3.26 21 | 3.18 13 | 2.66 70 | 2.88 82 | 2.74 70 | 3.15 83 | 6.90 84 | 2.45 57 | 1.79 60 | 4.52 94 | 1.71 15 | 2.59 76 | 4.04 78 | 1.44 4 |
CLG-TV [48] | 47.6 | 1.80 67 | 2.95 63 | 1.71 78 | 3.19 87 | 4.55 87 | 1.83 76 | 2.89 75 | 5.00 48 | 1.91 86 | 2.96 61 | 3.35 42 | 3.28 69 | 2.60 29 | 2.75 33 | 2.67 23 | 2.54 9 | 4.63 10 | 2.36 17 | 1.73 18 | 3.17 16 | 1.72 31 | 2.43 30 | 3.61 21 | 1.47 66 |
MDP-Flow [26] | 47.6 | 1.68 10 | 2.77 29 | 1.60 7 | 2.27 20 | 3.53 24 | 1.62 23 | 2.54 9 | 3.70 4 | 1.73 36 | 3.04 79 | 3.73 86 | 3.28 69 | 2.61 35 | 2.78 42 | 2.78 99 | 3.80 119 | 8.39 120 | 2.63 113 | 1.74 25 | 3.25 24 | 1.73 63 | 2.46 36 | 3.84 47 | 1.45 24 |
SVFilterOh [111] | 47.8 | 1.72 22 | 2.71 20 | 1.66 55 | 2.22 13 | 3.41 17 | 1.62 23 | 2.73 45 | 4.73 35 | 1.71 32 | 2.86 24 | 3.14 8 | 3.34 98 | 2.69 99 | 2.89 88 | 2.83 114 | 2.81 47 | 6.16 51 | 2.43 51 | 1.73 18 | 3.02 6 | 1.75 80 | 2.49 41 | 3.94 58 | 1.50 103 |
Classic+NL [31] | 47.9 | 1.80 67 | 2.85 46 | 1.68 63 | 2.43 38 | 3.68 35 | 1.63 30 | 2.64 21 | 5.45 74 | 1.68 13 | 2.86 24 | 3.33 37 | 3.22 31 | 2.64 51 | 2.84 57 | 2.72 54 | 3.05 69 | 6.39 58 | 2.46 63 | 1.78 56 | 4.29 84 | 1.71 15 | 2.57 66 | 4.03 74 | 1.45 24 |
SIOF [67] | 48.3 | 1.88 92 | 2.97 68 | 1.72 80 | 3.34 96 | 4.70 99 | 2.11 101 | 2.73 45 | 5.27 63 | 1.79 59 | 2.92 46 | 3.54 61 | 3.22 31 | 2.52 6 | 2.58 6 | 2.66 20 | 2.59 12 | 4.84 12 | 2.37 20 | 1.73 18 | 3.28 25 | 1.72 31 | 2.52 48 | 3.79 35 | 1.48 85 |
CostFilter [40] | 49.8 | 1.68 10 | 2.78 31 | 1.59 2 | 2.27 20 | 3.38 16 | 1.60 8 | 3.01 85 | 9.85 125 | 1.79 59 | 2.87 31 | 3.33 37 | 3.17 12 | 2.65 59 | 2.83 54 | 2.70 48 | 2.80 45 | 5.62 34 | 2.59 106 | 1.79 60 | 4.12 76 | 1.73 63 | 2.71 101 | 4.46 110 | 1.44 4 |
CBF [12] | 49.9 | 1.76 45 | 2.79 32 | 1.68 63 | 2.92 71 | 4.29 68 | 1.84 77 | 2.83 61 | 4.32 15 | 1.79 59 | 2.98 64 | 3.28 26 | 3.42 109 | 2.54 10 | 2.60 8 | 2.74 70 | 2.66 25 | 5.39 25 | 2.40 37 | 1.79 60 | 3.33 29 | 1.78 96 | 2.38 19 | 3.53 9 | 1.56 120 |
PGM-C [120] | 50.2 | 1.75 41 | 3.04 81 | 1.63 26 | 2.28 22 | 3.48 21 | 1.63 30 | 2.87 72 | 4.92 46 | 1.68 13 | 2.93 53 | 3.64 75 | 3.25 53 | 2.64 51 | 2.83 54 | 2.75 80 | 2.75 39 | 5.77 39 | 2.41 39 | 1.77 51 | 3.88 69 | 1.71 15 | 2.63 87 | 4.31 104 | 1.46 44 |
Kuang [131] | 50.2 | 1.75 41 | 3.06 86 | 1.63 26 | 2.39 33 | 3.61 32 | 1.62 23 | 2.89 75 | 5.63 82 | 1.73 36 | 2.94 56 | 3.70 83 | 3.25 53 | 2.63 44 | 2.81 48 | 2.74 70 | 2.73 35 | 5.74 37 | 2.37 20 | 1.83 82 | 4.01 73 | 1.76 87 | 2.46 36 | 3.83 43 | 1.44 4 |
Second-order prior [8] | 50.2 | 1.81 75 | 2.92 56 | 1.72 80 | 3.18 86 | 4.60 92 | 1.75 66 | 3.27 95 | 6.35 101 | 1.94 91 | 2.92 46 | 3.43 48 | 3.20 19 | 2.60 29 | 2.77 40 | 2.66 20 | 2.63 19 | 5.64 35 | 2.38 28 | 1.74 25 | 3.17 16 | 1.71 15 | 2.49 41 | 3.80 39 | 1.46 44 |
FC-2Layers-FF [74] | 50.8 | 1.73 30 | 2.80 34 | 1.63 26 | 2.05 3 | 2.95 2 | 1.62 23 | 2.60 13 | 5.12 52 | 1.68 13 | 2.83 11 | 3.32 35 | 3.20 19 | 2.67 84 | 2.90 90 | 2.75 80 | 3.51 114 | 7.68 114 | 2.56 100 | 1.81 74 | 4.87 104 | 1.72 31 | 2.53 53 | 4.00 71 | 1.46 44 |
RNLOD-Flow [121] | 51.2 | 1.72 22 | 2.73 22 | 1.64 45 | 2.72 61 | 4.21 64 | 1.65 39 | 2.79 55 | 6.73 105 | 1.79 59 | 2.80 5 | 3.06 2 | 3.20 19 | 2.68 88 | 2.93 103 | 2.74 70 | 3.06 71 | 6.61 69 | 2.47 66 | 1.75 33 | 3.39 35 | 1.72 31 | 2.59 76 | 3.98 64 | 1.45 24 |
p-harmonic [29] | 52.0 | 1.73 30 | 2.84 42 | 1.63 26 | 3.30 94 | 4.71 100 | 1.85 80 | 2.84 68 | 5.62 80 | 1.89 81 | 3.09 88 | 3.60 68 | 3.25 53 | 2.62 40 | 2.78 42 | 2.67 23 | 2.73 35 | 5.57 31 | 2.42 46 | 1.76 40 | 3.48 47 | 1.72 31 | 2.43 30 | 3.73 30 | 1.46 44 |
OAR-Flow [125] | 52.8 | 1.80 67 | 2.95 63 | 1.67 60 | 2.64 53 | 4.11 62 | 1.73 60 | 2.67 27 | 4.50 25 | 1.71 32 | 2.83 11 | 3.20 15 | 3.18 13 | 2.65 59 | 2.85 66 | 2.77 92 | 2.97 60 | 6.36 55 | 2.48 73 | 1.90 95 | 4.47 92 | 1.73 63 | 2.48 40 | 3.80 39 | 1.46 44 |
LDOF [28] | 53.2 | 1.93 99 | 3.01 76 | 1.83 109 | 2.79 65 | 3.86 42 | 2.27 107 | 3.01 85 | 5.20 60 | 1.92 88 | 2.99 66 | 3.56 62 | 3.28 69 | 2.55 13 | 2.64 11 | 2.68 32 | 2.60 13 | 5.11 16 | 2.35 14 | 1.76 40 | 3.55 50 | 1.72 31 | 2.44 33 | 3.74 31 | 1.47 66 |
S2D-Matching [84] | 53.4 | 1.77 54 | 2.91 53 | 1.66 55 | 2.82 66 | 4.29 68 | 1.67 46 | 2.68 31 | 5.38 70 | 1.72 34 | 2.86 24 | 3.24 18 | 3.23 39 | 2.66 70 | 2.87 78 | 2.68 32 | 3.36 101 | 7.27 96 | 2.54 90 | 1.76 40 | 3.33 29 | 1.70 7 | 2.56 63 | 4.03 74 | 1.46 44 |
ComplOF-FED-GPU [35] | 54.5 | 1.76 45 | 3.10 92 | 1.65 48 | 2.51 42 | 3.88 44 | 1.68 49 | 3.41 98 | 4.48 24 | 2.00 94 | 2.88 34 | 3.34 40 | 3.22 31 | 2.63 44 | 2.81 48 | 2.73 61 | 2.75 39 | 5.56 29 | 2.41 39 | 1.81 74 | 3.69 58 | 1.72 31 | 2.65 90 | 4.09 88 | 1.47 66 |
FESL [72] | 54.7 | 1.76 45 | 2.76 27 | 1.64 45 | 2.37 31 | 3.64 33 | 1.61 13 | 2.75 50 | 6.26 97 | 1.76 54 | 2.90 39 | 3.31 33 | 3.23 39 | 2.66 70 | 2.87 78 | 2.75 80 | 3.38 104 | 7.61 109 | 2.55 96 | 1.76 40 | 4.31 87 | 1.69 5 | 2.56 63 | 4.00 71 | 1.44 4 |
TCOF [69] | 55.0 | 1.76 45 | 2.74 24 | 1.63 26 | 3.50 111 | 5.03 120 | 1.89 82 | 2.60 13 | 4.72 34 | 1.66 6 | 2.89 36 | 3.33 37 | 3.26 64 | 2.60 29 | 2.76 36 | 2.65 12 | 3.06 71 | 6.77 77 | 2.42 46 | 1.82 77 | 4.62 97 | 1.71 15 | 2.64 89 | 4.06 82 | 1.49 90 |
Classic+CPF [83] | 56.3 | 1.73 30 | 2.74 24 | 1.61 16 | 2.44 39 | 3.69 36 | 1.63 30 | 2.73 45 | 5.55 76 | 1.73 36 | 2.81 8 | 3.10 6 | 3.13 1 | 2.79 113 | 3.10 115 | 2.77 92 | 3.37 102 | 7.37 98 | 2.49 80 | 1.86 88 | 4.62 97 | 1.69 5 | 2.72 103 | 4.42 107 | 1.44 4 |
Efficient-NL [60] | 57.6 | 1.72 22 | 2.65 13 | 1.62 22 | 2.64 53 | 4.05 55 | 1.63 30 | 3.58 104 | 5.43 72 | 2.12 100 | 2.87 31 | 3.40 45 | 3.18 13 | 2.61 35 | 2.80 46 | 2.73 61 | 3.33 97 | 7.51 106 | 2.45 57 | 1.82 77 | 4.50 93 | 1.72 31 | 2.72 103 | 4.13 93 | 1.45 24 |
TF+OM [100] | 59.5 | 1.78 62 | 2.92 56 | 1.69 73 | 2.42 36 | 3.70 39 | 1.90 85 | 2.86 69 | 5.64 83 | 1.74 53 | 3.03 77 | 3.64 75 | 3.28 69 | 2.64 51 | 2.80 46 | 2.72 54 | 2.76 42 | 5.80 40 | 2.43 51 | 1.82 77 | 3.85 65 | 1.73 63 | 2.49 41 | 3.76 32 | 1.49 90 |
MLDP_OF [89] | 59.7 | 1.71 18 | 2.71 20 | 1.62 22 | 2.73 63 | 4.21 64 | 1.66 43 | 2.60 13 | 4.46 20 | 1.68 13 | 3.03 77 | 3.34 40 | 3.39 105 | 2.65 59 | 2.84 57 | 2.75 80 | 3.27 93 | 6.95 87 | 2.68 121 | 1.80 67 | 3.64 56 | 1.77 93 | 2.54 54 | 3.98 64 | 1.50 103 |
FlowNet2 [122] | 61.7 | 2.11 115 | 3.47 117 | 1.80 104 | 2.64 53 | 3.87 43 | 1.90 85 | 2.94 79 | 4.90 44 | 1.80 70 | 2.94 56 | 3.77 88 | 3.28 69 | 2.66 70 | 2.85 66 | 2.71 50 | 2.63 19 | 5.39 25 | 2.37 20 | 1.81 74 | 4.12 76 | 1.72 31 | 2.45 35 | 3.84 47 | 1.46 44 |
Local-TV-L1 [65] | 62.0 | 2.00 107 | 2.99 73 | 1.89 114 | 3.48 105 | 4.69 98 | 2.35 111 | 2.72 43 | 4.41 18 | 1.73 36 | 3.04 79 | 3.43 48 | 3.45 110 | 2.60 29 | 2.76 36 | 2.73 61 | 2.86 52 | 5.94 45 | 2.61 111 | 1.75 33 | 3.40 37 | 1.72 31 | 2.34 12 | 3.54 10 | 1.49 90 |
TC-Flow [46] | 62.6 | 1.73 30 | 2.83 39 | 1.65 48 | 2.62 52 | 4.09 60 | 1.71 55 | 2.83 61 | 4.57 30 | 1.73 36 | 3.06 84 | 3.79 89 | 3.34 98 | 2.68 88 | 2.92 99 | 2.79 105 | 2.94 57 | 6.05 49 | 2.52 83 | 1.75 33 | 3.33 29 | 1.72 31 | 2.66 95 | 4.42 107 | 1.46 44 |
TriFlow [95] | 64.4 | 1.84 83 | 3.26 107 | 1.70 74 | 3.06 80 | 4.38 76 | 2.11 101 | 2.77 54 | 5.38 70 | 1.73 36 | 2.98 64 | 3.66 81 | 3.20 19 | 2.73 110 | 3.01 111 | 2.77 92 | 2.85 50 | 5.56 29 | 2.39 33 | 1.79 60 | 3.56 51 | 1.72 31 | 2.55 59 | 3.86 51 | 1.45 24 |
Sparse Occlusion [54] | 65.4 | 1.77 54 | 2.92 56 | 1.67 60 | 2.94 73 | 4.54 86 | 1.68 49 | 2.68 31 | 4.50 25 | 1.79 59 | 2.90 39 | 3.45 50 | 3.25 53 | 2.68 88 | 2.90 90 | 2.75 80 | 3.43 111 | 7.64 111 | 2.52 83 | 1.80 67 | 4.18 79 | 1.68 1 | 2.59 76 | 4.06 82 | 1.47 66 |
Modified CLG [34] | 65.5 | 1.76 45 | 2.79 32 | 1.72 80 | 3.57 115 | 4.75 103 | 2.43 116 | 3.15 90 | 7.06 110 | 1.95 93 | 3.02 75 | 3.67 82 | 3.25 53 | 2.61 35 | 2.75 33 | 2.68 32 | 3.05 69 | 6.71 74 | 2.48 73 | 1.76 40 | 3.42 41 | 1.72 31 | 2.52 48 | 3.79 35 | 1.47 66 |
EpicFlow [102] | 67.7 | 1.74 39 | 2.98 71 | 1.63 26 | 2.58 48 | 4.07 58 | 1.67 46 | 2.83 61 | 5.04 51 | 1.72 34 | 2.99 66 | 3.82 90 | 3.32 91 | 2.65 59 | 2.85 66 | 2.75 80 | 3.01 64 | 6.56 62 | 2.47 66 | 1.87 90 | 5.16 107 | 1.74 76 | 2.83 113 | 4.62 117 | 1.46 44 |
Classic++ [32] | 67.8 | 1.82 79 | 2.90 51 | 1.72 80 | 2.98 77 | 4.50 83 | 1.74 63 | 2.92 77 | 4.95 47 | 1.79 59 | 3.07 85 | 3.64 75 | 3.27 67 | 2.66 70 | 2.85 66 | 2.68 32 | 3.07 74 | 6.57 65 | 2.56 100 | 1.79 60 | 4.26 81 | 1.72 31 | 2.56 63 | 3.93 56 | 1.48 85 |
FlowNetS+ft+v [112] | 67.9 | 1.89 94 | 3.07 88 | 1.81 106 | 3.48 105 | 4.84 109 | 2.28 108 | 2.80 57 | 4.73 35 | 1.80 70 | 2.93 53 | 3.40 45 | 3.29 78 | 2.67 84 | 2.88 82 | 2.83 114 | 2.67 27 | 5.39 25 | 2.40 37 | 1.85 87 | 4.09 75 | 1.72 31 | 2.49 41 | 3.79 35 | 1.46 44 |
F-TV-L1 [15] | 68.3 | 1.95 103 | 3.21 102 | 1.85 112 | 3.34 96 | 4.63 93 | 1.97 90 | 3.02 88 | 5.30 66 | 2.03 97 | 3.01 73 | 3.56 62 | 3.29 78 | 2.64 51 | 2.84 57 | 2.66 20 | 2.68 29 | 5.36 24 | 2.42 46 | 1.82 77 | 4.37 89 | 1.75 80 | 2.35 14 | 3.51 7 | 1.48 85 |
RFlow [90] | 68.5 | 1.77 54 | 2.94 61 | 1.68 63 | 3.27 93 | 4.67 94 | 1.73 60 | 2.88 74 | 7.06 110 | 1.79 59 | 3.04 79 | 3.86 92 | 3.25 53 | 2.63 44 | 2.82 52 | 2.68 32 | 2.84 49 | 6.79 78 | 2.37 20 | 1.80 67 | 4.20 80 | 1.73 63 | 2.66 95 | 4.06 82 | 1.49 90 |
HBM-GC [105] | 69.5 | 1.81 75 | 2.84 42 | 1.74 92 | 2.65 56 | 4.20 63 | 1.70 52 | 2.48 3 | 3.66 2 | 1.68 13 | 3.02 75 | 3.63 73 | 3.32 91 | 2.72 109 | 2.97 108 | 2.86 119 | 4.21 128 | 9.74 128 | 2.66 117 | 1.76 40 | 3.30 27 | 1.74 76 | 2.43 30 | 3.84 47 | 1.50 103 |
OFH [38] | 72.5 | 1.81 75 | 2.98 71 | 1.68 63 | 2.92 71 | 4.30 70 | 1.72 58 | 2.96 83 | 5.62 80 | 1.78 56 | 2.88 34 | 3.36 44 | 3.18 13 | 2.66 70 | 2.88 82 | 2.73 61 | 3.01 64 | 6.44 59 | 2.52 83 | 2.05 112 | 5.99 112 | 1.76 87 | 2.85 114 | 4.49 111 | 1.47 66 |
IAOF [50] | 72.6 | 2.05 112 | 3.24 106 | 1.83 109 | 4.43 129 | 5.50 131 | 2.51 121 | 3.29 96 | 6.36 102 | 1.85 77 | 3.24 98 | 3.48 53 | 3.35 102 | 2.63 44 | 2.82 52 | 2.65 12 | 2.85 50 | 6.36 55 | 2.37 20 | 1.75 33 | 3.74 61 | 1.71 15 | 2.52 48 | 3.85 50 | 1.47 66 |
BlockOverlap [61] | 73.0 | 1.98 105 | 3.04 81 | 1.91 117 | 3.37 98 | 4.71 100 | 2.37 112 | 2.79 55 | 4.47 22 | 1.90 83 | 3.17 93 | 3.51 57 | 3.73 119 | 2.63 44 | 2.74 32 | 2.77 92 | 2.88 53 | 6.59 67 | 2.57 103 | 1.77 51 | 3.46 44 | 1.79 99 | 2.29 4 | 3.45 5 | 1.53 115 |
ROF-ND [107] | 74.1 | 1.77 54 | 2.69 18 | 1.63 26 | 2.88 68 | 4.46 81 | 1.70 52 | 2.67 27 | 5.18 56 | 1.73 36 | 3.35 107 | 4.80 117 | 3.33 95 | 2.62 40 | 2.79 45 | 2.75 80 | 3.44 112 | 7.78 117 | 2.52 83 | 1.91 97 | 3.85 65 | 1.80 101 | 3.00 118 | 4.80 118 | 1.47 66 |
Black & Anandan [4] | 74.6 | 2.01 110 | 3.03 80 | 1.86 113 | 3.86 122 | 5.04 123 | 2.25 106 | 4.13 113 | 6.30 99 | 2.49 111 | 3.26 101 | 3.86 92 | 3.24 48 | 2.66 70 | 2.84 57 | 2.72 54 | 2.72 32 | 6.16 51 | 2.35 14 | 1.82 77 | 3.56 51 | 1.72 31 | 2.49 41 | 3.71 28 | 1.47 66 |
AdaConv-v1 [126] | 75.0 | 2.22 121 | 4.18 127 | 2.00 120 | 3.13 84 | 3.92 47 | 2.63 125 | 4.33 116 | 5.53 75 | 3.46 126 | 4.12 123 | 4.92 118 | 4.19 126 | 2.53 9 | 2.66 14 | 2.59 1 | 2.49 7 | 4.58 9 | 2.30 2 | 1.93 100 | 4.96 105 | 1.93 121 | 2.25 3 | 3.42 3 | 1.55 118 |
Fusion [6] | 75.2 | 1.78 62 | 3.23 105 | 1.63 26 | 2.54 44 | 3.93 48 | 1.68 49 | 2.75 50 | 4.79 39 | 1.79 59 | 3.15 92 | 4.03 99 | 3.23 39 | 2.67 84 | 2.93 103 | 2.69 44 | 3.57 116 | 7.84 119 | 2.55 96 | 1.87 90 | 4.30 86 | 1.73 63 | 2.71 101 | 4.32 105 | 1.48 85 |
Ad-TV-NDC [36] | 75.6 | 2.31 124 | 3.15 98 | 2.22 125 | 3.85 121 | 4.86 111 | 2.52 122 | 2.87 72 | 5.55 76 | 1.85 77 | 3.25 99 | 3.57 64 | 3.41 107 | 2.68 88 | 2.86 74 | 2.74 70 | 2.70 30 | 5.61 32 | 2.47 66 | 1.77 51 | 3.34 32 | 1.72 31 | 2.39 25 | 3.58 16 | 1.50 103 |
ACK-Prior [27] | 75.9 | 1.73 30 | 2.93 59 | 1.61 16 | 2.46 41 | 3.89 45 | 1.66 43 | 4.37 117 | 4.91 45 | 2.75 116 | 3.05 82 | 3.64 75 | 3.34 98 | 2.68 88 | 2.86 74 | 2.79 105 | 3.21 90 | 6.38 57 | 2.57 103 | 1.87 90 | 3.71 59 | 1.81 104 | 2.66 95 | 4.03 74 | 1.54 116 |
CRTflow [80] | 77.4 | 1.89 94 | 3.09 91 | 1.76 93 | 3.26 92 | 4.84 109 | 1.84 77 | 3.01 85 | 5.81 89 | 2.00 94 | 2.99 66 | 3.45 50 | 3.31 85 | 2.68 88 | 2.91 98 | 2.79 105 | 2.63 19 | 5.25 21 | 2.41 39 | 1.80 67 | 4.29 84 | 1.74 76 | 2.57 66 | 4.05 80 | 1.49 90 |
Occlusion-TV-L1 [63] | 78.1 | 1.80 67 | 2.91 53 | 1.73 90 | 3.41 100 | 5.03 120 | 1.82 75 | 2.83 61 | 5.78 87 | 1.85 77 | 3.23 95 | 4.57 113 | 3.32 91 | 2.59 28 | 2.73 28 | 2.67 23 | 3.14 81 | 7.74 116 | 2.56 100 | 1.91 97 | 3.41 39 | 1.84 107 | 2.59 76 | 4.07 85 | 1.47 66 |
Adaptive [20] | 78.4 | 1.87 90 | 3.11 93 | 1.76 93 | 3.51 112 | 5.03 120 | 1.90 85 | 2.94 79 | 5.14 53 | 1.88 80 | 3.00 70 | 3.65 80 | 3.31 85 | 2.67 84 | 2.88 82 | 2.67 23 | 3.07 74 | 7.11 91 | 2.47 66 | 1.83 82 | 4.28 83 | 1.71 15 | 2.62 85 | 3.99 66 | 1.49 90 |
Steered-L1 [118] | 78.8 | 1.73 30 | 3.05 84 | 1.63 26 | 2.60 50 | 4.05 55 | 1.76 70 | 4.00 110 | 5.96 95 | 2.27 107 | 3.36 108 | 4.13 104 | 3.60 114 | 2.68 88 | 2.90 90 | 2.67 23 | 3.02 67 | 6.70 73 | 2.54 90 | 1.86 88 | 4.13 78 | 1.79 99 | 2.57 66 | 4.08 87 | 1.49 90 |
Filter Flow [19] | 79.2 | 1.93 99 | 3.01 76 | 1.81 106 | 3.49 109 | 4.76 105 | 2.37 112 | 2.93 78 | 5.33 68 | 1.90 83 | 3.25 99 | 3.62 72 | 3.41 107 | 2.62 40 | 2.73 28 | 2.78 99 | 2.83 48 | 6.08 50 | 2.43 51 | 1.84 84 | 3.90 70 | 1.76 87 | 2.57 66 | 3.83 43 | 1.56 120 |
Correlation Flow [75] | 80.8 | 1.73 30 | 2.89 49 | 1.60 7 | 3.23 90 | 4.83 108 | 1.70 52 | 2.67 27 | 4.47 22 | 1.73 36 | 2.95 58 | 3.50 54 | 3.29 78 | 2.73 110 | 2.92 99 | 2.83 114 | 4.11 124 | 9.07 124 | 2.60 109 | 2.06 113 | 6.41 114 | 1.87 114 | 2.79 110 | 4.33 106 | 1.49 90 |
GraphCuts [14] | 80.9 | 2.05 112 | 3.55 119 | 1.76 93 | 2.71 60 | 3.98 51 | 2.05 97 | 5.94 124 | 4.70 33 | 2.75 116 | 3.23 95 | 3.92 96 | 3.33 95 | 2.66 70 | 2.85 66 | 2.62 4 | 2.89 54 | 6.59 67 | 2.33 7 | 1.91 97 | 4.44 91 | 1.81 104 | 2.65 90 | 4.10 90 | 1.51 111 |
Complementary OF [21] | 81.5 | 1.75 41 | 3.20 101 | 1.61 16 | 2.55 45 | 4.05 55 | 1.66 43 | 5.55 121 | 7.07 112 | 3.02 119 | 2.95 58 | 3.64 75 | 3.25 53 | 2.69 99 | 2.94 106 | 2.75 80 | 3.01 64 | 6.76 76 | 2.48 73 | 2.04 111 | 5.69 110 | 1.75 80 | 3.37 124 | 5.64 128 | 1.47 66 |
BriefMatch [124] | 81.6 | 1.84 83 | 2.99 73 | 1.72 80 | 2.70 59 | 4.09 60 | 2.07 99 | 3.54 102 | 4.55 29 | 2.45 109 | 3.65 118 | 4.05 100 | 4.07 124 | 2.62 40 | 2.75 33 | 2.81 113 | 3.35 100 | 6.69 70 | 2.83 125 | 1.79 60 | 3.75 62 | 1.77 93 | 2.58 73 | 3.97 63 | 1.49 90 |
IAOF2 [51] | 82.7 | 2.00 107 | 3.27 108 | 1.78 100 | 3.48 105 | 5.01 117 | 2.11 101 | 2.75 50 | 5.81 89 | 1.78 56 | 3.10 90 | 3.74 87 | 3.30 82 | 2.88 121 | 3.33 125 | 2.72 54 | 3.39 106 | 7.64 111 | 2.45 57 | 1.76 40 | 3.52 49 | 1.70 7 | 2.61 82 | 4.03 74 | 1.47 66 |
HBpMotionGpu [43] | 83.2 | 2.13 116 | 3.48 118 | 1.96 118 | 3.80 120 | 5.07 126 | 2.47 117 | 2.71 39 | 5.28 65 | 1.73 36 | 3.26 101 | 4.60 115 | 3.31 85 | 2.65 59 | 2.86 74 | 2.77 92 | 3.14 81 | 7.55 107 | 2.51 81 | 1.72 11 | 3.06 10 | 1.71 15 | 2.69 100 | 4.18 98 | 1.52 113 |
2D-CLG [1] | 83.7 | 1.88 92 | 3.00 75 | 1.79 102 | 3.62 118 | 4.68 96 | 2.49 120 | 3.79 108 | 5.67 85 | 2.33 108 | 3.28 103 | 3.72 84 | 3.28 69 | 2.65 59 | 2.83 54 | 2.71 50 | 3.17 85 | 6.91 85 | 2.54 90 | 1.95 103 | 4.57 96 | 1.76 87 | 2.54 54 | 3.81 42 | 1.46 44 |
Nguyen [33] | 83.8 | 2.00 107 | 3.12 95 | 1.89 114 | 3.97 124 | 4.92 113 | 2.47 117 | 3.21 94 | 7.73 115 | 1.94 91 | 3.34 106 | 3.89 95 | 3.32 91 | 2.65 59 | 2.84 57 | 2.67 23 | 2.90 55 | 6.33 54 | 2.37 20 | 1.99 108 | 5.32 108 | 1.80 101 | 2.55 59 | 3.96 62 | 1.46 44 |
Aniso-Texture [82] | 85.6 | 1.72 22 | 2.87 47 | 1.63 26 | 3.41 100 | 5.01 117 | 1.89 82 | 3.70 105 | 5.00 48 | 1.84 75 | 3.40 110 | 5.27 119 | 3.64 116 | 2.74 112 | 3.03 113 | 2.79 105 | 4.28 130 | 9.61 127 | 2.77 124 | 1.75 33 | 3.60 54 | 1.70 7 | 2.76 108 | 4.43 109 | 1.47 66 |
CNN-flow-warp+ref [117] | 85.8 | 1.76 45 | 2.83 39 | 1.72 80 | 3.09 83 | 4.53 85 | 1.98 91 | 3.46 100 | 6.73 105 | 2.08 98 | 3.81 121 | 4.73 116 | 3.88 123 | 2.68 88 | 2.90 90 | 2.79 105 | 2.99 63 | 6.56 62 | 2.51 81 | 2.07 114 | 5.42 109 | 1.80 101 | 2.55 59 | 3.93 56 | 1.46 44 |
TriangleFlow [30] | 86.3 | 1.89 94 | 3.12 95 | 1.72 80 | 3.06 80 | 4.50 83 | 1.75 66 | 2.95 82 | 5.78 87 | 1.90 83 | 3.07 85 | 4.02 98 | 3.33 95 | 2.66 70 | 2.88 82 | 2.65 12 | 3.17 85 | 6.69 70 | 2.45 57 | 2.08 115 | 6.91 121 | 1.89 116 | 3.37 124 | 5.58 126 | 1.47 66 |
Horn & Schunck [3] | 87.3 | 1.95 103 | 3.08 89 | 1.78 100 | 3.94 123 | 4.99 115 | 2.37 112 | 4.00 110 | 6.86 108 | 2.68 114 | 3.53 112 | 4.32 108 | 3.28 69 | 2.71 108 | 2.90 90 | 2.73 61 | 2.75 39 | 5.82 41 | 2.37 20 | 1.93 100 | 3.96 71 | 1.77 93 | 2.59 76 | 3.83 43 | 1.49 90 |
Bartels [41] | 87.7 | 1.94 102 | 3.18 100 | 1.84 111 | 2.83 67 | 4.45 80 | 2.00 93 | 2.83 61 | 5.31 67 | 1.91 86 | 3.28 103 | 4.09 103 | 3.69 118 | 2.68 88 | 2.72 27 | 2.95 130 | 3.56 115 | 7.19 93 | 3.04 129 | 1.80 67 | 3.40 37 | 1.89 116 | 2.52 48 | 3.80 39 | 1.60 125 |
TI-DOFE [24] | 88.4 | 2.24 123 | 3.17 99 | 2.11 124 | 4.18 128 | 5.05 124 | 2.74 127 | 3.54 102 | 6.74 107 | 2.24 105 | 3.73 119 | 4.23 106 | 3.39 105 | 2.66 70 | 2.87 78 | 2.73 61 | 2.67 27 | 5.90 44 | 2.32 4 | 1.87 90 | 3.85 65 | 1.78 96 | 2.62 85 | 3.72 29 | 1.50 103 |
NL-TV-NCC [25] | 88.5 | 1.84 83 | 3.01 76 | 1.65 48 | 2.94 73 | 4.56 90 | 1.72 58 | 2.94 79 | 5.90 92 | 1.93 90 | 3.13 91 | 4.05 100 | 3.37 103 | 2.70 105 | 2.76 36 | 2.93 128 | 3.31 96 | 7.44 102 | 2.54 90 | 1.97 106 | 4.81 101 | 1.86 113 | 2.65 90 | 3.92 55 | 1.56 120 |
LocallyOriented [52] | 88.9 | 1.89 94 | 3.06 86 | 1.77 99 | 3.48 105 | 4.81 107 | 2.00 93 | 3.15 90 | 5.90 92 | 1.84 75 | 3.23 95 | 4.57 113 | 3.31 85 | 2.68 88 | 2.90 90 | 2.68 32 | 3.11 78 | 6.53 61 | 2.63 113 | 1.89 94 | 4.77 100 | 1.73 63 | 2.67 98 | 4.13 93 | 1.49 90 |
TV-L1-improved [17] | 89.1 | 1.83 82 | 3.02 79 | 1.72 80 | 3.53 114 | 4.99 115 | 1.96 89 | 3.70 105 | 5.14 53 | 2.20 104 | 3.00 70 | 3.58 67 | 3.27 67 | 2.69 99 | 2.92 99 | 2.68 32 | 3.23 91 | 7.48 103 | 2.45 57 | 2.13 117 | 6.96 122 | 1.85 110 | 2.65 90 | 4.10 90 | 1.50 103 |
SimpleFlow [49] | 89.5 | 1.77 54 | 2.96 65 | 1.66 55 | 2.95 75 | 4.32 72 | 1.71 55 | 5.71 122 | 9.23 121 | 2.71 115 | 2.91 43 | 3.51 57 | 3.28 69 | 2.68 88 | 2.90 90 | 2.74 70 | 3.87 120 | 8.55 121 | 2.58 105 | 2.57 127 | 11.2 129 | 2.13 128 | 3.16 122 | 5.29 122 | 1.45 24 |
HCIC-L [99] | 92.7 | 2.83 130 | 3.95 125 | 2.76 130 | 3.06 80 | 3.97 50 | 2.47 117 | 3.44 99 | 6.45 103 | 2.11 99 | 3.30 105 | 4.19 105 | 3.34 98 | 2.61 35 | 2.68 21 | 2.74 70 | 3.07 74 | 7.18 92 | 2.48 73 | 1.90 95 | 3.84 64 | 1.84 107 | 3.15 120 | 4.61 116 | 1.54 116 |
Shiralkar [42] | 94.4 | 1.87 90 | 3.21 102 | 1.68 63 | 3.43 104 | 4.75 103 | 1.77 71 | 3.72 107 | 7.09 113 | 2.13 101 | 3.76 120 | 5.83 122 | 3.29 78 | 2.69 99 | 2.97 108 | 2.65 12 | 3.39 106 | 7.20 94 | 2.55 96 | 2.23 119 | 6.65 118 | 1.78 96 | 3.03 119 | 4.95 121 | 1.44 4 |
StereoFlow [44] | 98.3 | 2.57 127 | 4.24 129 | 2.04 122 | 3.60 117 | 4.80 106 | 2.29 110 | 2.96 83 | 6.64 104 | 1.80 70 | 3.07 85 | 3.85 91 | 3.26 64 | 3.86 129 | 4.99 129 | 2.79 105 | 4.10 123 | 9.86 129 | 2.59 106 | 1.74 25 | 3.61 55 | 1.72 31 | 2.90 115 | 4.55 115 | 1.49 90 |
Rannacher [23] | 98.4 | 1.85 89 | 3.08 89 | 1.76 93 | 3.59 116 | 5.09 128 | 1.89 82 | 3.88 109 | 5.70 86 | 2.51 112 | 3.05 82 | 3.95 97 | 3.31 85 | 2.70 105 | 2.94 106 | 2.68 32 | 3.28 94 | 7.66 113 | 2.47 66 | 2.10 116 | 6.54 117 | 1.85 110 | 2.79 110 | 4.52 114 | 1.51 111 |
StereoOF-V1MT [119] | 100.3 | 1.90 98 | 3.40 115 | 1.67 60 | 3.17 85 | 4.55 87 | 1.78 73 | 4.31 115 | 6.26 97 | 2.51 112 | 4.40 124 | 6.00 126 | 3.74 120 | 2.87 120 | 3.26 122 | 2.83 114 | 3.67 118 | 6.94 86 | 2.88 127 | 2.16 118 | 6.46 116 | 1.96 122 | 2.61 82 | 3.99 66 | 1.44 4 |
SegOF [10] | 101.2 | 1.84 83 | 3.34 112 | 1.73 90 | 3.03 78 | 4.32 72 | 1.98 91 | 5.51 120 | 8.76 120 | 2.92 118 | 3.48 111 | 9.65 129 | 3.25 53 | 2.70 105 | 2.97 108 | 2.75 80 | 3.50 113 | 7.00 89 | 2.67 120 | 2.38 125 | 8.43 127 | 2.03 125 | 2.94 116 | 4.85 119 | 1.45 24 |
UnFlow [129] | 103.2 | 2.03 111 | 3.75 123 | 1.79 102 | 3.22 89 | 4.36 74 | 2.03 95 | 3.49 101 | 8.25 119 | 2.14 103 | 3.09 88 | 4.25 107 | 3.22 31 | 2.88 121 | 3.29 124 | 2.77 92 | 4.11 124 | 9.53 126 | 2.63 113 | 1.95 103 | 4.99 106 | 1.76 87 | 3.52 128 | 5.53 125 | 1.48 85 |
Dynamic MRF [7] | 107.2 | 1.80 67 | 3.33 111 | 1.65 48 | 3.04 79 | 4.68 96 | 1.75 66 | 4.08 112 | 7.74 116 | 2.45 109 | 4.01 122 | 6.17 128 | 3.86 121 | 2.80 114 | 3.18 118 | 2.78 99 | 4.14 126 | 8.87 122 | 2.74 123 | 2.29 122 | 7.41 124 | 1.90 120 | 2.94 116 | 4.49 111 | 1.50 103 |
Learning Flow [11] | 107.2 | 1.93 99 | 3.13 97 | 1.76 93 | 3.41 100 | 4.86 111 | 1.94 88 | 6.37 129 | 12.1 130 | 3.16 120 | 3.54 113 | 4.38 111 | 3.49 112 | 2.90 123 | 3.21 120 | 2.90 125 | 3.20 89 | 6.69 70 | 2.54 90 | 2.01 109 | 4.70 99 | 1.85 110 | 2.81 112 | 4.19 99 | 1.57 124 |
SPSA-learn [13] | 107.3 | 1.98 105 | 3.46 116 | 1.80 104 | 3.51 112 | 4.74 102 | 2.28 108 | 5.42 119 | 10.3 126 | 3.29 122 | 3.58 115 | 4.53 112 | 3.30 82 | 2.82 117 | 3.18 118 | 2.73 61 | 3.33 97 | 7.42 100 | 2.46 63 | 3.48 131 | 13.3 130 | 3.86 131 | 4.29 130 | 6.74 130 | 1.46 44 |
2bit-BM-tele [98] | 108.8 | 2.07 114 | 3.34 112 | 1.96 118 | 3.42 103 | 4.98 114 | 2.14 104 | 3.02 88 | 5.96 95 | 2.13 101 | 3.38 109 | 4.35 109 | 3.68 117 | 2.81 115 | 3.01 111 | 2.92 127 | 4.27 129 | 10.4 130 | 2.89 128 | 3.17 130 | 13.6 131 | 2.63 130 | 2.44 33 | 3.79 35 | 1.61 127 |
Adaptive flow [45] | 110.3 | 2.51 125 | 3.57 120 | 2.34 126 | 4.10 126 | 5.07 126 | 2.88 128 | 3.20 93 | 6.30 99 | 2.25 106 | 3.54 113 | 4.08 102 | 3.63 115 | 2.85 119 | 3.16 117 | 2.78 99 | 4.05 121 | 8.92 123 | 2.60 109 | 1.84 84 | 3.82 63 | 1.84 107 | 2.75 107 | 4.23 102 | 1.55 118 |
SILK [79] | 112.9 | 2.14 117 | 3.28 109 | 2.00 120 | 4.08 125 | 5.01 117 | 2.63 125 | 6.23 126 | 9.75 123 | 3.32 124 | 3.62 116 | 4.37 110 | 3.48 111 | 2.81 115 | 3.14 116 | 2.78 99 | 3.33 97 | 7.78 117 | 2.85 126 | 1.93 100 | 4.86 103 | 1.87 114 | 2.72 103 | 4.13 93 | 1.50 103 |
GroupFlow [9] | 112.9 | 2.23 122 | 4.26 130 | 1.89 114 | 3.20 88 | 4.49 82 | 2.21 105 | 6.25 127 | 10.6 128 | 4.16 129 | 3.63 117 | 5.92 123 | 3.56 113 | 3.12 128 | 3.87 128 | 2.86 119 | 4.20 127 | 9.27 125 | 2.66 117 | 2.28 121 | 7.36 123 | 1.76 87 | 3.43 126 | 5.62 127 | 1.44 4 |
Heeger++ [104] | 113.3 | 2.21 120 | 4.19 128 | 1.76 93 | 3.25 91 | 4.37 75 | 2.05 97 | 6.31 128 | 7.77 117 | 3.58 127 | 4.66 126 | 5.96 124 | 4.22 127 | 2.97 126 | 3.40 126 | 2.90 125 | 4.06 122 | 7.41 99 | 3.07 130 | 2.54 126 | 7.66 126 | 1.89 116 | 3.36 123 | 5.43 124 | 1.45 24 |
FOLKI [16] | 113.7 | 2.67 128 | 3.60 121 | 2.67 129 | 4.14 127 | 5.06 125 | 2.92 129 | 4.29 114 | 9.31 122 | 3.16 120 | 4.76 129 | 5.27 119 | 4.33 129 | 2.91 124 | 3.25 121 | 2.87 122 | 2.97 60 | 5.89 43 | 2.61 111 | 2.23 119 | 5.69 110 | 2.06 127 | 2.74 106 | 3.99 66 | 1.61 127 |
PGAM+LK [55] | 115.7 | 2.54 126 | 3.85 124 | 2.37 127 | 3.49 109 | 4.59 91 | 2.58 123 | 5.38 118 | 10.4 127 | 3.34 125 | 4.52 125 | 5.29 121 | 4.11 125 | 2.84 118 | 3.09 114 | 2.87 122 | 3.66 117 | 7.21 95 | 2.73 122 | 1.96 105 | 4.38 90 | 1.89 116 | 2.78 109 | 4.19 99 | 1.63 129 |
FFV1MT [106] | 117.8 | 2.16 118 | 4.10 126 | 1.82 108 | 3.40 99 | 4.42 79 | 2.41 115 | 5.77 123 | 9.75 123 | 3.31 123 | 4.66 126 | 5.96 124 | 4.22 127 | 2.93 125 | 3.26 122 | 2.94 129 | 3.42 109 | 6.83 81 | 2.65 116 | 2.66 128 | 7.57 125 | 2.00 123 | 3.46 127 | 5.41 123 | 1.63 129 |
SLK [47] | 118.7 | 2.17 119 | 3.38 114 | 2.06 123 | 3.74 119 | 4.55 87 | 2.59 124 | 6.05 125 | 8.14 118 | 3.70 128 | 4.66 126 | 6.07 127 | 3.86 121 | 3.09 127 | 3.67 127 | 2.83 114 | 3.42 109 | 7.28 97 | 2.66 117 | 2.35 123 | 6.75 119 | 2.03 125 | 3.15 120 | 4.91 120 | 1.56 120 |
Pyramid LK [2] | 122.1 | 2.69 129 | 3.63 122 | 2.65 128 | 4.52 130 | 5.19 129 | 3.28 130 | 9.47 130 | 7.52 114 | 6.37 130 | 10.2 131 | 17.4 130 | 10.8 131 | 4.45 131 | 6.09 131 | 2.86 119 | 3.13 80 | 6.71 74 | 2.59 106 | 2.36 124 | 6.88 120 | 2.02 124 | 4.64 131 | 7.19 131 | 1.60 125 |
Periodicity [78] | 130.3 | 3.05 131 | 6.22 131 | 2.83 131 | 5.00 131 | 5.35 130 | 3.62 131 | 11.4 131 | 14.2 131 | 10.8 131 | 9.29 130 | 17.8 131 | 6.38 130 | 4.38 130 | 5.73 130 | 3.29 131 | 4.39 131 | 10.6 131 | 3.11 131 | 2.85 129 | 11.0 128 | 2.34 129 | 4.04 129 | 6.02 129 | 2.26 131 |
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. |