Optical flow evaluation results |
Statistics:
Average
SD
R0.5
R1.0
R2.0
A90
A95
A99
Error type: endpoint angle interpolation normalized interpolation |
SD 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 | |
SepConv-v1 [127] | 24.6 | 2.15 88 | 3.44 90 | 0.69 1 | 1.70 29 | 2.16 25 | 1.17 27 | 8.00 65 | 3.78 2 | 7.22 42 | 2.71 57 | 2.44 3 | 2.23 39 | 1.63 7 | 1.81 8 | 1.22 1 | 1.80 1 | 2.53 1 | 1.09 1 | 3.88 10 | 6.01 10 | 1.90 12 | 5.43 35 | 6.86 36 | 0.76 1 |
NN-field [71] | 34.7 | 1.97 58 | 3.14 65 | 0.84 21 | 1.34 2 | 1.70 2 | 0.97 4 | 6.72 42 | 10.3 79 | 8.20 54 | 2.55 46 | 3.41 66 | 2.65 72 | 1.76 10 | 1.96 10 | 1.22 1 | 3.40 41 | 4.85 44 | 1.79 20 | 4.57 22 | 7.07 22 | 3.12 39 | 5.05 27 | 6.38 28 | 1.21 58 |
PMMST [114] | 35.3 | 1.96 57 | 3.12 58 | 0.81 5 | 1.60 19 | 2.09 19 | 1.03 16 | 6.35 36 | 9.32 68 | 7.35 43 | 2.93 74 | 3.99 99 | 2.65 72 | 1.64 9 | 1.82 9 | 1.23 4 | 3.08 23 | 4.39 23 | 1.53 6 | 4.97 30 | 7.68 30 | 2.63 21 | 5.04 25 | 6.37 27 | 1.34 74 |
SuperFlow [81] | 35.5 | 1.88 50 | 2.97 50 | 0.89 53 | 1.91 57 | 2.39 51 | 1.45 58 | 4.52 9 | 5.09 10 | 4.02 13 | 3.56 104 | 2.61 13 | 2.27 42 | 1.46 2 | 1.59 2 | 1.28 24 | 3.30 33 | 4.70 33 | 2.24 53 | 4.55 21 | 7.03 21 | 2.84 30 | 7.19 60 | 9.09 61 | 0.83 3 |
DeepFlow [86] | 36.1 | 2.03 77 | 3.23 78 | 0.83 11 | 1.95 61 | 2.44 61 | 1.58 75 | 4.26 6 | 5.34 16 | 2.24 3 | 2.77 64 | 2.57 10 | 1.98 26 | 3.30 37 | 3.75 37 | 1.27 19 | 2.92 18 | 4.16 18 | 1.69 14 | 3.34 4 | 5.16 4 | 1.58 6 | 8.21 78 | 10.4 79 | 1.27 65 |
ComplOF-FED-GPU [35] | 37.5 | 1.49 26 | 2.33 27 | 0.88 44 | 1.78 42 | 2.32 47 | 1.26 40 | 9.54 83 | 4.00 3 | 12.2 87 | 1.86 2 | 2.44 3 | 1.48 6 | 4.22 63 | 4.80 63 | 5.82 72 | 2.62 6 | 3.73 7 | 1.97 31 | 4.52 20 | 6.91 19 | 3.50 54 | 8.01 73 | 10.1 72 | 0.95 11 |
ALD-Flow [66] | 37.7 | 1.35 16 | 2.01 14 | 0.88 44 | 1.79 44 | 2.34 48 | 1.25 38 | 4.95 14 | 5.90 25 | 3.11 6 | 1.97 5 | 2.56 9 | 1.48 6 | 3.77 53 | 4.29 54 | 6.03 79 | 3.99 73 | 5.70 74 | 4.20 117 | 3.92 11 | 6.07 11 | 1.44 1 | 7.78 68 | 9.82 69 | 1.05 25 |
OAR-Flow [125] | 38.5 | 1.61 36 | 2.39 30 | 0.83 11 | 1.84 48 | 2.37 50 | 1.34 47 | 5.20 20 | 6.66 40 | 3.10 5 | 1.83 1 | 2.39 2 | 1.48 6 | 4.50 92 | 5.12 92 | 6.58 110 | 3.38 39 | 4.81 40 | 2.73 77 | 3.73 9 | 5.70 9 | 2.47 17 | 6.18 51 | 7.80 52 | 1.15 41 |
NNF-Local [87] | 40.6 | 1.08 4 | 1.64 7 | 0.79 2 | 1.30 1 | 1.64 1 | 0.98 6 | 8.95 77 | 13.9 98 | 11.2 79 | 2.52 40 | 3.37 65 | 2.49 56 | 1.45 1 | 1.58 1 | 1.22 1 | 4.26 92 | 6.08 93 | 2.22 51 | 5.38 41 | 8.33 41 | 5.57 108 | 4.99 22 | 6.31 23 | 1.26 64 |
IROF++ [58] | 41.7 | 1.07 2 | 1.58 2 | 0.81 5 | 1.67 26 | 2.16 25 | 1.08 22 | 6.49 39 | 7.78 52 | 6.90 40 | 1.93 3 | 2.55 8 | 2.04 28 | 4.40 76 | 5.00 76 | 6.59 117 | 3.77 61 | 5.38 62 | 1.84 23 | 3.98 12 | 6.16 12 | 2.66 24 | 10.5 122 | 13.3 123 | 1.15 41 |
TC-Flow [46] | 41.8 | 1.12 5 | 1.62 4 | 0.85 30 | 1.80 46 | 2.34 48 | 1.25 38 | 3.55 1 | 5.28 13 | 1.46 1 | 2.07 8 | 2.71 19 | 2.05 29 | 4.38 74 | 4.98 73 | 6.27 86 | 4.03 77 | 5.75 80 | 2.60 70 | 4.81 24 | 7.44 24 | 2.55 19 | 8.05 75 | 10.2 76 | 1.44 84 |
CLG-TV [48] | 43.2 | 1.83 48 | 2.89 49 | 0.97 83 | 2.20 89 | 2.74 97 | 1.65 84 | 5.14 19 | 6.68 41 | 5.81 32 | 2.23 16 | 2.69 16 | 2.60 67 | 4.17 62 | 4.74 62 | 1.25 10 | 2.65 9 | 3.77 9 | 1.47 4 | 3.24 3 | 5.01 3 | 1.66 7 | 9.56 107 | 12.1 107 | 0.96 13 |
Second-order prior [8] | 44.0 | 1.12 5 | 1.62 4 | 0.99 88 | 2.11 78 | 2.55 73 | 1.57 72 | 5.29 22 | 8.01 55 | 5.87 33 | 2.15 11 | 2.65 14 | 1.71 18 | 4.23 65 | 4.81 65 | 1.25 10 | 2.88 16 | 4.09 16 | 1.81 21 | 5.02 32 | 7.77 32 | 1.51 2 | 9.26 101 | 11.7 100 | 2.10 124 |
DeepFlow2 [108] | 44.4 | 2.32 100 | 3.69 102 | 0.87 39 | 1.90 53 | 2.43 57 | 1.30 43 | 4.25 3 | 5.73 21 | 2.18 2 | 3.00 79 | 3.42 69 | 1.92 23 | 3.44 40 | 3.91 41 | 2.60 50 | 2.24 3 | 3.17 3 | 1.48 5 | 3.41 5 | 5.27 6 | 2.92 34 | 8.45 83 | 10.7 84 | 2.03 120 |
FMOF [94] | 44.8 | 1.71 41 | 2.65 39 | 0.94 72 | 1.53 12 | 1.96 8 | 1.07 20 | 9.48 81 | 13.0 95 | 10.7 75 | 2.56 47 | 3.00 33 | 3.12 101 | 2.35 20 | 2.66 20 | 1.24 7 | 3.28 30 | 4.68 30 | 1.59 9 | 4.90 27 | 7.58 27 | 2.28 16 | 10.4 120 | 13.2 120 | 1.06 26 |
TC/T-Flow [76] | 45.0 | 1.54 30 | 2.36 28 | 1.06 101 | 1.85 49 | 2.41 55 | 1.44 56 | 5.40 25 | 7.43 48 | 5.66 28 | 2.57 48 | 2.47 5 | 1.48 6 | 4.45 84 | 5.06 84 | 6.55 100 | 3.37 38 | 4.80 39 | 1.36 2 | 4.27 14 | 6.49 14 | 4.17 75 | 7.94 70 | 10.0 71 | 0.94 10 |
SIOF [67] | 45.4 | 1.33 13 | 2.02 15 | 0.92 64 | 2.30 101 | 2.84 109 | 1.72 91 | 6.84 46 | 9.18 66 | 6.21 34 | 2.57 48 | 3.17 48 | 2.81 80 | 1.61 6 | 1.78 6 | 1.27 19 | 3.79 62 | 5.40 63 | 1.54 7 | 4.49 19 | 6.94 20 | 3.42 49 | 5.85 46 | 7.39 47 | 1.09 30 |
MLDP_OF [89] | 45.9 | 1.43 21 | 2.24 24 | 0.93 69 | 1.96 65 | 2.48 65 | 1.38 52 | 5.03 18 | 6.04 26 | 3.03 4 | 2.80 66 | 3.36 63 | 2.91 86 | 3.46 42 | 3.94 42 | 4.42 64 | 2.77 12 | 3.94 12 | 2.43 65 | 4.88 25 | 7.54 25 | 4.19 76 | 5.83 45 | 7.37 46 | 1.47 88 |
MDP-Flow2 [68] | 46.2 | 1.94 55 | 3.09 57 | 0.79 2 | 1.62 20 | 2.13 22 | 0.95 3 | 9.47 80 | 14.8 101 | 11.9 84 | 2.53 43 | 3.42 69 | 3.00 95 | 1.63 7 | 1.80 7 | 1.95 44 | 4.25 91 | 6.07 92 | 1.81 21 | 5.50 45 | 8.52 45 | 2.71 25 | 5.02 23 | 6.34 24 | 1.19 53 |
CBF [12] | 46.2 | 1.34 14 | 2.09 17 | 0.88 44 | 2.14 83 | 2.64 81 | 1.69 87 | 5.42 27 | 7.67 51 | 5.70 29 | 2.25 17 | 2.57 10 | 2.56 61 | 1.49 3 | 1.63 3 | 1.32 36 | 2.58 5 | 3.67 5 | 1.85 24 | 7.15 94 | 11.1 95 | 4.00 70 | 8.03 74 | 10.1 72 | 1.76 108 |
LME [70] | 46.5 | 1.93 53 | 3.07 54 | 0.81 5 | 1.63 21 | 2.14 23 | 1.13 25 | 5.40 25 | 7.66 50 | 5.53 27 | 2.44 29 | 3.25 53 | 2.60 67 | 4.48 89 | 5.10 90 | 6.48 92 | 4.83 112 | 6.89 112 | 1.65 10 | 4.28 15 | 6.62 15 | 3.05 38 | 5.03 24 | 6.35 25 | 1.23 61 |
WLIF-Flow [93] | 47.0 | 1.03 1 | 1.50 1 | 0.82 8 | 1.73 34 | 2.21 32 | 1.17 27 | 8.68 72 | 12.4 91 | 10.2 69 | 2.85 69 | 3.53 75 | 3.31 106 | 2.45 21 | 2.77 21 | 3.18 55 | 3.83 66 | 5.46 67 | 3.13 91 | 5.25 38 | 8.12 38 | 2.79 28 | 5.06 29 | 6.40 31 | 1.22 59 |
p-harmonic [29] | 47.6 | 1.81 47 | 2.87 47 | 0.84 21 | 2.26 98 | 2.86 110 | 2.12 112 | 4.57 10 | 5.80 22 | 3.76 10 | 2.95 75 | 2.80 22 | 2.74 76 | 3.65 49 | 4.15 49 | 1.26 16 | 3.01 20 | 4.28 20 | 2.08 39 | 4.42 18 | 6.84 18 | 4.12 73 | 8.75 88 | 11.1 89 | 0.96 13 |
OFH [38] | 47.6 | 1.41 20 | 2.17 20 | 0.91 62 | 1.96 65 | 2.47 64 | 1.32 45 | 7.12 51 | 9.37 69 | 6.61 37 | 2.03 7 | 2.69 16 | 1.47 3 | 4.26 67 | 4.84 67 | 5.89 73 | 3.02 21 | 4.29 21 | 2.26 54 | 6.29 67 | 9.37 60 | 6.74 117 | 6.88 59 | 8.69 60 | 0.98 18 |
IROF-TV [53] | 48.4 | 1.93 53 | 3.06 53 | 0.94 72 | 1.75 36 | 2.24 37 | 1.20 30 | 4.66 11 | 6.05 27 | 3.72 9 | 2.13 10 | 2.85 23 | 1.53 11 | 4.41 78 | 5.02 78 | 6.59 117 | 3.61 50 | 5.14 52 | 2.31 57 | 6.73 76 | 10.4 78 | 3.37 47 | 6.77 57 | 8.56 58 | 1.15 41 |
OFLAF [77] | 48.5 | 1.97 58 | 3.13 60 | 0.83 11 | 1.37 6 | 1.78 6 | 1.01 12 | 5.30 23 | 6.60 39 | 4.28 14 | 2.27 18 | 3.06 38 | 1.56 12 | 4.60 101 | 5.23 101 | 6.56 102 | 3.41 43 | 4.85 44 | 2.63 72 | 6.96 86 | 10.8 88 | 3.83 65 | 5.52 38 | 6.96 38 | 1.47 88 |
Aniso. Huber-L1 [22] | 50.1 | 1.44 22 | 2.23 22 | 0.90 60 | 2.23 92 | 2.73 94 | 1.50 65 | 5.43 28 | 6.83 43 | 5.46 25 | 2.49 35 | 2.94 29 | 3.01 96 | 4.09 60 | 4.65 60 | 1.25 10 | 4.29 93 | 6.12 94 | 1.38 3 | 4.08 13 | 6.31 13 | 1.72 8 | 10.2 117 | 12.9 118 | 0.82 2 |
Modified CLG [34] | 50.1 | 1.53 29 | 2.41 33 | 0.86 34 | 2.31 105 | 2.77 100 | 2.09 110 | 8.71 73 | 5.58 18 | 11.3 80 | 2.49 35 | 2.77 20 | 2.86 82 | 2.49 24 | 2.82 24 | 1.25 10 | 3.46 45 | 4.92 46 | 2.05 34 | 3.05 1 | 4.70 1 | 1.76 9 | 11.1 129 | 14.1 129 | 1.11 32 |
PGM-C [120] | 50.6 | 1.60 35 | 2.51 36 | 0.83 11 | 1.56 17 | 2.04 15 | 0.99 8 | 6.27 35 | 6.36 32 | 5.76 30 | 2.67 51 | 3.46 71 | 2.74 76 | 4.44 82 | 5.05 82 | 6.56 102 | 2.53 4 | 3.58 4 | 1.87 25 | 9.06 117 | 14.0 116 | 4.55 88 | 7.97 71 | 10.1 72 | 1.12 35 |
MDP-Flow [26] | 51.3 | 1.07 2 | 1.61 3 | 0.85 30 | 1.64 23 | 2.16 25 | 1.07 20 | 8.65 71 | 5.50 17 | 11.0 77 | 2.77 64 | 3.46 71 | 2.60 67 | 4.48 89 | 5.10 90 | 6.56 102 | 4.21 88 | 6.00 89 | 3.35 98 | 6.09 61 | 9.42 63 | 3.48 53 | 3.01 6 | 3.79 6 | 0.97 16 |
FlowNet2 [122] | 51.5 | 2.26 97 | 3.38 88 | 0.95 77 | 1.76 39 | 2.22 33 | 1.21 31 | 7.48 61 | 11.5 89 | 8.91 61 | 2.43 27 | 3.27 56 | 2.58 63 | 2.80 29 | 3.18 29 | 2.97 52 | 3.28 30 | 4.68 30 | 2.06 36 | 5.65 46 | 8.74 46 | 4.55 88 | 3.42 13 | 4.31 13 | 1.70 103 |
CostFilter [40] | 52.3 | 1.26 11 | 1.96 12 | 0.89 53 | 1.47 7 | 1.93 7 | 0.94 1 | 12.8 110 | 18.8 119 | 15.4 112 | 2.20 15 | 2.95 30 | 1.51 10 | 3.13 32 | 3.56 33 | 4.90 66 | 3.74 58 | 5.32 57 | 1.69 14 | 7.46 101 | 11.5 101 | 5.27 107 | 6.10 50 | 7.71 51 | 1.65 99 |
PH-Flow [101] | 52.6 | 1.97 58 | 3.13 60 | 0.83 11 | 1.55 15 | 2.05 16 | 1.22 33 | 8.79 75 | 13.5 97 | 10.6 73 | 2.33 21 | 3.14 45 | 2.10 34 | 1.78 11 | 1.99 11 | 2.14 46 | 5.44 123 | 7.77 123 | 4.46 120 | 5.15 35 | 7.97 35 | 3.91 67 | 5.25 32 | 6.63 33 | 1.47 88 |
COFM [59] | 53.2 | 1.36 17 | 1.97 13 | 0.88 44 | 1.59 18 | 2.08 18 | 1.23 35 | 9.49 82 | 15.4 105 | 12.1 86 | 3.08 87 | 4.19 109 | 1.60 14 | 2.21 18 | 2.50 18 | 2.01 45 | 3.92 70 | 5.58 71 | 2.13 44 | 6.58 75 | 10.2 77 | 5.93 113 | 2.85 4 | 3.60 4 | 1.77 109 |
CombBMOF [113] | 53.9 | 2.14 87 | 3.23 78 | 1.81 126 | 1.72 31 | 2.27 41 | 1.00 9 | 6.89 49 | 10.3 79 | 7.99 52 | 3.11 90 | 4.11 104 | 3.41 110 | 2.47 22 | 2.80 23 | 1.50 43 | 3.58 49 | 5.11 51 | 2.05 34 | 5.17 36 | 8.00 36 | 3.64 57 | 3.18 12 | 4.01 12 | 1.24 62 |
TV-L1-MCT [64] | 53.9 | 1.97 58 | 3.12 58 | 0.87 39 | 1.95 61 | 2.39 51 | 1.54 69 | 7.42 60 | 11.1 85 | 8.52 57 | 2.46 32 | 3.14 45 | 2.51 58 | 4.86 114 | 5.52 114 | 6.08 81 | 3.29 32 | 4.69 32 | 2.17 47 | 5.14 34 | 7.95 34 | 2.65 23 | 6.33 52 | 8.00 53 | 0.88 5 |
NL-TV-NCC [25] | 54.4 | 1.44 22 | 2.20 21 | 0.99 88 | 2.05 74 | 2.61 78 | 1.54 69 | 5.02 16 | 6.98 46 | 4.96 20 | 3.01 81 | 4.07 101 | 2.41 50 | 2.62 26 | 2.96 26 | 3.97 60 | 4.71 108 | 6.72 108 | 3.40 100 | 3.62 8 | 5.58 8 | 2.25 15 | 8.25 80 | 10.4 79 | 1.01 21 |
AdaConv-v1 [126] | 54.4 | 2.21 93 | 3.48 94 | 1.50 123 | 2.62 123 | 2.27 41 | 3.48 129 | 7.18 52 | 5.05 8 | 8.08 53 | 3.52 103 | 2.96 31 | 4.59 119 | 2.79 28 | 3.17 28 | 1.25 10 | 2.08 2 | 2.95 2 | 1.58 8 | 6.09 61 | 9.43 64 | 3.03 37 | 5.26 33 | 6.64 34 | 1.08 29 |
TCOF [69] | 55.1 | 1.34 14 | 2.05 16 | 0.84 21 | 2.47 116 | 3.10 129 | 2.36 121 | 6.61 40 | 8.86 60 | 6.94 41 | 2.52 40 | 3.35 62 | 2.27 42 | 3.95 57 | 4.49 57 | 1.29 29 | 3.12 25 | 4.45 25 | 1.96 30 | 7.32 99 | 11.3 100 | 3.45 50 | 5.79 42 | 7.31 43 | 1.25 63 |
CRTflow [80] | 56.2 | 1.83 48 | 2.87 47 | 1.01 94 | 2.30 101 | 2.90 115 | 2.26 115 | 5.96 32 | 6.95 44 | 5.40 24 | 2.45 31 | 3.05 37 | 2.28 45 | 4.40 76 | 5.01 77 | 6.50 95 | 3.19 27 | 4.55 28 | 1.67 12 | 6.35 69 | 9.83 69 | 2.60 20 | 7.70 67 | 9.73 68 | 0.91 7 |
NNF-EAC [103] | 56.8 | 2.01 74 | 3.16 71 | 1.02 95 | 1.75 36 | 2.29 44 | 1.17 27 | 11.4 97 | 18.5 116 | 14.6 105 | 4.57 122 | 6.10 125 | 2.97 91 | 3.85 55 | 4.38 55 | 1.23 4 | 2.78 13 | 3.95 13 | 1.65 10 | 5.19 37 | 8.03 37 | 2.23 14 | 5.05 27 | 6.38 28 | 1.29 68 |
nLayers [57] | 57.1 | 1.97 58 | 3.14 65 | 0.84 21 | 1.53 12 | 1.99 13 | 1.13 25 | 15.8 122 | 22.3 124 | 18.5 122 | 2.70 54 | 3.66 82 | 1.96 25 | 4.43 80 | 5.04 80 | 6.34 90 | 3.84 67 | 5.48 68 | 2.21 50 | 5.26 39 | 8.14 39 | 1.78 10 | 2.78 1 | 3.51 1 | 2.09 123 |
CPM-Flow [116] | 57.8 | 1.80 45 | 2.86 46 | 0.83 11 | 1.51 11 | 1.98 12 | 1.02 15 | 5.28 21 | 5.59 19 | 4.93 19 | 2.52 40 | 3.26 55 | 2.60 67 | 4.43 80 | 5.04 80 | 6.56 102 | 4.06 80 | 5.79 82 | 3.55 104 | 6.83 81 | 10.6 83 | 4.21 77 | 8.75 88 | 11.1 89 | 1.42 81 |
Classic++ [32] | 57.9 | 1.39 18 | 2.14 18 | 0.88 44 | 2.12 80 | 2.71 91 | 1.76 96 | 4.40 8 | 5.20 11 | 3.44 8 | 3.04 82 | 3.10 39 | 2.94 88 | 3.49 43 | 3.97 43 | 2.88 51 | 4.14 84 | 5.90 85 | 2.33 58 | 6.85 82 | 10.6 83 | 3.78 63 | 8.54 87 | 10.8 87 | 1.15 41 |
PMF [73] | 57.9 | 1.69 39 | 2.67 41 | 0.83 11 | 1.53 12 | 2.00 14 | 0.97 4 | 13.5 111 | 20.0 120 | 16.5 118 | 2.88 71 | 3.78 91 | 2.97 91 | 2.16 17 | 2.44 17 | 1.27 19 | 3.76 60 | 5.36 61 | 1.70 16 | 7.19 97 | 11.1 95 | 4.74 99 | 5.29 34 | 6.68 35 | 1.95 117 |
FlowFields [110] | 58.7 | 2.27 98 | 3.62 98 | 1.09 104 | 1.50 8 | 1.97 9 | 0.94 1 | 12.4 108 | 17.3 110 | 14.9 107 | 2.49 35 | 3.36 63 | 1.65 15 | 3.73 52 | 4.24 52 | 6.03 79 | 3.39 40 | 4.83 42 | 1.90 28 | 7.03 92 | 10.9 91 | 1.87 11 | 6.86 58 | 8.67 59 | 1.18 49 |
Ad-TV-NDC [36] | 59.1 | 2.61 114 | 4.13 116 | 1.16 112 | 2.45 113 | 2.78 102 | 2.34 120 | 4.25 3 | 6.07 28 | 3.78 11 | 3.58 106 | 4.45 115 | 2.51 58 | 1.87 12 | 2.10 12 | 1.33 40 | 3.36 37 | 4.79 38 | 2.42 63 | 3.41 5 | 5.27 6 | 1.54 3 | 9.04 94 | 11.4 94 | 0.97 16 |
Bartels [41] | 59.6 | 2.18 89 | 3.46 92 | 1.03 97 | 2.07 76 | 2.71 91 | 1.72 91 | 5.71 31 | 6.16 29 | 5.76 30 | 2.80 66 | 3.33 59 | 2.60 67 | 1.58 4 | 1.72 4 | 1.38 41 | 4.88 114 | 6.96 114 | 2.47 66 | 4.66 23 | 7.21 23 | 3.81 64 | 7.43 64 | 9.39 65 | 1.10 31 |
2DHMM-SAS [92] | 60.3 | 1.13 7 | 1.63 6 | 0.89 53 | 2.18 86 | 2.69 88 | 1.88 100 | 7.19 53 | 9.66 70 | 7.61 47 | 2.67 51 | 3.53 75 | 2.25 41 | 4.96 118 | 5.64 118 | 6.29 87 | 2.83 15 | 4.03 15 | 1.78 19 | 4.96 29 | 7.68 30 | 3.98 69 | 9.87 112 | 12.5 112 | 1.17 47 |
HAST [109] | 60.4 | 1.54 30 | 2.41 33 | 0.83 11 | 1.50 8 | 1.97 9 | 1.01 12 | 14.2 115 | 20.2 121 | 16.8 119 | 2.16 13 | 2.90 26 | 1.46 2 | 4.04 59 | 4.60 59 | 1.26 16 | 4.22 89 | 6.02 90 | 3.98 111 | 9.18 118 | 14.2 118 | 4.53 87 | 5.99 48 | 7.57 49 | 1.75 106 |
F-TV-L1 [15] | 60.6 | 2.41 105 | 3.82 106 | 0.97 83 | 2.33 106 | 2.89 113 | 1.91 102 | 4.72 13 | 6.38 33 | 4.58 17 | 2.49 35 | 2.90 26 | 2.98 94 | 2.25 19 | 2.54 19 | 1.30 31 | 2.73 11 | 3.88 11 | 2.40 62 | 6.37 71 | 9.85 71 | 3.69 60 | 11.1 129 | 14.1 129 | 0.92 9 |
Sparse Occlusion [54] | 60.6 | 2.21 93 | 3.52 95 | 0.93 69 | 2.12 80 | 2.73 94 | 1.46 60 | 4.67 12 | 6.57 37 | 4.98 22 | 2.16 13 | 2.85 23 | 1.75 19 | 5.06 121 | 5.76 121 | 6.58 110 | 2.64 8 | 3.74 8 | 2.23 52 | 6.21 66 | 9.61 67 | 3.46 52 | 9.82 111 | 12.4 111 | 0.95 11 |
ACK-Prior [27] | 61.4 | 1.52 28 | 2.37 29 | 1.10 105 | 1.90 53 | 2.51 68 | 1.04 17 | 12.1 105 | 8.87 61 | 14.9 107 | 2.53 43 | 3.02 34 | 2.32 47 | 4.80 110 | 5.46 110 | 6.63 122 | 4.03 77 | 5.74 79 | 2.68 74 | 5.80 49 | 8.97 49 | 4.01 71 | 3.02 7 | 3.80 7 | 1.01 21 |
EPPM w/o HM [88] | 61.8 | 1.72 42 | 2.70 43 | 1.27 119 | 1.65 24 | 2.18 29 | 1.30 43 | 14.6 117 | 15.2 104 | 13.1 92 | 2.76 61 | 3.69 84 | 2.07 31 | 1.98 13 | 2.22 13 | 1.28 24 | 3.63 51 | 5.17 53 | 2.42 63 | 8.88 115 | 13.7 115 | 8.97 120 | 5.63 39 | 7.11 40 | 1.17 47 |
LSM [39] | 61.9 | 1.15 8 | 1.70 9 | 0.89 53 | 1.72 31 | 2.11 20 | 1.37 50 | 7.40 59 | 11.1 85 | 8.52 57 | 1.96 4 | 2.60 12 | 1.79 20 | 4.95 117 | 5.63 117 | 6.24 84 | 4.37 94 | 6.23 95 | 2.08 39 | 6.95 85 | 10.7 86 | 5.20 105 | 8.77 90 | 11.1 89 | 1.40 76 |
ComponentFusion [96] | 62.4 | 1.44 22 | 2.27 25 | 0.80 4 | 1.68 27 | 2.23 36 | 1.00 9 | 11.0 92 | 14.9 102 | 11.7 83 | 2.43 27 | 3.28 57 | 1.47 3 | 3.77 53 | 4.27 53 | 4.63 65 | 4.18 86 | 5.96 87 | 3.42 101 | 10.2 124 | 15.8 124 | 11.7 126 | 6.07 49 | 7.66 50 | 1.51 93 |
Complementary OF [21] | 62.5 | 2.67 119 | 4.25 120 | 0.82 8 | 1.89 51 | 2.49 67 | 1.47 63 | 14.5 116 | 11.4 88 | 15.7 114 | 2.33 21 | 3.13 43 | 1.56 12 | 4.37 73 | 4.98 73 | 6.31 88 | 3.19 27 | 4.54 27 | 2.26 54 | 6.11 63 | 9.40 62 | 5.87 110 | 3.12 10 | 3.94 10 | 1.43 82 |
CNN-flow-warp+ref [117] | 62.7 | 1.79 44 | 2.83 44 | 0.87 39 | 2.11 78 | 2.69 88 | 1.69 87 | 5.64 30 | 6.46 36 | 6.31 35 | 3.11 90 | 2.91 28 | 2.85 81 | 4.52 96 | 5.14 96 | 6.58 110 | 3.34 36 | 4.75 36 | 2.76 80 | 6.52 74 | 10.1 74 | 2.88 32 | 8.42 81 | 10.6 82 | 1.07 27 |
FlowNetS+ft+v [112] | 62.8 | 2.75 124 | 4.38 124 | 0.92 64 | 3.05 127 | 2.83 107 | 3.29 127 | 4.97 15 | 3.66 1 | 4.97 21 | 1.97 5 | 2.34 1 | 2.24 40 | 4.78 108 | 5.44 108 | 6.58 110 | 2.93 19 | 4.17 19 | 1.74 18 | 6.15 64 | 9.52 65 | 2.63 21 | 9.27 102 | 11.7 100 | 0.98 18 |
SVFilterOh [111] | 63.4 | 2.18 89 | 3.45 91 | 0.92 64 | 1.66 25 | 2.19 31 | 1.22 33 | 15.7 121 | 22.4 125 | 18.5 122 | 2.48 34 | 3.32 58 | 2.35 48 | 4.46 85 | 5.08 86 | 6.58 110 | 3.66 52 | 5.21 54 | 2.37 61 | 5.40 43 | 8.36 43 | 1.54 3 | 3.87 16 | 4.89 16 | 1.83 112 |
Kuang [131] | 63.5 | 2.13 86 | 3.39 89 | 1.00 91 | 1.69 28 | 2.22 33 | 1.06 19 | 10.6 89 | 10.1 75 | 9.13 62 | 2.72 58 | 3.55 77 | 2.47 55 | 3.92 56 | 4.46 56 | 5.91 75 | 3.05 22 | 4.34 22 | 2.12 43 | 8.68 113 | 13.4 111 | 6.73 116 | 7.57 65 | 9.57 66 | 0.98 18 |
LDOF [28] | 63.9 | 2.50 110 | 3.97 110 | 0.94 72 | 3.47 129 | 2.44 61 | 3.90 130 | 6.75 45 | 5.87 23 | 7.37 44 | 2.12 9 | 2.70 18 | 2.27 42 | 2.09 15 | 2.35 15 | 1.30 31 | 3.15 26 | 4.48 26 | 2.14 45 | 11.8 126 | 18.3 126 | 11.2 124 | 5.67 40 | 7.16 41 | 2.25 126 |
DPOF [18] | 64.3 | 2.79 125 | 4.46 125 | 1.92 128 | 1.36 5 | 1.77 5 | 1.01 12 | 10.5 88 | 7.45 49 | 13.5 93 | 3.60 109 | 4.88 120 | 2.92 87 | 3.40 39 | 3.86 39 | 1.23 4 | 3.10 24 | 4.42 24 | 2.06 36 | 5.73 48 | 8.87 48 | 2.76 26 | 10.2 117 | 12.8 117 | 1.40 76 |
S2F-IF [123] | 64.5 | 2.28 99 | 3.63 99 | 0.95 77 | 1.55 15 | 2.05 16 | 1.09 24 | 8.10 66 | 12.9 93 | 10.1 68 | 2.33 21 | 3.13 43 | 1.93 24 | 4.33 72 | 4.93 72 | 6.57 108 | 2.66 10 | 3.77 9 | 2.10 42 | 6.50 73 | 10.1 74 | 4.58 92 | 10.3 119 | 13.0 119 | 1.91 114 |
Horn & Schunck [3] | 65.4 | 2.06 79 | 3.26 81 | 0.89 53 | 2.63 124 | 2.90 115 | 2.77 125 | 11.3 96 | 6.80 42 | 13.6 94 | 3.18 94 | 3.23 52 | 2.95 89 | 3.59 47 | 4.08 47 | 1.30 31 | 2.91 17 | 4.14 17 | 1.70 16 | 4.91 28 | 7.59 28 | 2.79 28 | 10.1 116 | 12.7 115 | 1.12 35 |
EpicFlow [102] | 65.4 | 2.07 81 | 3.30 83 | 0.83 11 | 1.75 36 | 2.29 44 | 1.23 35 | 5.47 29 | 4.72 7 | 4.44 16 | 5.04 126 | 6.87 126 | 3.09 98 | 4.46 85 | 5.08 86 | 6.56 102 | 2.81 14 | 4.00 14 | 2.00 32 | 9.31 119 | 14.4 119 | 5.04 101 | 8.21 78 | 10.4 79 | 1.18 49 |
Layers++ [37] | 65.6 | 1.97 58 | 3.13 60 | 0.84 21 | 1.34 2 | 1.72 3 | 0.98 6 | 10.9 91 | 17.8 113 | 13.8 97 | 2.74 59 | 3.71 85 | 2.06 30 | 6.74 130 | 7.67 130 | 8.88 129 | 5.04 116 | 7.20 116 | 4.08 113 | 5.81 50 | 8.98 50 | 1.54 3 | 5.06 29 | 6.39 30 | 1.19 53 |
Brox et al. [5] | 65.9 | 2.86 126 | 4.57 126 | 0.85 30 | 1.89 51 | 2.48 65 | 1.41 54 | 5.02 16 | 5.06 9 | 5.06 23 | 2.74 59 | 2.48 6 | 2.20 37 | 2.57 25 | 2.91 25 | 1.28 24 | 4.77 110 | 6.81 110 | 4.16 114 | 17.2 131 | 26.6 131 | 21.6 131 | 8.51 86 | 10.8 87 | 0.88 5 |
Nguyen [33] | 66.6 | 2.62 116 | 4.16 117 | 0.94 72 | 2.35 108 | 2.74 97 | 1.78 97 | 5.37 24 | 5.59 19 | 4.85 18 | 2.76 61 | 2.97 32 | 2.96 90 | 3.62 48 | 4.12 48 | 1.25 10 | 4.16 85 | 5.94 86 | 3.86 107 | 6.94 84 | 10.7 86 | 3.64 57 | 7.20 61 | 9.10 62 | 0.96 13 |
Sparse-NonSparse [56] | 67.4 | 1.98 66 | 3.14 65 | 0.85 30 | 1.72 31 | 2.11 20 | 1.36 49 | 8.49 69 | 13.4 96 | 10.4 71 | 2.15 11 | 2.87 25 | 1.47 3 | 4.86 114 | 5.52 114 | 6.00 77 | 4.42 98 | 6.31 99 | 2.81 84 | 6.82 80 | 10.5 80 | 4.64 96 | 8.92 93 | 11.3 93 | 1.19 53 |
SRR-TVOF-NL [91] | 67.8 | 1.49 26 | 2.29 26 | 0.95 77 | 1.85 49 | 2.40 53 | 1.41 54 | 14.1 114 | 20.2 121 | 16.3 116 | 2.54 45 | 3.41 66 | 1.81 21 | 4.52 96 | 5.14 96 | 6.60 120 | 4.49 101 | 6.40 102 | 1.88 26 | 5.35 40 | 8.28 40 | 3.57 56 | 5.81 44 | 7.34 45 | 1.50 92 |
FESL [72] | 68.2 | 1.16 9 | 1.69 8 | 0.87 39 | 1.71 30 | 2.18 29 | 1.21 31 | 6.73 43 | 8.98 62 | 6.68 39 | 2.81 68 | 3.41 66 | 3.29 105 | 5.02 120 | 5.71 120 | 6.58 110 | 4.39 96 | 6.26 97 | 2.29 56 | 6.36 70 | 9.84 70 | 4.42 84 | 9.75 108 | 12.3 108 | 1.31 70 |
Local-TV-L1 [65] | 68.3 | 1.56 33 | 2.40 31 | 0.98 87 | 2.43 112 | 2.91 118 | 2.04 108 | 4.29 7 | 5.30 14 | 3.22 7 | 2.27 18 | 2.65 14 | 2.08 32 | 5.34 122 | 6.08 123 | 6.63 122 | 3.66 52 | 5.02 50 | 2.69 75 | 7.43 100 | 11.5 101 | 4.10 72 | 7.36 63 | 9.29 64 | 1.92 115 |
DF-Auto [115] | 68.7 | 2.69 121 | 4.30 123 | 1.12 108 | 2.13 82 | 2.69 88 | 1.58 75 | 6.20 34 | 8.02 56 | 5.50 26 | 2.68 53 | 3.34 60 | 2.70 75 | 1.59 5 | 1.76 5 | 1.28 24 | 3.52 47 | 5.01 48 | 1.68 13 | 8.36 109 | 12.9 109 | 10.1 121 | 8.45 83 | 10.7 84 | 1.67 100 |
Fusion [6] | 69.0 | 2.25 96 | 3.58 97 | 1.33 120 | 1.95 61 | 2.53 70 | 1.23 35 | 8.48 68 | 5.32 15 | 10.5 72 | 2.89 72 | 3.75 87 | 1.67 16 | 4.65 104 | 5.29 104 | 3.65 58 | 5.17 119 | 7.37 119 | 4.89 123 | 5.87 54 | 9.09 54 | 2.49 18 | 3.09 9 | 3.89 9 | 1.40 76 |
TI-DOFE [24] | 69.2 | 2.32 100 | 3.67 100 | 1.10 105 | 2.56 122 | 2.92 119 | 2.11 111 | 4.25 3 | 4.63 6 | 4.31 15 | 3.59 107 | 3.76 90 | 2.86 82 | 3.24 36 | 3.69 36 | 1.32 36 | 4.55 102 | 6.50 104 | 3.36 99 | 4.41 17 | 6.81 17 | 2.76 26 | 9.18 98 | 11.6 98 | 1.11 32 |
Black & Anandan [4] | 70.1 | 2.44 106 | 3.86 107 | 1.02 95 | 2.49 118 | 2.93 120 | 1.98 104 | 13.5 111 | 7.92 53 | 14.3 100 | 3.14 92 | 3.16 47 | 2.55 60 | 3.13 32 | 3.55 32 | 1.27 19 | 3.51 46 | 5.01 48 | 2.20 48 | 5.13 33 | 7.94 33 | 3.20 41 | 7.26 62 | 9.17 63 | 1.89 113 |
Ramp [62] | 70.7 | 1.99 70 | 3.16 71 | 0.84 21 | 1.76 39 | 2.17 28 | 1.33 46 | 12.3 106 | 18.7 118 | 15.2 110 | 2.41 26 | 3.25 53 | 2.12 35 | 4.79 109 | 5.45 109 | 6.02 78 | 4.46 99 | 6.36 100 | 2.80 82 | 5.82 51 | 9.00 51 | 3.35 46 | 8.47 85 | 10.7 84 | 1.41 79 |
TriangleFlow [30] | 70.9 | 1.80 45 | 2.83 44 | 0.95 77 | 2.18 86 | 2.77 100 | 1.61 79 | 7.36 57 | 9.16 64 | 8.83 60 | 2.70 54 | 3.56 78 | 2.88 85 | 3.17 34 | 3.60 34 | 1.27 19 | 5.17 119 | 7.38 120 | 5.34 126 | 7.58 104 | 11.7 104 | 5.87 110 | 5.12 31 | 6.45 32 | 1.14 40 |
Occlusion-TV-L1 [63] | 71.0 | 2.12 84 | 3.37 87 | 1.25 118 | 2.30 101 | 2.90 115 | 1.91 102 | 4.07 2 | 5.27 12 | 3.82 12 | 3.60 109 | 4.67 118 | 3.14 103 | 3.03 30 | 3.45 30 | 1.24 7 | 3.73 56 | 5.32 57 | 2.95 88 | 6.99 89 | 9.92 72 | 4.22 78 | 9.25 100 | 11.7 100 | 1.12 35 |
Shiralkar [42] | 71.2 | 2.18 89 | 3.46 92 | 0.88 44 | 2.25 95 | 2.71 91 | 1.88 100 | 7.65 62 | 6.40 34 | 9.36 67 | 3.87 116 | 4.79 119 | 2.75 78 | 3.52 45 | 4.00 45 | 3.60 56 | 3.75 59 | 5.34 60 | 2.74 78 | 6.85 82 | 10.6 83 | 2.98 36 | 7.97 71 | 10.1 72 | 1.12 35 |
AGIF+OF [85] | 71.2 | 1.61 36 | 2.52 37 | 0.82 8 | 1.79 44 | 2.25 38 | 1.40 53 | 7.38 58 | 11.2 87 | 9.13 62 | 2.97 76 | 3.75 87 | 3.12 101 | 4.51 95 | 5.13 95 | 6.54 99 | 3.89 69 | 5.54 70 | 2.34 60 | 5.85 52 | 9.05 52 | 4.40 82 | 10.8 127 | 13.6 127 | 1.55 95 |
StereoOF-V1MT [119] | 71.6 | 1.70 40 | 2.65 39 | 0.87 39 | 2.10 77 | 2.64 81 | 1.74 95 | 9.86 85 | 4.53 5 | 12.2 87 | 4.12 118 | 4.14 106 | 4.98 122 | 4.38 74 | 4.98 73 | 6.21 82 | 3.56 48 | 4.72 34 | 3.00 90 | 7.15 94 | 11.1 95 | 4.77 100 | 6.63 55 | 8.37 56 | 1.02 24 |
Filter Flow [19] | 71.7 | 2.67 119 | 4.25 120 | 0.94 72 | 2.40 109 | 2.89 113 | 1.70 90 | 7.19 53 | 10.2 78 | 8.54 59 | 3.40 99 | 3.60 79 | 3.38 108 | 2.09 15 | 2.36 16 | 1.32 36 | 3.73 56 | 5.32 57 | 2.69 75 | 5.67 47 | 8.77 47 | 4.39 81 | 7.91 69 | 9.99 70 | 1.19 53 |
2D-CLG [1] | 72.0 | 2.03 77 | 3.22 77 | 0.89 53 | 2.24 93 | 2.67 84 | 1.80 98 | 7.75 63 | 4.01 4 | 9.13 62 | 3.05 83 | 2.78 21 | 3.46 111 | 6.30 129 | 7.17 129 | 9.00 130 | 2.62 6 | 3.71 6 | 2.33 58 | 6.41 72 | 9.92 72 | 3.52 55 | 9.35 103 | 11.8 103 | 1.13 39 |
Classic+CPF [83] | 73.0 | 1.98 66 | 3.14 65 | 0.86 34 | 1.82 47 | 2.26 40 | 1.47 63 | 7.22 55 | 10.0 74 | 7.55 46 | 2.31 20 | 3.11 42 | 1.44 1 | 4.80 110 | 5.46 110 | 6.56 102 | 4.55 102 | 6.48 103 | 2.85 86 | 6.73 76 | 10.4 78 | 4.62 95 | 10.4 120 | 13.2 120 | 1.63 98 |
BlockOverlap [61] | 73.2 | 2.06 79 | 3.26 81 | 1.03 97 | 2.22 90 | 2.67 84 | 1.85 99 | 8.84 76 | 6.32 30 | 11.0 77 | 4.26 120 | 3.93 97 | 5.84 123 | 4.23 65 | 4.81 65 | 1.31 35 | 4.01 76 | 5.71 77 | 3.22 93 | 4.98 31 | 7.64 29 | 4.44 85 | 4.57 21 | 5.77 22 | 1.72 105 |
Adaptive [20] | 73.3 | 2.51 111 | 3.99 111 | 0.93 69 | 2.42 110 | 3.03 123 | 2.24 114 | 6.42 37 | 9.95 73 | 7.88 48 | 2.89 72 | 3.63 80 | 3.26 104 | 4.30 69 | 4.89 69 | 1.26 16 | 4.07 82 | 5.81 83 | 3.25 94 | 6.07 60 | 9.39 61 | 2.94 35 | 7.57 65 | 9.57 66 | 0.91 7 |
PGAM+LK [55] | 73.7 | 2.12 84 | 3.32 85 | 1.16 112 | 2.33 106 | 2.79 103 | 2.01 106 | 18.8 128 | 29.3 131 | 23.5 131 | 4.83 124 | 3.93 97 | 6.42 125 | 2.76 27 | 3.13 27 | 1.47 42 | 3.40 41 | 4.83 42 | 2.49 67 | 3.41 5 | 5.25 5 | 2.90 33 | 6.38 53 | 8.05 54 | 1.15 41 |
Steered-L1 [118] | 73.8 | 1.89 51 | 3.01 52 | 0.89 53 | 1.93 59 | 2.51 68 | 1.50 65 | 17.3 125 | 23.1 127 | 20.6 127 | 3.65 111 | 4.27 111 | 3.39 109 | 3.51 44 | 3.99 44 | 3.99 61 | 3.32 34 | 4.73 35 | 1.95 29 | 7.20 98 | 11.1 95 | 4.57 90 | 8.42 81 | 10.6 82 | 1.01 21 |
TriFlow [95] | 74.0 | 2.63 118 | 4.19 119 | 1.07 103 | 2.05 74 | 2.61 78 | 1.57 72 | 6.62 41 | 10.1 75 | 7.49 45 | 2.70 54 | 3.10 39 | 2.86 82 | 4.46 85 | 5.07 85 | 6.53 97 | 3.72 55 | 5.30 56 | 1.88 26 | 5.90 55 | 9.13 55 | 4.57 90 | 9.77 110 | 12.3 108 | 1.19 53 |
ROF-ND [107] | 74.2 | 1.28 12 | 1.94 11 | 0.83 11 | 3.07 128 | 2.65 83 | 4.13 131 | 7.93 64 | 11.9 90 | 9.17 65 | 3.22 96 | 4.28 112 | 2.58 63 | 4.29 68 | 4.88 68 | 6.50 95 | 4.98 115 | 7.10 115 | 4.47 121 | 6.01 58 | 9.30 59 | 3.24 42 | 6.57 54 | 8.29 55 | 1.27 65 |
Efficient-NL [60] | 74.2 | 1.57 34 | 2.46 35 | 0.84 21 | 1.97 67 | 2.43 57 | 1.46 60 | 11.2 94 | 7.97 54 | 14.4 101 | 2.61 50 | 3.50 74 | 2.08 32 | 4.84 113 | 5.51 113 | 6.23 83 | 4.00 74 | 5.70 74 | 2.09 41 | 7.55 103 | 11.7 104 | 4.15 74 | 10.5 122 | 13.2 120 | 1.43 82 |
RNLOD-Flow [121] | 75.0 | 1.97 58 | 3.13 60 | 0.84 21 | 1.90 53 | 2.43 57 | 1.46 60 | 7.30 56 | 10.8 84 | 8.46 56 | 2.51 39 | 3.02 34 | 3.06 97 | 4.50 92 | 5.12 92 | 6.48 92 | 5.34 121 | 7.62 121 | 4.17 115 | 6.16 65 | 9.53 66 | 3.27 43 | 10.6 124 | 13.4 124 | 1.31 70 |
Heeger++ [104] | 75.5 | 3.50 130 | 5.53 130 | 2.53 129 | 2.00 69 | 2.45 63 | 1.37 50 | 8.20 67 | 5.88 24 | 9.23 66 | 3.50 101 | 3.20 50 | 3.83 115 | 4.31 70 | 4.90 70 | 6.32 89 | 3.45 44 | 4.82 41 | 2.97 89 | 7.84 107 | 10.1 74 | 6.00 114 | 3.91 18 | 4.93 18 | 1.44 84 |
FFV1MT [106] | 75.6 | 2.33 102 | 3.68 101 | 1.04 100 | 2.15 84 | 2.53 70 | 1.51 67 | 8.99 78 | 9.89 71 | 10.6 73 | 3.50 101 | 3.20 50 | 3.83 115 | 3.19 35 | 3.62 35 | 3.04 54 | 4.61 105 | 6.56 105 | 3.72 106 | 6.78 78 | 10.5 80 | 5.04 101 | 2.81 2 | 3.54 2 | 1.67 100 |
RFlow [90] | 75.7 | 1.17 10 | 1.77 10 | 0.88 44 | 2.16 85 | 2.68 86 | 1.68 86 | 10.8 90 | 15.8 107 | 12.8 90 | 3.06 85 | 3.81 92 | 3.10 99 | 4.14 61 | 4.71 61 | 1.30 31 | 4.00 74 | 5.70 74 | 2.75 79 | 6.80 79 | 10.5 80 | 3.39 48 | 10.0 115 | 12.7 115 | 1.92 115 |
IAOF2 [51] | 76.0 | 2.09 83 | 3.25 80 | 1.11 107 | 2.25 95 | 2.81 105 | 1.69 87 | 6.45 38 | 8.52 58 | 6.67 38 | 2.44 29 | 3.02 34 | 2.49 56 | 4.62 103 | 5.25 103 | 5.71 70 | 3.94 72 | 5.61 73 | 3.18 92 | 5.86 53 | 9.06 53 | 4.66 97 | 9.21 99 | 11.6 98 | 1.69 102 |
FlowFields+ [130] | 76.7 | 2.34 103 | 3.73 103 | 1.81 126 | 1.50 8 | 1.97 9 | 1.00 9 | 12.7 109 | 17.9 115 | 15.4 112 | 3.08 87 | 4.14 106 | 2.41 50 | 4.44 82 | 5.05 82 | 6.58 110 | 4.11 83 | 5.87 84 | 2.06 36 | 8.38 110 | 13.0 110 | 5.69 109 | 5.43 35 | 6.86 36 | 1.07 27 |
Classic+NL [31] | 76.8 | 1.97 58 | 3.13 60 | 0.88 44 | 1.78 42 | 2.22 33 | 1.44 56 | 11.8 102 | 17.8 113 | 14.5 104 | 2.47 33 | 3.34 60 | 2.43 52 | 4.81 112 | 5.47 112 | 5.90 74 | 4.41 97 | 6.29 98 | 2.58 69 | 7.03 92 | 10.9 91 | 5.12 104 | 9.04 94 | 11.4 94 | 1.18 49 |
BriefMatch [124] | 77.6 | 1.45 25 | 2.23 22 | 1.06 101 | 2.03 71 | 2.62 80 | 2.01 106 | 11.4 97 | 6.57 37 | 14.0 98 | 4.56 121 | 3.92 96 | 6.24 124 | 3.37 38 | 3.80 38 | 2.51 48 | 4.47 100 | 6.37 101 | 2.80 82 | 6.03 59 | 9.28 58 | 7.81 119 | 8.90 92 | 5.08 19 | 13.3 131 |
LocallyOriented [52] | 77.8 | 1.99 70 | 3.14 65 | 0.92 64 | 2.22 90 | 2.68 86 | 1.60 77 | 12.0 104 | 15.6 106 | 14.9 107 | 4.67 123 | 5.61 124 | 2.20 37 | 4.22 63 | 4.80 63 | 2.99 53 | 3.66 52 | 5.22 55 | 2.20 48 | 5.95 57 | 9.21 57 | 3.13 40 | 10.7 126 | 13.5 126 | 1.38 75 |
SLK [47] | 78.0 | 1.66 38 | 2.59 38 | 0.99 88 | 2.25 95 | 2.58 76 | 1.73 93 | 13.8 113 | 9.17 65 | 15.3 111 | 4.21 119 | 5.03 123 | 4.67 120 | 4.31 70 | 4.90 70 | 4.35 63 | 3.88 68 | 5.52 69 | 2.78 81 | 9.91 123 | 15.3 123 | 3.29 44 | 4.40 20 | 5.55 21 | 1.15 41 |
AggregFlow [97] | 78.9 | 2.69 121 | 4.28 122 | 0.86 34 | 1.76 39 | 2.31 46 | 1.34 47 | 11.2 94 | 16.4 108 | 13.7 95 | 3.22 96 | 4.29 113 | 2.65 72 | 2.08 14 | 2.34 14 | 2.34 47 | 3.26 29 | 4.65 29 | 2.14 45 | 12.4 127 | 19.2 127 | 15.8 129 | 9.95 113 | 12.6 113 | 2.03 120 |
FC-2Layers-FF [74] | 79.3 | 1.91 52 | 2.98 51 | 0.95 77 | 1.34 2 | 1.73 4 | 1.04 17 | 11.8 102 | 17.6 112 | 14.4 101 | 3.36 98 | 4.58 116 | 1.84 22 | 5.00 119 | 5.69 119 | 6.49 94 | 3.93 71 | 5.60 72 | 2.61 71 | 7.47 102 | 11.6 103 | 4.60 94 | 9.38 104 | 11.9 104 | 1.58 97 |
TV-L1-improved [17] | 79.4 | 1.54 30 | 2.40 31 | 1.00 91 | 2.46 115 | 3.09 127 | 2.33 118 | 11.4 97 | 6.97 45 | 14.4 101 | 2.35 24 | 2.50 7 | 2.44 53 | 4.48 89 | 5.09 88 | 1.32 36 | 4.23 90 | 6.03 91 | 3.29 96 | 7.02 90 | 10.9 91 | 3.93 68 | 9.39 105 | 11.9 104 | 2.01 119 |
S2D-Matching [84] | 79.6 | 1.98 66 | 3.14 65 | 0.90 60 | 2.02 70 | 2.55 73 | 1.63 82 | 10.3 87 | 15.0 103 | 11.9 84 | 2.76 61 | 3.74 86 | 2.15 36 | 4.02 58 | 4.57 58 | 5.30 68 | 5.99 127 | 8.55 127 | 5.16 124 | 5.45 44 | 8.43 44 | 4.59 93 | 9.75 108 | 12.3 108 | 1.41 79 |
HBM-GC [105] | 80.2 | 1.99 70 | 3.16 71 | 0.97 83 | 1.98 68 | 2.60 77 | 1.26 40 | 9.23 79 | 8.07 57 | 11.4 81 | 3.08 87 | 4.11 104 | 2.59 65 | 4.61 102 | 5.24 102 | 6.64 124 | 4.79 111 | 6.83 111 | 3.97 110 | 6.97 88 | 10.8 88 | 4.41 83 | 2.84 3 | 3.58 3 | 1.98 118 |
ProbFlowFields [128] | 80.5 | 2.07 81 | 3.30 83 | 0.86 34 | 1.63 21 | 2.15 24 | 1.08 22 | 15.1 119 | 21.3 123 | 17.8 121 | 2.85 69 | 3.87 94 | 2.56 61 | 4.41 78 | 5.02 78 | 6.55 100 | 4.37 94 | 6.23 95 | 3.86 107 | 8.64 111 | 13.4 111 | 11.3 125 | 5.51 37 | 6.96 38 | 1.75 106 |
Aniso-Texture [82] | 84.2 | 1.39 18 | 2.16 19 | 0.84 21 | 2.48 117 | 3.09 127 | 2.36 121 | 6.94 50 | 10.3 79 | 8.32 55 | 8.54 130 | 10.4 130 | 11.9 130 | 4.54 98 | 5.16 98 | 6.60 120 | 5.09 117 | 7.25 117 | 4.60 122 | 6.96 86 | 10.8 88 | 3.68 59 | 5.80 43 | 7.32 44 | 1.11 32 |
UnFlow [129] | 84.7 | 2.61 114 | 4.16 117 | 1.12 108 | 1.95 61 | 2.40 53 | 1.57 72 | 6.73 43 | 9.90 72 | 7.88 48 | 2.38 25 | 3.19 49 | 1.69 17 | 4.55 100 | 5.17 100 | 6.37 91 | 5.77 126 | 8.23 126 | 5.32 125 | 7.15 94 | 11.1 95 | 2.87 31 | 11.3 131 | 14.3 131 | 1.70 103 |
Adaptive flow [45] | 84.9 | 2.69 121 | 4.00 113 | 1.20 115 | 2.49 118 | 2.95 122 | 1.99 105 | 8.50 70 | 9.02 63 | 10.8 76 | 3.93 117 | 4.14 106 | 4.77 121 | 5.90 126 | 6.71 126 | 5.77 71 | 4.85 113 | 6.91 113 | 4.19 116 | 4.88 25 | 7.54 25 | 3.70 61 | 2.88 5 | 3.64 5 | 0.86 4 |
SILK [79] | 85.2 | 1.72 42 | 2.69 42 | 1.00 91 | 2.80 126 | 2.87 111 | 3.18 126 | 19.4 129 | 18.6 117 | 18.8 126 | 3.49 100 | 4.07 101 | 3.78 114 | 3.44 40 | 3.90 40 | 2.57 49 | 4.69 107 | 6.68 107 | 2.82 85 | 3.07 2 | 4.72 2 | 3.31 45 | 10.9 128 | 13.8 128 | 1.45 87 |
Rannacher [23] | 85.7 | 2.22 95 | 3.54 96 | 0.92 64 | 2.45 113 | 3.08 126 | 2.39 124 | 11.7 100 | 9.26 67 | 14.6 105 | 2.97 76 | 3.88 95 | 2.44 53 | 3.71 51 | 4.22 51 | 1.29 29 | 4.68 106 | 6.67 106 | 3.29 96 | 7.02 90 | 10.9 91 | 3.72 62 | 8.05 75 | 10.2 76 | 1.78 110 |
GraphCuts [14] | 85.8 | 2.47 109 | 3.90 108 | 1.03 97 | 1.91 57 | 2.43 57 | 1.56 71 | 11.0 92 | 6.33 31 | 13.7 95 | 2.97 76 | 3.47 73 | 3.11 100 | 5.47 124 | 6.22 124 | 7.89 126 | 3.33 35 | 4.75 36 | 2.03 33 | 8.66 112 | 13.4 111 | 6.07 115 | 9.96 114 | 12.6 113 | 1.18 49 |
Correlation Flow [75] | 86.5 | 2.01 74 | 3.18 74 | 0.91 62 | 2.18 86 | 2.73 94 | 1.60 77 | 8.77 74 | 12.9 93 | 10.3 70 | 3.00 79 | 4.07 101 | 2.00 27 | 3.52 45 | 4.00 45 | 4.02 62 | 6.31 128 | 9.00 128 | 6.21 128 | 7.72 106 | 11.9 107 | 5.07 103 | 8.86 91 | 11.2 92 | 2.61 129 |
TF+OM [100] | 86.6 | 1.95 56 | 3.07 54 | 1.54 124 | 1.73 34 | 2.25 38 | 1.28 42 | 6.86 48 | 10.4 82 | 7.88 48 | 3.72 115 | 4.88 120 | 2.97 91 | 5.54 125 | 6.30 125 | 8.26 128 | 4.18 86 | 5.97 88 | 2.90 87 | 7.63 105 | 11.8 106 | 5.25 106 | 8.05 75 | 10.2 76 | 2.03 120 |
FOLKI [16] | 87.2 | 1.98 66 | 3.08 56 | 1.24 117 | 2.52 121 | 2.83 107 | 2.33 118 | 10.1 86 | 8.57 59 | 12.8 90 | 4.91 125 | 4.36 114 | 6.54 126 | 3.10 31 | 3.52 31 | 3.73 59 | 8.35 130 | 11.9 130 | 9.26 131 | 4.40 16 | 6.80 16 | 5.91 112 | 9.08 96 | 11.5 96 | 1.22 59 |
Dynamic MRF [7] | 87.5 | 2.00 73 | 3.18 74 | 0.88 44 | 2.26 98 | 2.87 111 | 2.30 117 | 6.85 47 | 6.45 35 | 7.97 51 | 3.65 111 | 4.22 110 | 4.17 118 | 4.47 88 | 5.09 88 | 6.53 97 | 4.03 77 | 5.73 78 | 3.48 102 | 8.15 108 | 12.6 108 | 4.25 79 | 9.08 96 | 11.5 96 | 1.53 94 |
HCIC-L [99] | 88.4 | 2.19 92 | 3.36 86 | 1.38 121 | 1.90 53 | 2.28 43 | 1.62 81 | 15.4 120 | 22.7 126 | 18.7 124 | 3.66 113 | 4.96 122 | 2.28 45 | 2.47 22 | 2.79 22 | 1.28 24 | 5.47 124 | 7.81 124 | 5.56 127 | 10.4 125 | 16.2 125 | 10.4 123 | 5.04 25 | 6.36 26 | 2.40 128 |
IAOF [50] | 88.9 | 3.34 128 | 5.18 128 | 3.47 131 | 2.75 125 | 3.20 131 | 2.07 109 | 11.7 100 | 16.6 109 | 14.1 99 | 3.56 104 | 3.67 83 | 3.65 112 | 3.65 49 | 4.16 50 | 1.24 7 | 3.82 64 | 5.45 65 | 2.66 73 | 6.34 68 | 9.81 68 | 3.45 50 | 9.42 106 | 11.9 104 | 1.31 70 |
Learning Flow [11] | 89.0 | 2.38 104 | 3.78 105 | 0.95 77 | 2.24 93 | 2.80 104 | 1.63 82 | 20.4 131 | 24.8 129 | 20.8 128 | 3.05 83 | 3.10 39 | 2.35 48 | 4.76 107 | 5.41 107 | 5.57 69 | 3.79 62 | 5.41 64 | 3.26 95 | 5.92 56 | 9.16 56 | 3.90 66 | 10.6 124 | 13.4 124 | 1.44 84 |
SegOF [10] | 90.5 | 2.46 108 | 3.92 109 | 0.97 83 | 1.93 59 | 2.42 56 | 1.45 58 | 14.8 118 | 12.5 92 | 16.0 115 | 6.84 127 | 9.29 127 | 6.71 127 | 4.69 105 | 5.34 105 | 6.57 108 | 4.06 80 | 5.78 81 | 2.56 68 | 9.69 122 | 15.0 122 | 7.78 118 | 3.03 8 | 3.82 8 | 1.29 68 |
SimpleFlow [49] | 91.9 | 2.01 74 | 3.19 76 | 0.86 34 | 2.04 72 | 2.57 75 | 1.51 67 | 19.8 130 | 24.0 128 | 21.5 129 | 3.07 86 | 3.85 93 | 3.31 106 | 4.93 116 | 5.61 116 | 6.26 85 | 5.48 125 | 7.81 124 | 3.88 109 | 9.39 120 | 14.5 120 | 10.2 122 | 3.62 14 | 4.57 14 | 1.31 70 |
StereoFlow [44] | 95.6 | 2.98 127 | 4.65 127 | 1.15 110 | 2.26 98 | 2.74 97 | 1.61 79 | 6.11 33 | 7.37 47 | 6.48 36 | 3.16 93 | 4.04 100 | 2.76 79 | 6.10 128 | 6.94 128 | 8.09 127 | 5.13 118 | 7.31 118 | 4.06 112 | 9.04 116 | 14.0 116 | 4.66 97 | 6.69 56 | 8.45 57 | 1.56 96 |
HBpMotionGpu [43] | 95.7 | 2.51 111 | 3.99 111 | 1.40 122 | 2.50 120 | 3.10 129 | 2.21 113 | 9.83 84 | 14.5 100 | 11.5 82 | 3.59 107 | 4.62 117 | 2.59 65 | 7.38 131 | 8.39 131 | 11.8 131 | 5.43 122 | 7.75 122 | 4.40 119 | 5.39 42 | 8.33 41 | 2.06 13 | 5.92 47 | 7.48 48 | 1.47 88 |
GroupFlow [9] | 98.1 | 2.44 106 | 3.75 104 | 1.61 125 | 2.04 72 | 2.53 70 | 1.67 85 | 12.3 106 | 10.6 83 | 12.5 89 | 9.06 131 | 10.6 131 | 12.0 131 | 4.69 105 | 5.34 105 | 6.59 117 | 4.72 109 | 6.72 108 | 4.23 118 | 8.80 114 | 13.6 114 | 4.48 86 | 3.88 17 | 4.90 17 | 1.81 111 |
SPSA-learn [13] | 98.8 | 3.49 129 | 5.31 129 | 1.15 110 | 2.30 101 | 2.81 105 | 1.73 93 | 16.0 123 | 14.3 99 | 16.8 119 | 3.70 114 | 3.65 81 | 4.08 117 | 4.54 98 | 5.16 98 | 5.09 67 | 3.82 64 | 5.45 65 | 3.56 105 | 16.4 130 | 25.3 130 | 19.6 130 | 4.18 19 | 5.27 20 | 2.10 124 |
Pyramid LK [2] | 99.4 | 2.62 116 | 4.11 114 | 1.23 116 | 3.77 130 | 2.94 121 | 2.29 116 | 17.2 124 | 10.1 75 | 16.4 117 | 7.25 128 | 9.29 127 | 8.44 128 | 5.34 122 | 6.07 122 | 3.60 56 | 4.58 104 | 4.97 47 | 3.49 103 | 9.56 121 | 14.8 121 | 4.29 80 | 3.73 15 | 4.70 15 | 1.28 67 |
2bit-BM-tele [98] | 106.7 | 2.60 113 | 4.11 114 | 1.19 114 | 2.42 110 | 3.07 125 | 2.37 123 | 18.2 127 | 27.8 130 | 23.0 130 | 3.19 95 | 3.75 87 | 3.71 113 | 4.50 92 | 5.12 92 | 5.98 76 | 6.47 129 | 9.23 129 | 6.38 129 | 14.3 128 | 22.1 128 | 15.2 128 | 3.15 11 | 3.97 11 | 2.39 127 |
Periodicity [78] | 120.4 | 5.15 131 | 7.86 131 | 3.24 130 | 5.39 131 | 3.06 124 | 3.39 128 | 17.5 126 | 17.3 110 | 18.7 124 | 7.38 129 | 9.91 129 | 8.96 129 | 5.97 127 | 6.80 127 | 6.94 125 | 8.36 131 | 11.9 130 | 8.85 130 | 15.0 129 | 23.2 129 | 14.1 127 | 5.71 41 | 7.19 42 | 4.90 130 |
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. |