Optical flow evaluation results |
Statistics:
Average
SD
R2.5
R5.0
R10.0
A90
A95
A99
Error type: endpoint angle interpolation normalized interpolation |
Average 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 | |
MDP-Flow2 [68] | 13.7 | 2.89 5 | 5.38 6 | 1.19 2 | 3.47 9 | 5.07 11 | 1.26 1 | 3.66 12 | 6.10 36 | 2.48 65 | 5.20 2 | 7.48 11 | 3.14 12 | 10.2 3 | 12.8 5 | 3.61 24 | 6.13 21 | 11.8 18 | 2.31 24 | 7.36 6 | 16.8 4 | 1.49 2 | 7.75 21 | 12.1 21 | 1.69 8 |
PMMST [114] | 14.0 | 2.90 6 | 5.43 8 | 1.20 6 | 3.50 13 | 5.05 10 | 1.27 6 | 3.56 3 | 5.46 3 | 1.82 30 | 5.38 18 | 7.92 37 | 3.41 81 | 10.2 3 | 12.8 5 | 3.60 18 | 5.76 3 | 11.0 3 | 2.26 5 | 7.39 7 | 16.9 7 | 1.53 13 | 7.57 7 | 11.8 7 | 1.72 36 |
SepConv-v1 [127] | 20.6 | 2.52 1 | 4.83 1 | 1.11 1 | 3.56 21 | 5.04 9 | 1.90 107 | 4.17 48 | 4.15 1 | 2.86 79 | 5.41 27 | 6.81 1 | 3.88 111 | 10.2 3 | 12.8 5 | 3.37 2 | 5.47 2 | 10.4 2 | 2.21 1 | 6.88 1 | 15.6 1 | 1.72 68 | 6.63 1 | 10.3 1 | 1.62 1 |
DeepFlow [86] | 22.6 | 2.98 18 | 5.67 23 | 1.22 43 | 3.88 53 | 5.78 52 | 1.52 62 | 3.62 5 | 5.93 30 | 1.34 3 | 5.39 23 | 7.20 6 | 3.17 16 | 11.0 33 | 13.9 38 | 3.63 34 | 5.91 8 | 11.3 7 | 2.29 16 | 7.14 2 | 16.3 2 | 1.49 2 | 7.80 27 | 12.2 25 | 1.70 14 |
CBF [12] | 25.8 | 2.83 2 | 5.20 2 | 1.23 61 | 3.97 62 | 5.79 55 | 1.56 64 | 3.62 5 | 5.47 4 | 1.60 17 | 5.21 3 | 7.12 3 | 3.29 51 | 10.1 1 | 12.6 1 | 3.62 28 | 5.97 11 | 11.5 11 | 2.31 24 | 7.76 35 | 17.8 36 | 1.61 40 | 7.60 10 | 11.9 10 | 1.76 83 |
DeepFlow2 [108] | 27.3 | 2.99 21 | 5.65 20 | 1.22 43 | 3.88 53 | 5.79 55 | 1.48 57 | 3.62 5 | 6.03 31 | 1.34 3 | 5.38 18 | 7.44 10 | 3.22 32 | 11.0 33 | 13.8 33 | 3.67 40 | 5.83 4 | 11.2 4 | 2.25 4 | 7.60 19 | 17.4 21 | 1.50 4 | 7.82 28 | 12.2 25 | 1.77 92 |
SuperFlow [81] | 28.2 | 2.94 10 | 5.56 16 | 1.24 72 | 3.99 65 | 5.78 52 | 1.67 83 | 4.06 38 | 5.55 6 | 1.83 31 | 5.52 37 | 7.07 2 | 3.20 26 | 10.2 3 | 12.7 3 | 3.68 43 | 6.13 21 | 11.8 18 | 2.24 2 | 7.68 27 | 17.5 25 | 1.77 79 | 7.44 5 | 11.6 4 | 1.69 8 |
NN-field [71] | 29.4 | 2.98 18 | 5.70 24 | 1.20 6 | 3.31 3 | 4.73 3 | 1.26 1 | 4.69 79 | 5.91 28 | 2.03 42 | 5.99 85 | 9.13 98 | 3.57 97 | 10.3 9 | 12.8 5 | 3.60 18 | 6.24 33 | 12.0 30 | 2.31 24 | 7.39 7 | 16.9 7 | 1.54 18 | 7.69 18 | 12.0 16 | 1.72 36 |
NNF-Local [87] | 29.4 | 2.92 9 | 5.51 14 | 1.19 2 | 3.30 2 | 4.71 2 | 1.26 1 | 3.65 10 | 5.91 28 | 2.29 58 | 5.76 63 | 8.70 82 | 3.55 95 | 10.3 9 | 12.9 11 | 3.60 18 | 6.42 51 | 12.4 50 | 2.34 37 | 7.57 16 | 17.4 21 | 1.74 70 | 7.61 11 | 11.9 10 | 1.72 36 |
Aniso. Huber-L1 [22] | 29.8 | 2.95 13 | 5.44 10 | 1.24 72 | 4.42 93 | 6.27 93 | 1.67 83 | 3.79 23 | 5.70 13 | 1.50 10 | 5.31 8 | 7.42 9 | 3.24 41 | 11.1 43 | 14.0 51 | 3.61 24 | 5.91 8 | 11.4 9 | 2.24 2 | 7.60 19 | 17.3 15 | 1.51 7 | 7.62 13 | 11.9 10 | 1.73 45 |
CLG-TV [48] | 30.7 | 2.94 10 | 5.45 11 | 1.25 81 | 4.26 81 | 6.17 81 | 1.60 70 | 3.68 16 | 5.73 15 | 1.73 26 | 5.36 14 | 7.41 8 | 3.32 67 | 11.1 43 | 14.0 51 | 3.57 7 | 5.88 7 | 11.3 7 | 2.26 5 | 7.58 17 | 17.0 11 | 1.57 31 | 7.75 21 | 12.1 21 | 1.72 36 |
IROF-TV [53] | 31.7 | 3.07 36 | 5.91 45 | 1.23 61 | 3.71 37 | 5.47 34 | 1.40 38 | 3.70 18 | 6.27 42 | 1.58 16 | 5.25 5 | 7.60 18 | 3.17 16 | 11.0 33 | 13.9 38 | 4.47 97 | 6.37 47 | 12.4 50 | 2.30 21 | 7.79 39 | 17.9 39 | 1.50 4 | 7.63 14 | 11.9 10 | 1.66 2 |
LME [70] | 32.2 | 2.95 13 | 5.59 18 | 1.19 2 | 3.68 35 | 5.50 38 | 1.38 33 | 4.06 38 | 7.00 77 | 1.71 25 | 5.38 18 | 7.92 37 | 3.18 19 | 11.2 56 | 14.1 59 | 4.51 117 | 6.29 38 | 12.2 38 | 2.31 24 | 7.33 4 | 16.8 4 | 1.51 7 | 7.83 29 | 12.3 29 | 1.70 14 |
IROF++ [58] | 32.6 | 3.03 27 | 5.77 32 | 1.20 6 | 3.59 24 | 5.31 22 | 1.33 24 | 4.32 61 | 6.61 58 | 2.25 54 | 5.06 1 | 7.14 4 | 3.16 15 | 11.0 33 | 13.9 38 | 4.44 93 | 6.34 43 | 12.3 45 | 2.27 10 | 7.54 15 | 17.3 15 | 1.64 51 | 8.09 51 | 12.7 52 | 1.69 8 |
NNF-EAC [103] | 33.7 | 3.01 24 | 5.60 19 | 1.25 81 | 3.63 27 | 5.36 27 | 1.29 13 | 4.17 48 | 7.03 79 | 2.99 82 | 5.50 36 | 7.96 39 | 3.28 49 | 11.2 56 | 14.1 59 | 3.60 18 | 5.86 6 | 11.2 4 | 2.26 5 | 7.43 11 | 17.0 11 | 1.54 18 | 7.79 26 | 12.2 25 | 1.73 45 |
CombBMOF [113] | 34.2 | 3.16 65 | 5.88 41 | 1.24 72 | 3.54 17 | 5.24 17 | 1.34 28 | 4.01 34 | 6.45 53 | 2.20 51 | 5.62 55 | 8.22 55 | 3.29 51 | 10.7 19 | 13.5 20 | 3.62 28 | 6.20 29 | 11.9 26 | 2.27 10 | 7.78 38 | 17.3 15 | 1.56 28 | 7.75 21 | 12.1 21 | 1.71 26 |
DF-Auto [115] | 34.2 | 2.94 10 | 5.34 4 | 1.23 61 | 3.99 65 | 5.84 59 | 1.65 77 | 3.85 28 | 6.73 61 | 1.55 15 | 5.38 18 | 7.54 13 | 3.25 43 | 10.4 12 | 13.0 12 | 3.70 45 | 6.17 28 | 11.9 26 | 2.28 13 | 7.94 49 | 18.2 50 | 1.75 74 | 7.68 16 | 12.0 16 | 1.71 26 |
WLIF-Flow [93] | 36.2 | 2.95 13 | 5.53 15 | 1.20 6 | 3.66 32 | 5.41 30 | 1.39 35 | 4.26 57 | 7.17 85 | 2.54 68 | 5.30 7 | 7.57 16 | 3.29 51 | 10.7 19 | 13.5 20 | 3.70 45 | 6.74 92 | 13.1 89 | 2.48 89 | 7.40 9 | 16.9 7 | 1.53 13 | 7.87 35 | 12.3 29 | 1.69 8 |
ALD-Flow [66] | 36.5 | 3.28 90 | 6.45 91 | 1.24 72 | 3.81 44 | 5.73 51 | 1.41 42 | 3.62 5 | 6.28 43 | 1.35 5 | 5.58 46 | 8.39 66 | 3.04 4 | 10.8 22 | 13.5 20 | 4.15 76 | 5.96 10 | 11.4 9 | 2.29 16 | 7.34 5 | 16.8 4 | 1.51 7 | 8.25 70 | 12.9 63 | 1.70 14 |
Second-order prior [8] | 36.8 | 2.91 8 | 5.39 7 | 1.24 72 | 4.26 81 | 6.21 86 | 1.56 64 | 3.82 25 | 6.34 45 | 1.62 18 | 5.39 23 | 7.68 20 | 3.04 4 | 11.1 43 | 13.9 38 | 3.59 11 | 6.14 24 | 11.9 26 | 2.31 24 | 7.61 21 | 17.4 21 | 1.63 50 | 7.90 37 | 12.4 39 | 1.78 96 |
PH-Flow [101] | 37.2 | 3.12 51 | 6.01 60 | 1.20 6 | 3.39 5 | 4.94 6 | 1.28 11 | 3.70 18 | 6.43 48 | 2.48 65 | 5.23 4 | 7.58 17 | 3.22 32 | 10.4 12 | 13.1 14 | 3.62 28 | 6.84 102 | 13.3 100 | 2.47 85 | 7.84 41 | 18.1 46 | 1.58 34 | 7.87 35 | 12.3 29 | 1.73 45 |
p-harmonic [29] | 37.9 | 3.00 22 | 5.72 26 | 1.21 19 | 4.33 86 | 6.24 91 | 1.69 88 | 3.60 4 | 6.07 35 | 1.39 6 | 5.70 56 | 7.87 32 | 3.29 51 | 11.0 33 | 13.8 33 | 3.63 34 | 6.02 14 | 11.6 14 | 2.34 37 | 7.67 25 | 17.5 25 | 1.70 63 | 7.92 41 | 12.4 39 | 1.72 36 |
Brox et al. [5] | 38.8 | 3.08 39 | 5.94 50 | 1.21 19 | 3.83 47 | 5.67 44 | 1.45 51 | 3.93 30 | 5.76 17 | 1.67 20 | 5.32 9 | 7.19 5 | 3.22 32 | 10.6 16 | 13.4 18 | 3.56 5 | 6.60 77 | 12.7 65 | 2.42 72 | 8.61 87 | 19.7 89 | 3.04 126 | 7.43 4 | 11.6 4 | 1.68 6 |
SIOF [67] | 40.1 | 3.06 34 | 5.74 30 | 1.24 72 | 4.40 92 | 6.40 103 | 1.63 75 | 4.17 48 | 7.43 96 | 1.93 36 | 5.40 26 | 7.75 25 | 3.44 84 | 10.1 1 | 12.6 1 | 3.58 9 | 6.10 18 | 11.8 18 | 2.29 16 | 7.52 13 | 17.2 14 | 1.53 13 | 7.96 45 | 12.5 48 | 1.73 45 |
FMOF [94] | 40.4 | 3.16 65 | 5.92 48 | 1.23 61 | 3.48 10 | 5.07 11 | 1.28 11 | 4.59 75 | 6.82 64 | 2.78 75 | 5.71 58 | 8.42 67 | 3.40 79 | 10.4 12 | 13.0 12 | 3.67 40 | 6.49 58 | 12.6 59 | 2.28 13 | 7.64 23 | 17.5 25 | 1.48 1 | 8.06 50 | 12.6 49 | 1.67 4 |
Local-TV-L1 [65] | 41.8 | 3.00 22 | 5.47 12 | 1.30 107 | 4.43 95 | 6.23 90 | 1.75 96 | 3.50 1 | 5.35 2 | 1.45 7 | 5.39 23 | 7.56 14 | 3.29 51 | 11.2 56 | 14.1 59 | 3.91 65 | 6.16 26 | 11.8 18 | 2.47 85 | 7.67 25 | 17.6 29 | 1.55 23 | 7.57 7 | 11.8 7 | 1.76 83 |
MDP-Flow [26] | 42.2 | 2.86 3 | 5.34 4 | 1.20 6 | 3.49 12 | 5.15 14 | 1.34 28 | 4.01 34 | 5.51 5 | 2.28 56 | 5.58 46 | 7.91 36 | 3.33 70 | 11.2 56 | 14.0 51 | 4.49 105 | 6.72 87 | 13.1 89 | 2.54 106 | 7.71 30 | 17.7 33 | 1.74 70 | 7.83 29 | 12.3 29 | 1.70 14 |
OAR-Flow [125] | 44.3 | 3.13 57 | 5.95 52 | 1.22 43 | 3.83 47 | 5.70 47 | 1.48 57 | 3.65 10 | 6.06 32 | 1.16 1 | 5.60 51 | 8.48 71 | 3.03 1 | 11.2 56 | 14.1 59 | 4.51 117 | 6.12 20 | 11.8 18 | 2.41 69 | 7.97 52 | 17.9 39 | 1.59 35 | 8.11 52 | 12.7 52 | 1.71 26 |
Modified CLG [34] | 47.0 | 2.87 4 | 5.32 3 | 1.24 72 | 4.51 100 | 6.21 86 | 1.96 116 | 4.15 46 | 6.45 53 | 2.67 71 | 5.56 43 | 7.69 21 | 3.64 102 | 10.8 22 | 13.5 20 | 3.63 34 | 6.36 46 | 12.3 45 | 2.39 57 | 7.46 12 | 17.1 13 | 1.56 28 | 7.86 32 | 12.3 29 | 1.75 72 |
Ad-TV-NDC [36] | 47.5 | 3.23 81 | 5.70 24 | 1.44 124 | 4.78 118 | 6.46 106 | 1.92 110 | 3.67 13 | 5.86 24 | 1.50 10 | 5.97 81 | 8.14 53 | 3.51 91 | 10.8 22 | 13.5 20 | 3.63 34 | 6.24 33 | 12.0 30 | 2.40 59 | 7.70 28 | 17.3 15 | 1.51 7 | 7.48 6 | 11.7 6 | 1.73 45 |
F-TV-L1 [15] | 47.7 | 3.30 92 | 6.36 85 | 1.29 104 | 4.39 91 | 6.32 99 | 1.62 74 | 3.80 24 | 5.90 27 | 1.76 27 | 5.61 52 | 7.97 41 | 3.31 64 | 10.9 29 | 13.6 25 | 3.59 11 | 5.84 5 | 11.2 4 | 2.33 35 | 7.70 28 | 17.6 29 | 1.79 81 | 7.61 11 | 11.9 10 | 1.78 96 |
2DHMM-SAS [92] | 49.2 | 3.10 45 | 5.91 45 | 1.21 19 | 4.10 71 | 6.05 72 | 1.46 55 | 4.38 65 | 7.10 82 | 2.07 43 | 5.38 18 | 7.78 29 | 3.22 32 | 11.3 71 | 14.3 80 | 4.42 89 | 6.33 40 | 12.2 38 | 2.26 5 | 7.95 51 | 18.2 50 | 1.64 51 | 8.19 58 | 12.8 57 | 1.70 14 |
CPM-Flow [116] | 49.5 | 3.17 70 | 6.31 80 | 1.21 19 | 3.54 17 | 5.26 19 | 1.31 19 | 4.22 54 | 5.88 26 | 1.45 7 | 6.11 91 | 9.48 109 | 3.31 64 | 11.1 43 | 13.9 38 | 4.50 108 | 6.28 37 | 12.1 35 | 2.32 32 | 7.66 24 | 17.6 29 | 1.74 70 | 8.18 56 | 12.8 57 | 1.76 83 |
LDOF [28] | 50.5 | 3.03 27 | 5.66 22 | 1.28 99 | 4.06 69 | 5.53 40 | 2.40 127 | 4.32 61 | 6.43 48 | 2.00 39 | 5.45 32 | 7.56 14 | 3.60 100 | 10.2 3 | 12.7 3 | 3.59 11 | 6.39 48 | 12.4 50 | 2.29 16 | 8.36 76 | 19.4 83 | 2.21 109 | 7.57 7 | 11.8 7 | 1.86 120 |
TC/T-Flow [76] | 50.6 | 3.21 78 | 6.24 74 | 1.22 43 | 3.90 57 | 5.86 61 | 1.43 47 | 3.69 17 | 5.83 20 | 1.50 10 | 5.88 74 | 8.93 89 | 3.15 13 | 11.1 43 | 13.9 38 | 4.50 108 | 6.23 30 | 12.0 30 | 2.26 5 | 8.61 87 | 19.0 72 | 1.93 92 | 8.16 55 | 12.8 57 | 1.70 14 |
COFM [59] | 51.0 | 3.03 27 | 5.76 31 | 1.22 43 | 3.55 20 | 5.21 16 | 1.32 22 | 3.82 25 | 6.98 75 | 2.81 76 | 5.41 27 | 7.97 41 | 3.30 59 | 10.8 22 | 13.6 25 | 3.62 28 | 7.01 116 | 13.7 113 | 2.40 59 | 8.00 56 | 18.5 60 | 1.98 95 | 7.91 38 | 12.4 39 | 1.80 110 |
Layers++ [37] | 51.9 | 2.96 16 | 5.56 16 | 1.22 43 | 3.29 1 | 4.64 1 | 1.26 1 | 4.07 40 | 7.24 86 | 3.08 85 | 5.48 33 | 8.10 49 | 3.25 43 | 12.0 125 | 15.2 125 | 4.62 127 | 7.29 119 | 14.3 119 | 2.44 77 | 7.63 22 | 17.5 25 | 1.54 18 | 7.84 31 | 12.3 29 | 1.70 14 |
AdaConv-v1 [126] | 52.1 | 3.57 113 | 6.88 107 | 1.41 121 | 4.34 88 | 5.67 44 | 2.52 129 | 5.00 89 | 5.86 24 | 2.98 81 | 6.91 115 | 8.89 88 | 4.89 125 | 10.2 3 | 12.8 5 | 3.21 1 | 5.33 1 | 10.1 1 | 2.27 10 | 7.30 3 | 16.6 3 | 1.92 91 | 6.94 2 | 10.8 2 | 1.67 4 |
ComplOF-FED-GPU [35] | 52.6 | 3.23 81 | 6.40 86 | 1.22 43 | 3.73 40 | 5.62 43 | 1.44 49 | 5.23 95 | 6.06 32 | 3.23 93 | 5.53 38 | 8.25 56 | 3.29 51 | 11.1 43 | 13.9 38 | 4.21 78 | 6.11 19 | 11.8 18 | 2.32 32 | 8.16 63 | 18.5 60 | 1.61 40 | 8.29 76 | 12.9 63 | 1.71 26 |
TV-L1-MCT [64] | 52.7 | 3.17 70 | 6.05 63 | 1.22 43 | 3.87 50 | 5.82 57 | 1.40 38 | 4.48 72 | 7.75 106 | 2.24 53 | 5.37 16 | 7.76 27 | 3.24 41 | 11.6 109 | 14.7 114 | 4.31 82 | 6.08 16 | 11.7 16 | 2.31 24 | 8.07 59 | 18.6 63 | 2.15 107 | 7.68 16 | 12.0 16 | 1.68 6 |
FlowFields [110] | 52.8 | 3.15 63 | 6.30 78 | 1.21 19 | 3.57 22 | 5.34 25 | 1.32 22 | 4.73 80 | 6.89 69 | 3.23 93 | 5.85 69 | 8.96 92 | 3.08 6 | 10.8 22 | 13.6 25 | 4.19 77 | 6.57 70 | 12.8 76 | 2.36 46 | 7.72 31 | 17.8 36 | 1.67 58 | 8.20 60 | 12.9 63 | 1.74 64 |
nLayers [57] | 52.9 | 3.03 27 | 5.72 26 | 1.21 19 | 3.48 10 | 5.09 13 | 1.31 19 | 5.60 100 | 7.52 99 | 4.26 114 | 5.61 52 | 8.33 61 | 3.29 51 | 11.6 109 | 14.6 109 | 4.31 82 | 6.66 81 | 12.9 83 | 2.40 59 | 7.58 17 | 17.3 15 | 1.59 35 | 7.94 42 | 12.4 39 | 1.69 8 |
Kuang [131] | 53.5 | 3.12 51 | 6.20 72 | 1.21 19 | 3.67 34 | 5.48 37 | 1.37 32 | 5.23 95 | 6.84 65 | 2.75 74 | 5.98 83 | 9.19 102 | 3.13 11 | 11.0 33 | 13.9 38 | 4.12 74 | 6.23 30 | 12.0 30 | 2.31 24 | 7.89 47 | 18.2 50 | 3.07 127 | 8.23 67 | 12.9 63 | 1.71 26 |
CRTflow [80] | 53.6 | 3.09 43 | 5.91 45 | 1.27 95 | 4.35 89 | 6.31 97 | 1.68 86 | 4.15 46 | 7.26 87 | 1.84 32 | 5.33 11 | 7.51 12 | 3.38 75 | 11.0 33 | 13.8 33 | 4.48 99 | 6.09 17 | 11.7 16 | 2.30 21 | 8.55 85 | 19.8 90 | 1.55 23 | 8.19 58 | 12.8 57 | 1.72 36 |
DPOF [18] | 54.4 | 3.34 99 | 6.82 103 | 1.29 104 | 3.40 6 | 4.93 5 | 1.29 13 | 5.00 89 | 6.36 46 | 3.40 96 | 5.86 70 | 8.94 90 | 3.51 91 | 11.0 33 | 13.8 33 | 3.59 11 | 6.56 67 | 12.7 65 | 2.28 13 | 7.99 53 | 18.2 50 | 1.55 23 | 8.24 68 | 12.9 63 | 1.70 14 |
TC-Flow [46] | 54.8 | 3.31 94 | 6.70 100 | 1.22 43 | 3.91 59 | 5.95 63 | 1.45 51 | 3.64 9 | 5.84 21 | 1.28 2 | 5.70 56 | 8.50 73 | 3.22 32 | 11.2 56 | 14.1 59 | 4.44 93 | 6.34 43 | 12.3 45 | 2.41 69 | 7.79 39 | 17.9 39 | 1.55 23 | 8.42 90 | 13.2 92 | 1.74 64 |
AGIF+OF [85] | 55.3 | 3.12 51 | 5.95 52 | 1.20 6 | 3.64 29 | 5.39 28 | 1.40 38 | 3.96 32 | 6.44 52 | 2.28 56 | 5.48 33 | 8.03 45 | 3.25 43 | 11.4 81 | 14.3 80 | 4.49 105 | 6.91 107 | 13.5 108 | 2.37 50 | 7.85 43 | 17.9 39 | 1.54 18 | 8.44 94 | 13.2 92 | 1.73 45 |
Classic++ [32] | 56.4 | 3.05 32 | 5.85 37 | 1.24 72 | 4.08 70 | 6.08 73 | 1.52 62 | 3.74 21 | 5.58 9 | 1.53 14 | 5.72 60 | 8.12 51 | 3.21 28 | 11.4 81 | 14.3 80 | 3.74 55 | 6.68 83 | 13.0 85 | 2.42 72 | 8.35 75 | 19.2 76 | 1.62 47 | 8.21 62 | 12.9 63 | 1.73 45 |
BlockOverlap [61] | 56.5 | 2.98 18 | 5.47 12 | 1.33 114 | 4.38 90 | 6.09 74 | 1.88 106 | 4.26 57 | 5.57 8 | 3.14 88 | 5.56 43 | 7.32 7 | 4.14 118 | 11.1 43 | 13.9 38 | 3.77 58 | 6.41 49 | 12.3 45 | 2.54 106 | 7.75 33 | 17.4 21 | 3.02 125 | 7.32 3 | 11.4 3 | 1.78 96 |
PGM-C [120] | 56.5 | 3.17 70 | 6.29 76 | 1.21 19 | 3.58 23 | 5.32 23 | 1.33 24 | 5.01 91 | 6.14 39 | 1.90 35 | 6.14 94 | 9.63 114 | 3.23 38 | 11.2 56 | 14.1 59 | 4.50 108 | 6.14 24 | 11.8 18 | 2.34 37 | 8.20 65 | 18.9 69 | 1.59 35 | 8.46 97 | 13.3 98 | 1.73 45 |
Sparse-NonSparse [56] | 56.9 | 3.07 36 | 5.88 41 | 1.21 19 | 3.61 25 | 5.33 24 | 1.33 24 | 4.29 60 | 7.47 97 | 2.19 50 | 5.37 16 | 7.74 23 | 3.21 28 | 11.5 96 | 14.5 101 | 4.36 85 | 6.66 81 | 12.9 83 | 2.41 69 | 8.69 93 | 20.1 95 | 1.67 58 | 8.27 73 | 13.0 75 | 1.70 14 |
ProbFlowFields [128] | 57.8 | 3.15 63 | 6.32 82 | 1.21 19 | 3.53 14 | 5.26 19 | 1.29 13 | 5.03 92 | 7.35 92 | 3.73 103 | 5.43 30 | 7.97 41 | 3.25 43 | 11.1 43 | 14.0 51 | 4.50 108 | 6.48 56 | 12.6 59 | 2.55 108 | 7.99 53 | 18.4 59 | 2.57 118 | 7.78 25 | 12.2 25 | 1.75 72 |
MLDP_OF [89] | 58.4 | 3.08 39 | 5.98 57 | 1.21 19 | 4.01 67 | 6.01 70 | 1.49 60 | 3.67 13 | 6.14 39 | 1.47 9 | 5.78 64 | 8.13 52 | 3.95 113 | 11.3 71 | 14.2 71 | 3.87 62 | 6.71 85 | 13.0 85 | 2.51 99 | 7.73 32 | 17.7 33 | 1.71 65 | 8.18 56 | 12.8 57 | 1.76 83 |
FlowNetS+ft+v [112] | 58.6 | 3.07 36 | 5.81 34 | 1.28 99 | 4.57 107 | 6.29 95 | 2.41 128 | 4.01 34 | 5.64 11 | 2.13 48 | 5.55 41 | 7.77 28 | 3.88 111 | 11.3 71 | 14.2 71 | 4.46 96 | 5.99 13 | 11.5 11 | 2.35 43 | 8.63 90 | 20.0 93 | 1.62 47 | 7.70 19 | 12.0 16 | 1.74 64 |
PMF [73] | 58.9 | 3.14 59 | 6.13 68 | 1.20 6 | 3.73 40 | 5.60 41 | 1.27 6 | 5.24 97 | 8.98 118 | 3.76 104 | 5.75 61 | 8.56 78 | 3.28 49 | 10.8 22 | 13.6 25 | 3.62 28 | 6.55 64 | 12.7 65 | 2.35 43 | 8.41 81 | 19.5 86 | 1.64 51 | 8.57 104 | 13.4 103 | 1.70 14 |
S2F-IF [123] | 59.0 | 3.26 87 | 6.66 99 | 1.20 6 | 3.53 14 | 5.25 18 | 1.29 13 | 4.11 43 | 6.64 59 | 2.34 60 | 5.89 75 | 9.06 96 | 3.08 6 | 11.4 81 | 14.3 80 | 4.51 117 | 6.41 49 | 12.4 50 | 2.40 59 | 7.84 41 | 18.1 46 | 1.76 77 | 8.33 83 | 13.1 84 | 1.75 72 |
HAST [109] | 59.3 | 3.01 24 | 5.73 28 | 1.21 19 | 3.45 8 | 5.01 7 | 1.27 6 | 6.39 113 | 8.24 112 | 4.09 109 | 5.43 30 | 7.96 39 | 3.03 1 | 11.2 56 | 14.2 71 | 3.59 11 | 7.47 121 | 14.7 121 | 2.47 85 | 8.68 92 | 20.1 95 | 1.53 13 | 8.35 86 | 13.1 84 | 1.77 92 |
OFLAF [77] | 59.4 | 3.10 45 | 5.98 57 | 1.20 6 | 3.44 7 | 5.03 8 | 1.26 1 | 3.73 20 | 5.82 19 | 1.66 19 | 5.33 11 | 7.74 23 | 3.10 9 | 11.6 109 | 14.7 114 | 4.50 108 | 6.58 74 | 12.8 76 | 2.48 89 | 9.33 114 | 21.6 115 | 2.06 103 | 8.45 96 | 13.2 92 | 1.80 110 |
Sparse Occlusion [54] | 59.8 | 3.16 65 | 6.18 71 | 1.23 61 | 4.14 77 | 6.24 91 | 1.45 51 | 3.67 13 | 5.84 21 | 1.52 13 | 5.61 52 | 8.26 57 | 3.15 13 | 11.5 96 | 14.4 91 | 4.48 99 | 6.26 35 | 12.1 35 | 2.46 82 | 8.52 83 | 19.6 88 | 1.54 18 | 8.28 75 | 13.0 75 | 1.75 72 |
FlowFields+ [130] | 59.8 | 3.14 59 | 6.26 75 | 1.22 43 | 3.54 17 | 5.27 21 | 1.30 18 | 4.74 82 | 7.10 82 | 3.20 91 | 6.01 87 | 9.35 105 | 3.11 10 | 11.1 43 | 13.9 38 | 4.50 108 | 6.57 70 | 12.8 76 | 2.40 59 | 7.89 47 | 18.2 50 | 1.80 82 | 8.22 64 | 12.9 63 | 1.73 45 |
Filter Flow [19] | 60.3 | 3.13 57 | 5.90 43 | 1.28 99 | 4.56 106 | 6.38 102 | 1.85 104 | 4.22 54 | 6.28 43 | 2.10 46 | 5.91 76 | 7.97 41 | 3.44 84 | 10.4 12 | 13.1 14 | 3.69 44 | 6.43 53 | 12.5 55 | 2.40 59 | 8.17 64 | 18.8 68 | 1.62 47 | 7.94 42 | 12.4 39 | 1.78 96 |
TF+OM [100] | 60.6 | 3.33 98 | 6.83 104 | 1.25 81 | 3.65 30 | 5.43 32 | 1.47 56 | 3.82 25 | 6.43 48 | 1.68 22 | 6.01 87 | 9.04 95 | 3.19 23 | 11.2 56 | 14.1 59 | 4.38 87 | 6.46 55 | 12.5 55 | 2.34 37 | 8.30 74 | 19.2 76 | 1.86 85 | 8.05 49 | 12.6 49 | 1.75 72 |
Black & Anandan [4] | 63.0 | 3.22 80 | 5.87 39 | 1.30 107 | 4.82 120 | 6.55 110 | 1.78 100 | 7.16 117 | 7.10 82 | 3.93 106 | 6.25 99 | 8.49 72 | 3.35 73 | 10.9 29 | 13.7 30 | 3.56 5 | 6.33 40 | 12.2 38 | 2.37 50 | 8.23 68 | 18.6 63 | 1.64 51 | 7.67 15 | 11.9 10 | 1.69 8 |
EpicFlow [102] | 63.9 | 3.17 70 | 6.34 83 | 1.21 19 | 3.79 43 | 5.70 47 | 1.44 49 | 4.28 59 | 5.73 15 | 1.67 20 | 6.37 104 | 10.1 119 | 3.39 78 | 11.2 56 | 14.1 59 | 4.50 108 | 6.23 30 | 12.0 30 | 2.38 55 | 8.11 61 | 18.5 60 | 1.76 77 | 8.76 114 | 13.8 114 | 1.74 64 |
RNLOD-Flow [121] | 64.1 | 3.06 34 | 5.87 39 | 1.21 19 | 3.96 60 | 5.97 69 | 1.42 44 | 4.39 67 | 8.08 109 | 2.44 63 | 5.35 13 | 7.75 25 | 3.18 19 | 11.5 96 | 14.5 101 | 4.49 105 | 6.71 85 | 13.1 89 | 2.43 75 | 7.85 43 | 18.0 44 | 2.18 108 | 8.44 94 | 13.2 92 | 1.73 45 |
LSM [39] | 64.2 | 3.12 51 | 6.05 63 | 1.21 19 | 3.68 35 | 5.47 34 | 1.33 24 | 4.38 65 | 7.66 104 | 2.01 40 | 5.55 41 | 8.19 54 | 3.19 23 | 11.5 96 | 14.5 101 | 4.43 90 | 6.83 99 | 13.3 100 | 2.37 50 | 8.70 94 | 20.1 95 | 1.72 68 | 8.34 85 | 13.1 84 | 1.71 26 |
TCOF [69] | 64.3 | 3.12 51 | 5.94 50 | 1.21 19 | 4.60 110 | 6.64 117 | 1.76 98 | 4.13 44 | 7.30 88 | 1.81 28 | 5.42 29 | 7.88 33 | 3.25 43 | 11.3 71 | 14.2 71 | 3.63 34 | 6.42 51 | 12.4 50 | 2.36 46 | 9.08 111 | 21.0 111 | 1.59 35 | 8.37 87 | 13.1 84 | 1.76 83 |
Ramp [62] | 65.1 | 3.11 49 | 5.96 54 | 1.22 43 | 3.61 25 | 5.34 25 | 1.40 38 | 4.91 86 | 8.45 115 | 3.20 91 | 5.29 6 | 7.66 19 | 3.21 28 | 11.5 96 | 14.5 101 | 4.31 82 | 6.88 106 | 13.4 104 | 2.48 89 | 8.73 99 | 20.2 99 | 1.52 12 | 8.29 76 | 13.0 75 | 1.73 45 |
AggregFlow [97] | 65.7 | 3.80 120 | 8.08 121 | 1.23 61 | 3.87 50 | 5.83 58 | 1.43 47 | 4.21 53 | 6.79 62 | 2.85 78 | 6.11 91 | 9.36 106 | 3.31 64 | 10.6 16 | 13.3 16 | 3.67 40 | 6.13 21 | 11.8 18 | 2.34 37 | 8.70 94 | 19.8 90 | 2.30 113 | 8.27 73 | 13.0 75 | 1.75 72 |
Fusion [6] | 65.9 | 3.04 31 | 5.86 38 | 1.22 43 | 3.75 42 | 5.47 34 | 1.42 44 | 4.08 41 | 5.55 6 | 3.08 85 | 5.80 66 | 8.10 49 | 3.19 23 | 11.4 81 | 14.3 80 | 3.73 52 | 6.99 113 | 13.7 113 | 2.60 114 | 8.40 80 | 19.4 83 | 1.65 56 | 8.50 99 | 13.3 98 | 1.80 110 |
Classic+NL [31] | 66.0 | 3.10 45 | 5.92 48 | 1.23 61 | 3.66 32 | 5.40 29 | 1.39 35 | 4.78 84 | 8.42 114 | 3.01 83 | 5.36 14 | 7.78 29 | 3.30 59 | 11.5 96 | 14.5 101 | 4.24 79 | 6.73 88 | 13.1 89 | 2.40 59 | 8.74 100 | 20.2 99 | 1.70 63 | 8.29 76 | 13.0 75 | 1.71 26 |
ComponentFusion [96] | 66.9 | 3.41 102 | 7.08 110 | 1.20 6 | 3.63 27 | 5.44 33 | 1.27 6 | 4.20 52 | 6.49 55 | 2.43 62 | 5.59 49 | 8.38 65 | 3.32 67 | 11.4 81 | 14.4 91 | 4.11 73 | 6.26 35 | 12.1 35 | 2.35 43 | 9.30 113 | 21.6 115 | 2.80 123 | 8.68 108 | 13.6 109 | 1.73 45 |
Bartels [41] | 67.2 | 3.48 107 | 7.24 114 | 1.30 107 | 4.02 68 | 6.12 78 | 1.68 86 | 3.74 21 | 5.80 18 | 1.95 37 | 5.87 72 | 8.44 69 | 3.78 110 | 10.3 9 | 12.8 5 | 3.75 57 | 6.77 95 | 13.0 85 | 2.73 128 | 7.53 14 | 17.3 15 | 2.72 121 | 8.13 53 | 12.7 52 | 1.77 92 |
CNN-flow-warp+ref [117] | 67.5 | 2.90 6 | 5.43 8 | 1.25 81 | 4.10 71 | 5.95 63 | 1.83 103 | 4.92 87 | 7.63 103 | 2.45 64 | 6.13 93 | 7.85 31 | 3.72 107 | 11.3 71 | 14.2 71 | 4.51 117 | 6.03 15 | 11.6 14 | 2.46 82 | 9.00 107 | 20.8 110 | 1.65 56 | 7.91 38 | 12.4 39 | 1.76 83 |
Occlusion-TV-L1 [63] | 67.6 | 3.14 59 | 6.13 68 | 1.25 81 | 4.47 99 | 6.61 113 | 1.66 80 | 3.51 2 | 5.71 14 | 1.70 24 | 6.33 101 | 9.58 113 | 3.51 91 | 11.0 33 | 13.9 38 | 3.57 7 | 6.48 56 | 12.6 59 | 2.52 103 | 8.36 76 | 18.1 46 | 2.00 98 | 8.32 82 | 13.0 75 | 1.79 104 |
2D-CLG [1] | 67.8 | 3.01 24 | 5.65 20 | 1.28 99 | 4.59 109 | 6.17 81 | 1.95 115 | 5.18 93 | 6.06 32 | 3.15 90 | 6.01 87 | 7.88 33 | 3.97 114 | 11.4 81 | 14.4 91 | 4.69 128 | 5.98 12 | 11.5 11 | 2.45 79 | 8.89 105 | 20.5 103 | 1.67 58 | 7.74 20 | 12.0 16 | 1.71 26 |
HBM-GC [105] | 68.2 | 3.08 39 | 5.90 43 | 1.26 91 | 3.97 62 | 6.04 71 | 1.41 42 | 3.92 29 | 5.62 10 | 2.87 80 | 5.54 40 | 8.03 45 | 3.21 28 | 11.7 118 | 14.7 114 | 4.58 126 | 7.66 126 | 15.0 126 | 2.69 125 | 8.36 76 | 19.3 79 | 1.55 23 | 7.86 32 | 12.3 29 | 1.76 83 |
Horn & Schunck [3] | 68.6 | 3.16 65 | 5.83 35 | 1.26 91 | 4.91 121 | 6.65 118 | 1.92 110 | 6.13 109 | 6.85 66 | 3.53 99 | 6.80 112 | 9.10 97 | 3.57 97 | 10.9 29 | 13.7 30 | 3.59 11 | 6.16 26 | 11.9 26 | 2.32 32 | 8.63 90 | 19.5 86 | 1.84 84 | 7.91 38 | 12.3 29 | 1.73 45 |
Efficient-NL [60] | 69.2 | 3.05 32 | 5.77 32 | 1.21 19 | 3.90 57 | 5.84 59 | 1.38 33 | 5.90 106 | 6.94 72 | 4.19 111 | 5.59 49 | 8.09 48 | 3.20 26 | 11.5 96 | 14.4 91 | 4.40 88 | 6.87 103 | 13.4 104 | 2.40 59 | 8.85 102 | 20.5 103 | 1.68 61 | 8.57 104 | 13.4 103 | 1.66 2 |
Classic+CPF [83] | 69.3 | 3.12 51 | 5.96 54 | 1.21 19 | 3.72 39 | 5.51 39 | 1.39 35 | 4.39 67 | 7.38 95 | 2.27 55 | 5.32 9 | 7.70 22 | 3.18 19 | 11.7 118 | 14.8 120 | 4.50 108 | 7.18 118 | 14.0 118 | 2.45 79 | 8.79 101 | 20.2 99 | 1.57 31 | 8.71 112 | 13.7 111 | 1.73 45 |
RFlow [90] | 70.1 | 3.08 39 | 5.99 59 | 1.23 61 | 4.33 86 | 6.31 97 | 1.66 80 | 4.83 85 | 7.32 89 | 3.14 88 | 5.87 72 | 8.72 83 | 3.47 87 | 11.1 43 | 14.0 51 | 3.60 18 | 6.54 63 | 12.7 65 | 2.39 57 | 8.54 84 | 19.8 90 | 1.61 40 | 8.26 72 | 12.9 63 | 1.80 110 |
FC-2Layers-FF [74] | 70.7 | 3.18 74 | 6.16 70 | 1.22 43 | 3.33 4 | 4.73 3 | 1.35 30 | 4.34 64 | 7.09 81 | 3.11 87 | 5.56 43 | 8.29 58 | 3.29 51 | 11.5 96 | 14.5 101 | 4.48 99 | 7.00 114 | 13.7 113 | 2.48 89 | 8.92 106 | 20.6 106 | 1.71 65 | 8.30 79 | 13.0 75 | 1.73 45 |
SRR-TVOF-NL [91] | 70.7 | 3.32 97 | 6.46 92 | 1.23 61 | 3.96 60 | 5.96 66 | 1.59 67 | 4.68 77 | 7.90 108 | 3.52 98 | 5.99 85 | 8.77 84 | 3.23 38 | 11.2 56 | 14.1 59 | 4.45 95 | 6.79 97 | 13.2 97 | 2.31 24 | 7.88 45 | 18.0 44 | 1.50 4 | 8.37 87 | 13.1 84 | 1.75 72 |
FESL [72] | 71.1 | 3.16 65 | 6.02 61 | 1.21 19 | 3.65 30 | 5.42 31 | 1.35 30 | 4.39 67 | 7.61 102 | 2.18 49 | 5.71 58 | 8.35 63 | 3.30 59 | 11.6 109 | 14.7 114 | 4.51 117 | 6.73 88 | 13.1 89 | 2.47 85 | 8.70 94 | 20.1 95 | 1.56 28 | 8.42 90 | 13.2 92 | 1.75 72 |
TriFlow [95] | 71.8 | 3.71 119 | 7.95 120 | 1.25 81 | 4.31 85 | 6.36 101 | 1.71 91 | 4.05 37 | 6.86 68 | 1.84 32 | 6.21 98 | 9.44 108 | 3.17 16 | 11.3 71 | 14.2 71 | 4.48 99 | 6.76 94 | 13.1 89 | 2.29 16 | 8.01 58 | 18.2 50 | 1.75 74 | 8.24 68 | 12.9 63 | 1.70 14 |
OFH [38] | 72.4 | 3.18 74 | 6.29 76 | 1.23 61 | 4.11 73 | 5.96 66 | 1.61 72 | 4.68 77 | 8.40 113 | 1.68 22 | 5.84 68 | 8.99 93 | 3.03 1 | 11.3 71 | 14.2 71 | 4.25 81 | 6.30 39 | 12.2 38 | 2.40 59 | 8.59 86 | 19.3 79 | 1.89 88 | 8.55 101 | 13.4 103 | 1.97 125 |
Nguyen [33] | 73.1 | 3.26 87 | 6.11 67 | 1.33 114 | 4.94 122 | 6.51 108 | 1.91 109 | 4.09 42 | 7.32 89 | 1.96 38 | 6.19 97 | 8.53 75 | 3.60 100 | 11.1 43 | 13.9 38 | 3.58 9 | 6.55 64 | 12.7 65 | 2.36 46 | 9.44 116 | 21.8 119 | 1.80 82 | 7.86 32 | 12.3 29 | 1.74 64 |
CostFilter [40] | 73.8 | 3.46 106 | 7.24 114 | 1.19 2 | 3.71 37 | 5.60 41 | 1.27 6 | 5.63 102 | 9.41 123 | 3.86 105 | 6.37 104 | 10.1 119 | 3.23 38 | 11.2 56 | 14.0 51 | 3.78 59 | 6.35 45 | 12.2 38 | 2.40 59 | 8.86 104 | 20.6 106 | 1.69 62 | 8.80 115 | 13.8 114 | 1.74 64 |
S2D-Matching [84] | 73.8 | 3.21 78 | 6.22 73 | 1.22 43 | 3.97 62 | 5.95 63 | 1.48 57 | 4.57 74 | 7.70 105 | 2.84 77 | 5.48 33 | 8.06 47 | 3.48 89 | 11.4 81 | 14.3 80 | 4.14 75 | 6.97 111 | 13.6 111 | 2.56 112 | 8.09 60 | 18.6 63 | 1.74 70 | 8.21 62 | 12.9 63 | 1.76 83 |
SVFilterOh [111] | 74.2 | 3.23 81 | 6.35 84 | 1.23 61 | 3.53 14 | 5.19 15 | 1.31 19 | 5.91 107 | 8.20 111 | 4.22 112 | 5.75 61 | 8.52 74 | 3.43 82 | 11.4 81 | 14.3 80 | 4.53 123 | 6.97 111 | 13.6 111 | 2.38 55 | 7.94 49 | 18.3 58 | 1.57 31 | 8.31 81 | 13.0 75 | 1.79 104 |
Steered-L1 [118] | 74.2 | 2.97 17 | 5.73 28 | 1.21 19 | 3.81 44 | 5.72 50 | 1.60 70 | 8.15 120 | 9.24 120 | 6.46 127 | 6.42 106 | 9.21 103 | 4.28 121 | 11.4 81 | 14.3 80 | 3.80 60 | 6.52 61 | 12.7 65 | 2.43 75 | 8.20 65 | 19.0 72 | 2.54 116 | 8.33 83 | 13.1 84 | 1.70 14 |
FlowNet2 [122] | 75.0 | 4.84 127 | 10.1 128 | 1.29 104 | 4.11 73 | 6.13 79 | 1.61 72 | 4.73 80 | 7.06 80 | 2.36 61 | 6.36 103 | 10.0 117 | 3.38 75 | 11.2 56 | 14.1 59 | 3.71 48 | 6.44 54 | 12.5 55 | 2.33 35 | 8.45 82 | 19.4 83 | 1.61 40 | 8.03 48 | 12.6 49 | 1.77 92 |
IAOF [50] | 75.1 | 3.53 111 | 6.60 98 | 1.32 112 | 5.39 129 | 7.19 129 | 1.96 116 | 5.81 104 | 7.32 89 | 3.63 101 | 6.15 95 | 8.34 62 | 3.72 107 | 11.1 43 | 14.0 51 | 3.60 18 | 6.50 59 | 12.6 59 | 2.34 37 | 8.28 72 | 19.0 72 | 1.53 13 | 7.94 42 | 12.4 39 | 1.73 45 |
Adaptive [20] | 75.3 | 3.24 84 | 6.44 89 | 1.25 81 | 4.57 107 | 6.61 113 | 1.72 92 | 3.94 31 | 6.12 38 | 1.81 28 | 5.86 70 | 8.66 81 | 3.47 87 | 11.6 109 | 14.6 109 | 3.59 11 | 6.55 64 | 12.7 65 | 2.51 99 | 9.03 108 | 20.6 106 | 1.59 35 | 8.13 53 | 12.7 52 | 1.78 96 |
Complementary OF [21] | 76.9 | 3.48 107 | 7.32 118 | 1.20 6 | 3.89 55 | 5.96 66 | 1.45 51 | 8.94 124 | 6.94 72 | 5.45 122 | 6.33 101 | 10.0 117 | 3.09 8 | 11.3 71 | 14.2 71 | 4.24 79 | 6.33 40 | 12.3 45 | 2.42 72 | 8.62 89 | 19.3 79 | 1.75 74 | 9.07 121 | 14.3 122 | 1.72 36 |
TV-L1-improved [17] | 77.5 | 3.09 43 | 6.03 62 | 1.25 81 | 4.55 105 | 6.59 112 | 1.70 89 | 5.88 105 | 5.66 12 | 4.09 109 | 5.53 38 | 7.88 33 | 3.22 32 | 11.4 81 | 14.4 91 | 3.61 24 | 6.73 88 | 13.1 89 | 2.51 99 | 9.48 117 | 22.1 121 | 1.94 93 | 8.25 70 | 12.9 63 | 1.79 104 |
TI-DOFE [24] | 78.2 | 3.41 102 | 6.44 89 | 1.44 124 | 5.20 127 | 6.82 126 | 2.01 120 | 4.19 51 | 6.41 47 | 1.88 34 | 6.98 116 | 9.50 110 | 3.70 105 | 10.8 22 | 13.6 25 | 3.61 24 | 6.59 75 | 12.8 76 | 2.36 46 | 8.13 62 | 18.2 50 | 1.77 79 | 8.53 100 | 12.4 39 | 2.33 129 |
EPPM w/o HM [88] | 79.4 | 3.35 100 | 6.86 106 | 1.21 19 | 3.85 49 | 5.88 62 | 1.29 13 | 7.03 115 | 9.47 125 | 3.97 108 | 6.15 95 | 9.51 111 | 3.38 75 | 10.6 16 | 13.3 16 | 3.62 28 | 7.00 114 | 13.7 113 | 2.37 50 | 8.85 102 | 20.5 103 | 2.62 120 | 8.42 90 | 13.2 92 | 1.76 83 |
BriefMatch [124] | 80.3 | 3.25 86 | 6.49 93 | 1.25 81 | 3.87 50 | 5.67 44 | 1.97 118 | 6.16 110 | 6.17 41 | 4.79 118 | 6.83 114 | 8.37 64 | 5.73 128 | 11.0 33 | 13.8 33 | 3.73 52 | 6.75 93 | 13.0 85 | 2.61 116 | 7.99 53 | 17.9 39 | 3.29 128 | 8.22 64 | 12.8 57 | 2.32 128 |
GraphCuts [14] | 80.8 | 3.65 118 | 7.01 109 | 1.27 95 | 3.89 55 | 5.71 49 | 1.59 67 | 7.54 118 | 5.84 21 | 4.31 115 | 5.98 83 | 8.42 67 | 3.45 86 | 11.4 81 | 14.4 91 | 4.09 71 | 6.56 67 | 12.8 76 | 2.30 21 | 8.70 94 | 20.2 99 | 1.98 95 | 8.59 107 | 13.5 108 | 1.73 45 |
NL-TV-NCC [25] | 82.3 | 3.37 101 | 6.58 97 | 1.24 72 | 4.23 80 | 6.41 104 | 1.49 60 | 4.39 67 | 6.68 60 | 2.07 43 | 7.19 121 | 11.2 126 | 3.35 73 | 10.7 19 | 13.4 18 | 4.00 68 | 6.95 108 | 13.4 104 | 2.44 77 | 9.06 109 | 20.0 93 | 2.13 106 | 8.42 90 | 13.1 84 | 1.78 96 |
IAOF2 [51] | 83.8 | 3.43 104 | 6.70 100 | 1.28 99 | 4.62 112 | 6.77 124 | 1.74 94 | 4.41 71 | 6.89 69 | 2.12 47 | 5.97 81 | 8.53 75 | 3.33 70 | 11.6 109 | 14.7 114 | 4.06 70 | 6.87 103 | 13.4 104 | 2.51 99 | 8.26 69 | 18.7 67 | 1.61 40 | 8.22 64 | 12.9 63 | 1.74 64 |
TriangleFlow [30] | 84.1 | 3.24 84 | 6.31 80 | 1.26 91 | 4.29 84 | 6.29 95 | 1.66 80 | 4.67 76 | 6.85 66 | 2.48 65 | 5.78 64 | 8.47 70 | 3.30 59 | 11.4 81 | 14.4 91 | 3.47 3 | 6.63 80 | 12.8 76 | 2.37 50 | 9.67 121 | 22.5 122 | 2.08 104 | 9.69 127 | 15.2 127 | 1.90 122 |
LocallyOriented [52] | 85.5 | 3.29 91 | 6.53 95 | 1.26 91 | 4.64 113 | 6.69 120 | 1.74 94 | 5.61 101 | 7.56 100 | 3.67 102 | 6.73 110 | 9.84 116 | 3.18 19 | 11.5 96 | 14.4 91 | 3.71 48 | 6.57 70 | 12.7 65 | 2.45 79 | 8.71 98 | 19.3 79 | 1.71 65 | 8.40 89 | 13.1 84 | 1.72 36 |
Correlation Flow [75] | 86.8 | 3.27 89 | 6.50 94 | 1.20 6 | 4.42 93 | 6.56 111 | 1.65 77 | 3.98 33 | 6.10 36 | 2.30 59 | 5.93 78 | 8.94 90 | 3.32 67 | 11.6 109 | 14.6 109 | 3.84 61 | 7.63 125 | 14.8 123 | 2.65 123 | 9.95 125 | 23.0 125 | 2.01 100 | 8.73 113 | 13.7 111 | 1.71 26 |
ROF-ND [107] | 87.0 | 3.18 74 | 5.83 35 | 1.21 19 | 4.13 76 | 6.13 79 | 1.92 110 | 4.22 54 | 7.51 98 | 2.22 52 | 7.10 117 | 10.8 121 | 3.53 94 | 11.4 81 | 14.3 80 | 4.48 99 | 6.95 108 | 13.5 108 | 2.53 104 | 8.21 67 | 18.6 63 | 1.90 89 | 9.08 122 | 14.2 121 | 1.81 117 |
StereoOF-V1MT [119] | 87.9 | 3.56 112 | 7.20 113 | 1.22 43 | 4.27 83 | 6.18 84 | 1.70 89 | 6.10 108 | 6.80 63 | 3.43 97 | 7.17 120 | 9.52 112 | 4.01 117 | 11.2 56 | 14.1 59 | 4.43 90 | 6.61 79 | 12.5 55 | 2.60 114 | 9.49 118 | 21.6 115 | 2.05 101 | 8.01 47 | 12.4 39 | 1.78 96 |
HBpMotionGpu [43] | 88.1 | 3.63 116 | 7.28 116 | 1.35 118 | 4.78 118 | 6.69 120 | 1.92 110 | 4.33 63 | 7.01 78 | 2.56 69 | 6.46 107 | 9.81 115 | 3.40 79 | 11.5 96 | 14.4 91 | 5.69 131 | 6.83 99 | 13.3 100 | 2.55 108 | 7.40 9 | 16.9 7 | 1.51 7 | 8.30 79 | 13.0 75 | 1.79 104 |
ACK-Prior [27] | 88.3 | 3.30 92 | 6.56 96 | 1.21 19 | 3.81 44 | 5.78 52 | 1.42 44 | 7.13 116 | 6.90 71 | 5.04 119 | 6.02 90 | 8.78 85 | 3.70 105 | 11.7 118 | 14.7 114 | 4.57 125 | 6.95 108 | 13.5 108 | 2.50 95 | 8.36 76 | 19.2 76 | 2.53 115 | 8.56 103 | 13.4 103 | 1.73 45 |
Aniso-Texture [82] | 90.0 | 3.11 49 | 6.09 65 | 1.21 19 | 4.51 100 | 6.62 115 | 1.75 96 | 4.77 83 | 6.43 48 | 2.08 45 | 7.44 123 | 10.9 122 | 4.80 123 | 11.6 109 | 14.6 109 | 4.51 117 | 7.49 122 | 14.7 121 | 2.71 127 | 8.28 72 | 19.1 75 | 1.61 40 | 8.68 108 | 13.6 109 | 1.74 64 |
Dynamic MRF [7] | 91.7 | 3.19 77 | 6.41 87 | 1.22 43 | 4.11 73 | 6.21 86 | 1.56 64 | 5.37 98 | 7.35 92 | 2.70 72 | 6.74 111 | 9.18 101 | 4.19 119 | 11.1 43 | 13.9 38 | 4.48 99 | 7.02 117 | 13.7 113 | 2.62 119 | 9.26 112 | 21.4 114 | 2.23 110 | 8.57 104 | 13.3 98 | 1.80 110 |
FOLKI [16] | 91.9 | 3.64 117 | 7.12 111 | 1.65 127 | 5.22 128 | 6.72 123 | 2.36 126 | 5.20 94 | 8.08 109 | 3.96 107 | 7.93 125 | 9.33 104 | 5.52 127 | 11.2 56 | 14.0 51 | 3.70 45 | 6.56 67 | 12.6 59 | 2.74 129 | 8.00 56 | 18.2 50 | 2.88 124 | 7.96 45 | 12.3 29 | 1.78 96 |
Shiralkar [42] | 92.9 | 3.57 113 | 7.31 117 | 1.22 43 | 4.46 98 | 6.33 100 | 1.65 77 | 5.49 99 | 6.98 75 | 2.73 73 | 7.42 122 | 10.9 122 | 3.43 82 | 11.5 96 | 14.4 91 | 3.73 52 | 6.57 70 | 12.7 65 | 2.48 89 | 9.58 119 | 21.9 120 | 1.88 87 | 9.18 124 | 14.4 123 | 1.75 72 |
SILK [79] | 93.2 | 3.45 105 | 6.85 105 | 1.36 119 | 5.11 125 | 6.70 122 | 2.21 124 | 11.1 128 | 9.96 126 | 6.24 126 | 6.49 108 | 8.82 86 | 3.59 99 | 11.4 81 | 14.3 80 | 3.54 4 | 6.87 103 | 13.3 100 | 2.63 120 | 7.76 35 | 17.7 33 | 1.87 86 | 8.20 60 | 12.7 52 | 1.80 110 |
SimpleFlow [49] | 93.6 | 3.10 45 | 5.97 56 | 1.22 43 | 4.19 79 | 6.11 77 | 1.64 76 | 9.91 127 | 9.43 124 | 6.53 128 | 5.58 46 | 8.29 58 | 3.30 59 | 11.6 109 | 14.6 109 | 4.43 90 | 7.42 120 | 14.6 120 | 2.56 112 | 10.7 128 | 25.2 128 | 2.73 122 | 9.16 123 | 14.4 123 | 1.73 45 |
Learning Flow [11] | 94.3 | 3.14 59 | 6.09 65 | 1.27 95 | 4.51 100 | 6.53 109 | 1.67 83 | 11.5 131 | 12.9 131 | 7.17 131 | 6.31 100 | 8.30 60 | 3.66 103 | 11.7 118 | 14.8 120 | 3.89 63 | 6.59 75 | 12.8 76 | 2.48 89 | 8.27 71 | 18.9 69 | 1.96 94 | 8.68 108 | 13.4 103 | 1.80 110 |
Rannacher [23] | 94.4 | 3.31 94 | 6.72 102 | 1.25 81 | 4.60 110 | 6.66 119 | 1.72 92 | 6.36 112 | 6.54 56 | 4.25 113 | 5.91 76 | 8.87 87 | 3.49 90 | 11.5 96 | 14.5 101 | 3.63 34 | 6.73 88 | 13.1 89 | 2.53 104 | 9.35 115 | 21.7 118 | 1.98 95 | 8.70 111 | 13.7 111 | 1.75 72 |
Adaptive flow [45] | 96.3 | 3.60 115 | 6.30 78 | 1.54 126 | 5.14 126 | 6.79 125 | 2.14 123 | 4.52 73 | 6.60 57 | 3.01 83 | 6.54 109 | 8.64 80 | 4.23 120 | 12.1 127 | 15.2 125 | 4.09 71 | 7.57 123 | 14.9 125 | 2.64 122 | 7.75 33 | 17.8 36 | 2.28 112 | 8.47 98 | 13.3 98 | 1.71 26 |
UnFlow [129] | 97.0 | 4.05 122 | 8.73 125 | 1.31 110 | 4.44 97 | 6.28 94 | 1.87 105 | 4.92 87 | 7.36 94 | 2.62 70 | 5.95 80 | 9.00 94 | 3.27 48 | 12.0 125 | 15.2 125 | 4.37 86 | 7.59 124 | 14.8 123 | 2.61 116 | 7.77 37 | 17.6 29 | 1.64 51 | 10.4 128 | 15.4 128 | 2.33 129 |
StereoFlow [44] | 97.9 | 5.35 131 | 10.3 130 | 1.42 123 | 5.03 124 | 7.21 130 | 1.76 98 | 4.14 45 | 6.94 72 | 2.01 40 | 5.83 67 | 8.55 77 | 3.33 70 | 13.7 129 | 17.3 129 | 4.70 129 | 8.71 130 | 17.2 130 | 2.70 126 | 7.88 45 | 18.1 46 | 1.61 40 | 8.82 117 | 13.9 118 | 1.79 104 |
2bit-BM-tele [98] | 99.1 | 3.31 94 | 6.41 87 | 1.34 117 | 4.53 104 | 6.62 115 | 1.80 101 | 6.23 111 | 9.24 120 | 6.19 125 | 5.94 79 | 8.59 79 | 3.55 95 | 11.3 71 | 14.2 71 | 4.03 69 | 7.72 127 | 15.1 127 | 3.02 130 | 12.2 131 | 28.7 131 | 4.77 131 | 7.76 24 | 12.1 21 | 1.82 119 |
FFV1MT [106] | 103.0 | 4.09 124 | 8.38 122 | 1.31 110 | 4.68 115 | 6.18 84 | 2.02 121 | 6.95 114 | 11.5 128 | 3.35 95 | 7.12 118 | 9.16 99 | 3.98 115 | 11.3 71 | 14.1 59 | 3.74 55 | 6.77 95 | 12.7 65 | 2.50 95 | 9.59 120 | 21.0 111 | 2.05 101 | 8.87 118 | 13.8 114 | 1.90 122 |
SPSA-learn [13] | 103.6 | 3.89 121 | 7.79 119 | 1.27 95 | 4.43 95 | 6.17 81 | 1.81 102 | 9.03 125 | 8.47 116 | 5.47 123 | 6.80 112 | 9.40 107 | 3.72 107 | 11.5 96 | 14.5 101 | 3.91 65 | 6.51 60 | 12.6 59 | 2.46 82 | 11.9 129 | 27.9 130 | 4.54 130 | 10.5 130 | 16.5 130 | 1.75 72 |
SegOF [10] | 106.2 | 3.51 109 | 7.12 111 | 1.32 112 | 4.17 78 | 6.10 75 | 1.59 67 | 8.69 122 | 7.75 106 | 5.15 120 | 8.58 128 | 14.3 129 | 4.29 122 | 11.7 118 | 14.8 120 | 4.50 108 | 6.79 97 | 13.2 97 | 2.50 95 | 10.1 126 | 23.5 126 | 2.55 117 | 8.80 115 | 13.8 114 | 1.72 36 |
PGAM+LK [55] | 107.2 | 4.08 123 | 8.41 123 | 1.65 127 | 4.74 117 | 6.45 105 | 2.27 125 | 8.87 123 | 12.2 129 | 6.88 129 | 8.06 127 | 10.9 122 | 4.83 124 | 11.4 81 | 14.3 80 | 3.90 64 | 6.83 99 | 13.2 97 | 2.55 108 | 8.26 69 | 18.9 69 | 2.27 111 | 8.55 101 | 13.3 98 | 1.90 122 |
Heeger++ [104] | 107.5 | 4.76 126 | 9.63 127 | 1.33 114 | 4.65 114 | 6.22 89 | 1.90 107 | 7.84 119 | 9.26 122 | 3.57 100 | 7.12 118 | 9.16 99 | 3.98 115 | 11.9 124 | 15.0 124 | 4.47 97 | 6.52 61 | 12.2 38 | 2.61 116 | 9.82 122 | 20.6 106 | 2.00 98 | 9.02 120 | 14.0 119 | 1.79 104 |
SLK [47] | 109.3 | 3.51 109 | 6.96 108 | 1.41 121 | 4.72 116 | 6.10 75 | 1.98 119 | 9.84 126 | 7.59 101 | 5.20 121 | 7.98 126 | 11.0 125 | 6.14 129 | 11.8 123 | 14.9 123 | 3.71 48 | 6.60 77 | 12.7 65 | 2.50 95 | 9.87 124 | 22.8 124 | 2.08 104 | 8.94 119 | 14.0 119 | 2.03 126 |
HCIC-L [99] | 110.2 | 4.98 129 | 9.28 126 | 1.77 130 | 4.97 123 | 6.87 128 | 2.11 122 | 5.70 103 | 10.0 127 | 4.41 117 | 7.85 124 | 11.8 127 | 3.68 104 | 10.9 29 | 13.7 30 | 3.72 51 | 8.18 129 | 16.1 129 | 2.55 108 | 9.06 109 | 21.0 111 | 2.58 119 | 9.57 126 | 15.0 126 | 1.81 117 |
Pyramid LK [2] | 119.0 | 4.16 125 | 8.44 124 | 1.74 129 | 5.83 130 | 6.82 126 | 2.76 130 | 11.4 129 | 8.60 117 | 5.89 124 | 12.4 131 | 16.7 130 | 7.03 131 | 14.3 130 | 18.1 130 | 3.92 67 | 6.69 84 | 12.2 38 | 2.63 120 | 10.3 127 | 24.0 127 | 2.45 114 | 11.1 131 | 17.4 131 | 2.55 131 |
GroupFlow [9] | 120.7 | 4.94 128 | 10.2 129 | 1.36 119 | 4.51 100 | 6.50 107 | 1.92 110 | 8.67 121 | 9.13 119 | 4.38 116 | 8.83 129 | 13.0 128 | 5.40 126 | 12.9 128 | 16.3 128 | 4.53 123 | 7.89 128 | 15.5 128 | 2.65 123 | 9.85 123 | 22.6 123 | 1.91 90 | 9.52 125 | 14.9 125 | 1.88 121 |
Periodicity [78] | 130.0 | 5.27 130 | 11.1 131 | 1.83 131 | 7.09 131 | 7.33 131 | 2.86 131 | 11.4 129 | 12.2 129 | 7.13 130 | 10.5 130 | 17.1 131 | 6.14 129 | 14.9 131 | 19.0 131 | 4.71 130 | 9.13 131 | 17.9 131 | 3.16 131 | 11.9 129 | 27.8 129 | 3.76 129 | 10.4 128 | 15.8 129 | 2.29 127 |
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. |