Optical flow evaluation results |
Statistics:
Average
SD
R2.5
R5.0
R10.0
A90
A95
A99
Error type: endpoint angle interpolation normalized interpolation |
SD 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] | 22.0 | 8.94 31 | 13.9 29 | 1.56 1 | 7.36 7 | 9.57 9 | 2.68 3 | 15.9 34 | 20.5 74 | 16.3 74 | 13.4 15 | 18.2 29 | 8.87 29 | 24.1 7 | 27.3 7 | 12.1 17 | 17.6 29 | 25.0 34 | 6.01 21 | 22.8 11 | 34.9 12 | 7.11 4 | 20.9 17 | 26.2 17 | 7.78 16 |
CBF [12] | 22.3 | 8.02 2 | 12.4 2 | 1.75 38 | 8.18 49 | 10.3 32 | 4.84 74 | 13.2 5 | 15.5 14 | 9.17 18 | 11.5 1 | 15.0 2 | 7.94 18 | 22.8 1 | 25.8 1 | 12.2 28 | 16.2 9 | 22.9 10 | 6.24 38 | 23.4 21 | 35.8 21 | 8.06 50 | 21.2 22 | 26.7 24 | 8.24 56 |
PMMST [114] | 24.2 | 8.93 27 | 14.0 36 | 1.57 2 | 7.21 3 | 9.34 4 | 2.66 1 | 14.1 17 | 16.2 25 | 11.0 35 | 15.6 76 | 21.2 87 | 14.7 114 | 24.1 7 | 27.3 7 | 12.1 17 | 16.3 11 | 23.0 11 | 5.91 12 | 22.6 9 | 34.5 9 | 7.52 19 | 19.9 4 | 25.0 4 | 8.18 44 |
SepConv-v1 [127] | 26.6 | 8.19 4 | 12.5 3 | 5.08 127 | 7.89 25 | 9.09 3 | 8.08 125 | 20.8 88 | 12.1 1 | 18.4 89 | 12.5 7 | 14.8 1 | 11.7 84 | 23.5 4 | 26.6 4 | 9.13 2 | 14.1 2 | 19.9 2 | 5.12 2 | 21.1 1 | 32.2 1 | 8.37 56 | 17.2 1 | 21.5 1 | 6.71 5 |
DeepFlow [86] | 27.5 | 8.76 16 | 13.7 19 | 1.59 3 | 8.08 43 | 10.4 42 | 4.72 65 | 13.8 12 | 18.1 47 | 7.83 11 | 12.1 2 | 15.2 4 | 8.41 23 | 28.8 60 | 32.7 60 | 12.1 17 | 17.0 22 | 24.1 24 | 5.99 19 | 21.6 3 | 33.0 4 | 7.45 12 | 23.0 66 | 28.9 67 | 7.80 18 |
DeepFlow2 [108] | 31.3 | 8.59 11 | 13.4 12 | 1.64 8 | 8.06 41 | 10.4 42 | 4.45 57 | 13.8 12 | 18.6 51 | 8.12 13 | 12.4 4 | 16.1 6 | 10.6 51 | 28.5 54 | 32.3 53 | 12.3 39 | 16.7 14 | 23.6 14 | 5.92 14 | 23.0 15 | 35.0 13 | 7.49 15 | 22.6 53 | 28.4 54 | 8.47 105 |
SuperFlow [81] | 31.8 | 8.83 20 | 13.8 21 | 1.68 12 | 8.19 50 | 10.3 32 | 5.04 80 | 16.0 36 | 16.6 28 | 10.9 33 | 14.9 55 | 15.1 3 | 7.67 12 | 23.0 3 | 26.0 2 | 12.5 45 | 18.0 41 | 25.5 43 | 6.55 62 | 23.6 25 | 35.8 21 | 9.80 82 | 20.7 12 | 26.0 13 | 8.12 33 |
CLG-TV [48] | 34.8 | 8.34 8 | 12.9 7 | 1.98 84 | 8.74 71 | 10.8 63 | 4.75 70 | 14.0 15 | 16.0 20 | 9.23 19 | 12.4 4 | 16.1 6 | 9.95 42 | 29.7 83 | 33.7 83 | 12.0 14 | 16.8 15 | 23.9 18 | 5.46 3 | 22.2 5 | 32.9 3 | 8.02 46 | 21.8 38 | 27.4 39 | 8.33 79 |
SIOF [67] | 35.5 | 8.78 18 | 13.5 14 | 1.80 56 | 8.97 82 | 11.2 88 | 4.51 59 | 16.7 49 | 23.2 95 | 11.6 39 | 13.2 11 | 17.7 20 | 9.61 38 | 23.7 6 | 26.8 6 | 11.8 8 | 17.8 34 | 25.2 37 | 5.98 18 | 23.4 21 | 35.9 24 | 7.33 7 | 22.1 41 | 27.8 43 | 8.15 37 |
AdaConv-v1 [126] | 36.2 | 9.51 65 | 14.1 43 | 4.99 126 | 9.04 85 | 9.51 8 | 9.70 127 | 18.8 76 | 13.8 3 | 18.3 88 | 14.5 44 | 16.5 11 | 15.2 118 | 25.9 19 | 29.4 19 | 7.81 1 | 13.1 1 | 18.4 1 | 5.63 6 | 21.4 2 | 32.7 2 | 7.45 12 | 17.4 2 | 21.8 2 | 6.73 7 |
Aniso. Huber-L1 [22] | 36.9 | 8.22 5 | 12.7 5 | 1.84 66 | 9.12 94 | 11.1 83 | 5.11 82 | 13.6 9 | 16.3 27 | 7.58 9 | 12.2 3 | 16.1 6 | 9.16 32 | 29.8 89 | 33.8 88 | 12.7 50 | 16.9 18 | 23.9 18 | 5.57 5 | 23.2 18 | 35.5 18 | 7.30 6 | 21.8 38 | 27.4 39 | 8.32 78 |
CombBMOF [113] | 38.1 | 9.74 80 | 14.3 51 | 3.85 122 | 7.82 20 | 10.2 24 | 3.81 35 | 16.2 38 | 19.1 59 | 12.8 53 | 13.8 24 | 18.5 34 | 10.2 47 | 26.5 20 | 30.0 20 | 12.2 28 | 17.8 34 | 25.2 37 | 6.09 29 | 23.1 16 | 35.2 16 | 7.64 26 | 21.3 25 | 26.7 24 | 8.21 53 |
MDP-Flow [26] | 38.3 | 8.27 6 | 12.8 6 | 1.74 33 | 7.26 5 | 9.42 5 | 3.90 38 | 17.2 57 | 16.1 24 | 15.0 69 | 13.6 19 | 18.0 25 | 10.9 58 | 28.8 60 | 32.7 60 | 15.3 94 | 17.9 40 | 25.2 37 | 7.36 89 | 23.6 25 | 36.1 26 | 12.2 101 | 20.6 9 | 25.9 10 | 8.07 23 |
LME [70] | 38.8 | 8.97 33 | 14.0 36 | 1.62 5 | 8.07 42 | 10.5 49 | 3.69 28 | 16.9 52 | 17.7 40 | 9.29 21 | 14.5 44 | 19.6 57 | 9.68 39 | 29.2 77 | 33.1 77 | 15.3 94 | 18.1 43 | 25.7 45 | 6.15 31 | 22.7 10 | 34.7 10 | 7.37 9 | 21.0 19 | 26.4 20 | 8.20 49 |
NN-field [71] | 39.0 | 9.03 39 | 14.1 43 | 1.74 33 | 7.01 2 | 9.05 2 | 2.74 8 | 18.3 71 | 19.1 59 | 12.6 51 | 16.8 101 | 22.7 106 | 15.8 120 | 24.2 9 | 27.5 9 | 12.1 17 | 17.8 34 | 25.1 35 | 6.07 26 | 23.1 16 | 35.4 17 | 7.69 30 | 20.6 9 | 25.9 10 | 8.36 89 |
WLIF-Flow [93] | 39.2 | 8.64 12 | 13.4 12 | 1.69 17 | 7.89 25 | 10.2 24 | 3.94 39 | 17.0 53 | 22.0 84 | 14.5 65 | 13.7 22 | 18.4 32 | 11.5 75 | 26.7 23 | 30.3 23 | 12.3 39 | 19.8 96 | 28.0 97 | 8.12 112 | 22.4 7 | 34.2 7 | 7.58 23 | 21.1 21 | 26.4 20 | 7.62 14 |
NNF-Local [87] | 39.5 | 8.84 22 | 13.8 21 | 1.61 4 | 7.25 4 | 9.44 6 | 2.76 9 | 14.6 23 | 19.3 64 | 14.5 65 | 16.0 89 | 21.6 94 | 15.8 120 | 24.2 9 | 27.5 9 | 12.2 28 | 18.4 51 | 26.0 53 | 6.42 54 | 24.2 32 | 37.1 36 | 9.54 75 | 20.3 7 | 25.4 7 | 8.27 67 |
p-harmonic [29] | 39.8 | 8.89 25 | 13.9 29 | 1.68 12 | 8.86 76 | 10.9 70 | 5.20 88 | 13.4 6 | 17.5 37 | 6.45 3 | 13.7 22 | 17.9 24 | 10.0 44 | 28.9 65 | 32.8 66 | 12.8 51 | 17.6 29 | 24.9 31 | 6.53 61 | 22.9 12 | 35.0 13 | 8.88 61 | 22.5 50 | 28.3 52 | 8.10 29 |
Second-order prior [8] | 40.3 | 8.06 3 | 12.5 3 | 1.93 76 | 8.80 72 | 11.0 74 | 4.80 73 | 12.8 2 | 16.2 25 | 7.51 8 | 12.6 8 | 16.7 12 | 6.25 4 | 28.9 65 | 32.8 66 | 12.2 28 | 18.1 43 | 25.7 45 | 6.10 30 | 23.3 19 | 35.5 18 | 9.35 72 | 22.7 57 | 28.6 61 | 8.45 103 |
IROF-TV [53] | 40.4 | 8.93 27 | 13.9 29 | 1.82 62 | 8.15 47 | 10.6 54 | 4.01 42 | 13.9 14 | 17.6 38 | 8.70 16 | 13.3 12 | 18.0 25 | 9.04 30 | 28.5 54 | 32.3 53 | 15.3 94 | 18.5 55 | 26.2 57 | 6.57 63 | 24.7 47 | 37.8 49 | 6.81 3 | 22.2 45 | 28.0 48 | 6.71 5 |
ALD-Flow [66] | 41.6 | 10.4 96 | 16.0 94 | 1.76 40 | 7.99 35 | 10.3 32 | 3.78 33 | 14.1 17 | 19.3 64 | 6.64 4 | 16.1 91 | 21.9 98 | 5.92 2 | 26.5 20 | 30.0 20 | 14.0 67 | 16.9 18 | 23.9 18 | 6.23 37 | 22.5 8 | 34.4 8 | 7.50 17 | 23.2 74 | 29.2 79 | 8.09 26 |
OAR-Flow [125] | 41.6 | 9.14 47 | 14.0 36 | 1.71 23 | 7.90 27 | 10.1 18 | 4.04 45 | 14.3 20 | 18.8 54 | 5.59 1 | 16.6 99 | 22.6 104 | 6.23 3 | 27.7 40 | 31.4 40 | 15.3 94 | 15.9 5 | 22.4 5 | 6.89 74 | 24.2 32 | 36.4 29 | 7.80 36 | 22.9 64 | 28.8 65 | 8.15 37 |
Ad-TV-NDC [36] | 41.9 | 9.09 43 | 13.8 21 | 2.24 104 | 9.50 110 | 11.1 83 | 6.94 121 | 14.2 19 | 15.4 13 | 6.85 5 | 14.5 44 | 18.6 35 | 9.51 37 | 27.4 33 | 31.1 33 | 12.3 39 | 18.2 48 | 25.8 51 | 6.40 53 | 22.9 12 | 34.7 10 | 7.43 10 | 20.6 9 | 25.8 9 | 8.26 64 |
IROF++ [58] | 43.1 | 8.58 10 | 13.3 10 | 1.68 12 | 7.99 35 | 10.4 42 | 3.84 36 | 17.3 58 | 18.5 49 | 12.5 49 | 12.4 4 | 16.7 12 | 9.15 31 | 28.4 53 | 32.3 53 | 15.3 94 | 19.5 89 | 27.6 90 | 6.06 24 | 23.3 19 | 35.6 20 | 8.55 60 | 23.4 82 | 29.4 86 | 7.79 17 |
DF-Auto [115] | 43.8 | 9.30 57 | 14.4 60 | 1.99 85 | 8.37 57 | 10.6 54 | 4.99 76 | 15.5 31 | 22.3 86 | 8.88 17 | 13.3 12 | 17.6 19 | 9.86 41 | 25.7 16 | 29.1 16 | 13.9 65 | 18.1 43 | 25.7 45 | 5.96 17 | 25.2 54 | 38.6 59 | 10.8 93 | 20.7 12 | 25.9 10 | 8.09 26 |
Modified CLG [34] | 44.3 | 7.87 1 | 12.2 1 | 1.68 12 | 8.96 80 | 10.7 60 | 5.94 115 | 16.8 50 | 16.7 30 | 15.9 73 | 13.3 12 | 16.4 10 | 12.6 103 | 27.6 38 | 31.3 38 | 11.9 9 | 18.8 66 | 26.6 67 | 6.50 59 | 22.3 6 | 34.0 6 | 7.67 28 | 22.2 45 | 27.9 45 | 8.64 109 |
Brox et al. [5] | 44.5 | 9.33 59 | 14.7 65 | 1.62 5 | 7.86 23 | 10.1 18 | 4.14 48 | 15.9 34 | 16.0 20 | 10.4 29 | 13.5 18 | 17.7 20 | 8.77 27 | 26.8 24 | 30.4 24 | 11.9 9 | 19.1 77 | 27.0 78 | 9.52 124 | 28.6 104 | 43.6 102 | 23.0 129 | 19.9 4 | 25.0 4 | 8.05 22 |
NNF-EAC [103] | 46.8 | 9.00 36 | 14.0 36 | 1.99 85 | 7.79 18 | 10.2 24 | 2.85 12 | 17.5 62 | 25.1 109 | 19.2 93 | 15.4 69 | 20.6 75 | 11.6 79 | 29.9 93 | 33.9 93 | 12.1 17 | 16.5 12 | 23.4 12 | 5.99 19 | 22.9 12 | 35.1 15 | 7.51 18 | 20.9 17 | 26.2 17 | 8.42 100 |
Local-TV-L1 [65] | 49.5 | 8.65 13 | 13.3 10 | 1.90 72 | 9.07 88 | 11.0 74 | 5.04 80 | 13.1 4 | 15.3 12 | 8.62 15 | 12.8 9 | 17.0 14 | 7.89 17 | 30.8 119 | 35.0 119 | 15.5 121 | 18.4 51 | 26.0 53 | 6.98 78 | 23.9 29 | 36.5 31 | 7.66 27 | 21.4 28 | 26.9 28 | 8.40 96 |
F-TV-L1 [15] | 49.7 | 10.4 96 | 16.2 96 | 1.94 78 | 9.02 83 | 11.2 88 | 4.72 65 | 14.6 23 | 16.7 30 | 11.0 35 | 14.2 33 | 18.9 44 | 10.3 48 | 27.5 35 | 31.2 36 | 12.3 39 | 16.0 6 | 22.6 6 | 6.38 51 | 23.9 29 | 36.6 32 | 9.23 68 | 21.3 25 | 26.7 24 | 10.2 122 |
PH-Flow [101] | 50.8 | 9.30 57 | 14.3 51 | 1.70 21 | 7.70 15 | 10.1 18 | 2.82 11 | 14.9 27 | 20.6 77 | 14.8 67 | 14.3 37 | 19.4 53 | 11.5 75 | 25.0 13 | 28.3 13 | 12.2 28 | 21.6 124 | 30.7 125 | 9.38 123 | 25.0 52 | 38.3 54 | 7.76 34 | 22.6 53 | 28.4 54 | 8.15 37 |
FMOF [94] | 51.6 | 9.22 53 | 13.9 29 | 1.96 80 | 7.58 11 | 9.87 14 | 2.87 13 | 19.5 79 | 22.4 87 | 17.7 83 | 15.3 65 | 20.6 75 | 12.5 102 | 24.5 11 | 27.7 11 | 13.7 61 | 19.3 85 | 27.3 86 | 6.05 23 | 24.6 45 | 37.7 47 | 6.64 2 | 23.4 82 | 29.4 86 | 6.98 8 |
Filter Flow [19] | 52.5 | 9.35 61 | 14.5 61 | 1.79 52 | 9.19 96 | 11.1 83 | 5.50 105 | 17.6 65 | 16.8 33 | 12.2 43 | 14.0 29 | 18.0 25 | 11.3 65 | 24.6 12 | 27.9 12 | 12.2 28 | 18.4 51 | 26.0 53 | 7.54 95 | 24.8 48 | 37.9 51 | 7.77 35 | 21.5 30 | 27.0 31 | 8.40 96 |
CRTflow [80] | 54.8 | 8.75 15 | 13.6 16 | 2.04 89 | 9.27 100 | 11.5 108 | 5.28 89 | 16.2 38 | 22.5 89 | 9.27 20 | 12.8 9 | 17.0 14 | 11.5 75 | 27.0 26 | 30.6 26 | 15.3 94 | 17.6 29 | 24.9 31 | 6.06 24 | 27.8 92 | 42.7 95 | 7.62 25 | 23.4 82 | 29.4 86 | 8.16 42 |
CNN-flow-warp+ref [117] | 55.1 | 8.33 7 | 13.0 8 | 2.06 93 | 8.26 54 | 10.3 32 | 5.85 112 | 18.3 71 | 22.7 91 | 11.1 37 | 13.6 19 | 16.0 5 | 11.1 61 | 29.1 75 | 33.0 75 | 15.3 94 | 15.7 3 | 22.1 3 | 6.96 77 | 28.2 98 | 43.1 98 | 7.67 28 | 21.8 38 | 27.3 37 | 8.49 106 |
TC/T-Flow [76] | 55.3 | 9.42 63 | 14.6 64 | 2.39 108 | 8.67 69 | 11.2 88 | 4.00 40 | 13.6 9 | 16.0 20 | 8.03 12 | 17.5 112 | 23.5 115 | 10.8 55 | 27.3 31 | 31.0 31 | 15.3 94 | 17.4 27 | 24.6 27 | 5.89 11 | 25.8 68 | 37.8 49 | 9.59 77 | 22.8 60 | 28.7 63 | 8.13 35 |
COFM [59] | 55.5 | 8.95 32 | 13.8 21 | 1.90 72 | 7.42 8 | 9.61 10 | 3.19 21 | 15.3 29 | 22.1 85 | 16.3 74 | 15.4 69 | 20.9 81 | 14.6 112 | 26.8 24 | 30.4 24 | 12.2 28 | 21.4 121 | 30.3 121 | 6.26 41 | 26.3 75 | 40.4 77 | 10.4 89 | 20.8 14 | 26.1 14 | 8.36 89 |
ComplOF-FED-GPU [35] | 55.6 | 9.91 88 | 15.5 88 | 1.77 47 | 7.74 16 | 10.1 18 | 4.25 53 | 19.8 83 | 17.7 40 | 17.0 77 | 15.3 65 | 20.7 78 | 11.8 85 | 28.2 51 | 32.0 51 | 14.5 71 | 16.2 9 | 22.8 9 | 5.95 16 | 26.2 74 | 39.6 75 | 9.25 69 | 22.7 57 | 28.4 54 | 8.25 61 |
Sparse Occlusion [54] | 56.1 | 9.75 81 | 15.2 79 | 2.05 92 | 8.71 70 | 11.2 88 | 4.19 50 | 13.5 7 | 15.9 19 | 7.80 10 | 14.6 49 | 19.7 62 | 7.51 10 | 30.4 105 | 34.5 106 | 15.3 94 | 16.1 7 | 22.7 7 | 6.27 42 | 26.9 84 | 41.1 86 | 7.45 12 | 23.2 74 | 29.2 79 | 8.12 33 |
2DHMM-SAS [92] | 56.5 | 8.83 20 | 13.6 16 | 1.76 40 | 8.88 78 | 11.3 95 | 4.29 54 | 17.5 62 | 20.9 78 | 12.5 49 | 14.5 44 | 19.6 57 | 11.3 65 | 30.1 98 | 34.1 97 | 15.1 84 | 17.6 29 | 24.9 31 | 5.84 7 | 25.2 54 | 38.7 61 | 8.23 54 | 23.1 69 | 29.1 72 | 8.16 42 |
2D-CLG [1] | 56.7 | 8.51 9 | 13.2 9 | 1.76 40 | 8.84 75 | 10.4 42 | 5.71 110 | 19.4 78 | 15.6 16 | 15.0 69 | 14.2 33 | 16.3 9 | 14.0 108 | 31.1 123 | 35.3 123 | 20.9 130 | 16.1 7 | 22.7 7 | 6.34 47 | 27.7 91 | 42.3 90 | 8.19 53 | 21.4 28 | 26.9 28 | 8.13 35 |
LDOF [28] | 56.8 | 8.85 23 | 13.8 21 | 2.04 89 | 10.2 126 | 9.70 13 | 10.8 131 | 17.0 53 | 20.4 73 | 12.0 40 | 13.4 15 | 17.4 18 | 12.3 97 | 22.9 2 | 26.0 2 | 11.9 9 | 18.9 70 | 26.7 70 | 6.27 42 | 30.1 118 | 46.3 120 | 16.0 113 | 19.7 3 | 24.7 3 | 8.89 113 |
Horn & Schunck [3] | 57.0 | 8.92 26 | 13.6 16 | 1.73 26 | 9.79 119 | 11.4 99 | 6.31 118 | 24.1 107 | 18.7 52 | 18.6 91 | 15.8 85 | 19.4 53 | 11.1 61 | 28.0 45 | 31.8 46 | 10.4 5 | 17.8 34 | 25.2 37 | 5.54 4 | 25.3 59 | 38.4 57 | 9.70 79 | 22.1 41 | 27.7 42 | 8.27 67 |
PMF [73] | 57.2 | 9.35 61 | 14.5 61 | 1.77 47 | 7.80 19 | 10.1 18 | 2.68 3 | 24.0 106 | 28.7 118 | 22.5 114 | 15.3 65 | 20.6 75 | 11.6 79 | 25.7 16 | 29.2 17 | 12.1 17 | 19.1 77 | 27.0 78 | 5.92 14 | 27.6 89 | 42.4 92 | 9.09 66 | 23.1 69 | 29.0 71 | 6.47 1 |
CPM-Flow [116] | 57.3 | 9.82 85 | 15.4 87 | 1.69 17 | 7.60 12 | 9.90 15 | 3.04 18 | 15.6 33 | 15.7 17 | 7.43 7 | 16.9 103 | 23.0 110 | 12.0 90 | 27.6 38 | 31.3 38 | 15.3 94 | 18.5 55 | 26.2 57 | 7.13 84 | 23.4 21 | 35.8 21 | 9.99 84 | 23.8 98 | 29.9 100 | 8.37 91 |
FlowNetS+ft+v [112] | 57.8 | 9.02 38 | 14.1 43 | 2.07 96 | 10.0 124 | 11.0 74 | 9.60 126 | 16.3 41 | 14.4 4 | 13.5 58 | 13.8 24 | 17.7 20 | 13.3 106 | 29.7 83 | 33.8 88 | 15.3 94 | 16.8 15 | 23.8 16 | 6.25 40 | 27.8 92 | 42.6 93 | 7.83 40 | 20.4 8 | 25.5 8 | 8.24 56 |
Black & Anandan [4] | 58.0 | 9.24 55 | 14.1 43 | 1.95 79 | 9.65 117 | 11.4 99 | 5.28 89 | 28.3 116 | 24.2 101 | 20.2 102 | 14.8 53 | 18.7 39 | 10.5 50 | 27.7 40 | 31.5 41 | 9.57 4 | 19.0 74 | 27.0 78 | 6.35 49 | 24.2 32 | 36.7 33 | 8.42 58 | 21.0 19 | 26.3 19 | 6.55 2 |
TC-Flow [46] | 58.5 | 10.9 104 | 17.1 105 | 1.71 23 | 8.86 76 | 11.6 111 | 4.00 40 | 13.0 3 | 16.0 20 | 6.24 2 | 15.6 76 | 21.1 84 | 8.58 24 | 27.9 42 | 31.7 44 | 15.1 84 | 18.7 63 | 26.4 64 | 6.72 67 | 24.6 45 | 37.6 44 | 7.95 44 | 23.4 82 | 29.4 86 | 8.28 70 |
Fusion [6] | 58.8 | 8.82 19 | 13.8 21 | 2.62 111 | 7.96 33 | 10.1 18 | 4.47 58 | 16.5 44 | 13.6 2 | 17.3 82 | 14.0 29 | 18.1 28 | 9.97 43 | 29.8 89 | 33.8 88 | 12.8 51 | 19.4 86 | 27.4 87 | 10.1 127 | 26.4 76 | 40.4 77 | 8.14 51 | 21.7 34 | 27.2 35 | 10.1 121 |
OFLAF [77] | 59.1 | 9.70 75 | 15.0 73 | 1.69 17 | 7.94 32 | 10.4 42 | 2.73 7 | 14.3 20 | 15.0 9 | 10.2 25 | 13.8 24 | 18.6 35 | 8.40 22 | 30.0 95 | 34.0 95 | 15.4 114 | 17.0 22 | 23.9 18 | 6.73 68 | 30.1 118 | 46.1 118 | 13.9 108 | 23.4 82 | 29.3 83 | 9.45 117 |
MLDP_OF [89] | 59.1 | 9.06 40 | 14.1 43 | 1.83 65 | 8.81 73 | 11.3 95 | 4.78 72 | 14.0 15 | 17.6 38 | 8.56 14 | 15.5 73 | 20.3 71 | 15.8 120 | 29.7 83 | 33.7 83 | 13.6 57 | 19.1 77 | 27.0 78 | 5.86 9 | 23.8 27 | 36.3 27 | 8.15 52 | 23.2 74 | 29.1 72 | 8.25 61 |
PGM-C [120] | 59.4 | 9.70 75 | 15.2 79 | 1.69 17 | 7.84 22 | 10.2 24 | 3.70 29 | 21.2 90 | 17.2 35 | 12.3 46 | 17.4 109 | 23.6 116 | 8.69 25 | 28.0 45 | 31.8 46 | 15.3 94 | 16.6 13 | 23.4 12 | 6.17 32 | 26.4 76 | 40.5 81 | 8.04 48 | 24.3 114 | 30.5 116 | 8.34 82 |
EpicFlow [102] | 59.5 | 9.69 74 | 15.2 79 | 1.67 11 | 7.90 27 | 10.2 24 | 4.37 56 | 16.0 36 | 14.5 5 | 9.75 23 | 19.1 121 | 25.8 123 | 12.3 97 | 27.9 42 | 31.6 42 | 15.3 94 | 16.9 18 | 23.9 18 | 6.21 36 | 24.9 50 | 38.0 52 | 10.3 88 | 24.6 119 | 30.9 120 | 8.30 74 |
Kuang [131] | 59.7 | 9.55 68 | 15.0 73 | 1.96 80 | 7.63 13 | 9.90 15 | 4.02 44 | 23.6 104 | 19.5 66 | 17.0 77 | 16.3 94 | 21.9 98 | 7.62 11 | 28.3 52 | 32.1 52 | 13.9 65 | 17.3 25 | 24.4 25 | 4.83 1 | 24.5 39 | 37.6 44 | 20.5 124 | 23.9 105 | 30.1 109 | 8.20 49 |
TV-L1-MCT [64] | 60.7 | 9.18 50 | 14.2 49 | 1.78 49 | 8.53 63 | 11.1 83 | 3.70 29 | 17.7 67 | 23.3 98 | 13.6 59 | 14.4 41 | 19.5 56 | 11.6 79 | 30.5 110 | 34.6 108 | 13.8 62 | 18.1 43 | 25.7 45 | 6.02 22 | 25.8 68 | 39.5 73 | 15.0 111 | 21.7 34 | 27.3 37 | 7.99 20 |
Bartels [41] | 61.0 | 12.7 119 | 20.1 120 | 2.13 101 | 8.52 62 | 11.0 74 | 4.96 75 | 13.5 7 | 14.5 5 | 10.2 25 | 14.4 41 | 18.9 44 | 10.8 55 | 23.5 4 | 26.6 4 | 12.9 54 | 19.0 74 | 26.9 75 | 6.94 76 | 24.5 39 | 37.5 42 | 19.7 123 | 23.4 82 | 29.4 86 | 8.31 76 |
AGIF+OF [85] | 61.1 | 9.07 41 | 14.0 36 | 1.78 49 | 7.93 31 | 10.3 32 | 3.78 33 | 14.4 22 | 17.8 42 | 12.4 47 | 14.9 55 | 20.2 68 | 11.4 70 | 28.9 65 | 32.8 66 | 15.3 94 | 20.0 98 | 28.3 98 | 6.98 78 | 25.5 62 | 39.0 64 | 7.74 32 | 23.9 105 | 30.1 109 | 8.28 70 |
S2F-IF [123] | 62.4 | 10.3 94 | 16.3 98 | 1.79 52 | 7.83 21 | 10.2 24 | 2.90 15 | 17.0 53 | 20.0 70 | 13.9 62 | 16.1 91 | 21.6 94 | 6.69 5 | 29.2 77 | 33.2 78 | 15.3 94 | 16.8 15 | 23.7 15 | 6.34 47 | 24.9 50 | 38.2 53 | 10.7 91 | 23.9 105 | 30.0 107 | 8.35 87 |
BlockOverlap [61] | 62.8 | 9.09 43 | 14.3 51 | 2.04 89 | 8.96 80 | 10.9 70 | 5.37 99 | 18.1 69 | 15.5 14 | 18.0 86 | 14.2 33 | 17.2 16 | 14.0 108 | 28.9 65 | 32.8 66 | 13.8 62 | 18.8 66 | 26.7 70 | 7.92 106 | 24.8 48 | 37.2 37 | 21.0 126 | 20.0 6 | 25.1 6 | 8.38 92 |
OFH [38] | 63.7 | 9.54 67 | 15.0 73 | 1.74 33 | 8.49 61 | 10.6 54 | 5.13 83 | 18.1 69 | 24.9 107 | 10.4 29 | 17.4 109 | 23.7 118 | 5.72 1 | 28.7 58 | 32.5 56 | 14.6 74 | 17.6 29 | 24.8 29 | 5.85 8 | 26.0 72 | 39.2 68 | 10.2 86 | 22.7 57 | 28.5 59 | 14.1 128 |
nLayers [57] | 63.8 | 9.15 48 | 14.3 51 | 1.76 40 | 7.42 8 | 9.62 11 | 3.57 25 | 27.8 114 | 29.9 121 | 25.8 124 | 15.9 87 | 21.5 92 | 11.9 86 | 30.2 99 | 34.3 100 | 14.7 77 | 20.3 104 | 28.8 105 | 6.45 56 | 23.5 24 | 36.0 25 | 7.87 42 | 21.6 31 | 27.1 32 | 8.10 29 |
HAST [109] | 63.9 | 8.87 24 | 13.8 21 | 1.76 40 | 7.34 6 | 9.50 7 | 2.70 5 | 28.8 118 | 28.6 117 | 24.0 119 | 14.9 55 | 20.2 68 | 7.68 13 | 28.9 65 | 32.8 66 | 12.1 17 | 21.3 120 | 30.2 120 | 7.57 97 | 28.6 104 | 43.9 105 | 7.55 21 | 22.8 60 | 28.7 63 | 8.43 102 |
TCOF [69] | 64.2 | 9.34 60 | 14.3 51 | 1.89 69 | 9.50 110 | 11.7 117 | 5.42 100 | 16.2 38 | 21.7 82 | 10.3 27 | 13.8 24 | 18.6 35 | 9.45 36 | 30.4 105 | 34.6 108 | 13.6 57 | 18.2 48 | 25.7 45 | 6.20 35 | 28.5 101 | 43.5 101 | 7.54 20 | 22.9 64 | 28.8 65 | 8.18 44 |
Layers++ [37] | 64.6 | 8.93 27 | 14.0 36 | 1.76 40 | 6.74 1 | 8.61 1 | 2.71 6 | 18.3 71 | 25.8 111 | 19.3 94 | 15.3 65 | 20.8 79 | 11.3 65 | 33.1 128 | 37.6 128 | 19.8 128 | 21.6 124 | 30.6 124 | 8.73 118 | 24.4 37 | 37.4 40 | 7.81 38 | 21.6 31 | 27.1 32 | 8.09 26 |
DPOF [18] | 65.4 | 11.0 105 | 17.4 109 | 3.88 123 | 7.78 17 | 10.2 24 | 3.01 16 | 18.7 74 | 18.1 47 | 18.4 89 | 16.5 97 | 22.4 102 | 14.6 112 | 28.8 60 | 32.7 60 | 12.1 17 | 18.9 70 | 26.7 70 | 6.18 34 | 25.2 54 | 38.4 57 | 7.59 24 | 23.6 93 | 29.6 93 | 8.07 23 |
FlowFields [110] | 65.9 | 9.98 89 | 15.7 90 | 2.08 98 | 7.96 33 | 10.4 42 | 3.62 26 | 23.1 98 | 23.2 95 | 20.3 104 | 16.0 89 | 21.5 92 | 7.08 8 | 27.0 26 | 30.6 26 | 14.2 69 | 19.2 83 | 27.1 83 | 6.08 28 | 24.4 37 | 37.4 40 | 10.2 86 | 23.2 74 | 29.2 79 | 8.35 87 |
Classic++ [32] | 66.5 | 9.48 64 | 14.9 69 | 1.80 56 | 8.59 64 | 11.0 74 | 4.61 61 | 13.7 11 | 15.0 9 | 9.57 22 | 14.4 41 | 19.0 46 | 8.76 26 | 29.9 93 | 33.9 93 | 13.6 57 | 20.2 102 | 28.7 103 | 6.87 73 | 27.4 87 | 42.0 87 | 9.63 78 | 23.8 98 | 29.9 100 | 8.34 82 |
SRR-TVOF-NL [91] | 67.1 | 9.65 72 | 14.8 66 | 1.82 62 | 8.21 52 | 10.6 54 | 4.76 71 | 22.7 96 | 28.1 115 | 21.9 109 | 15.6 76 | 20.9 81 | 9.18 33 | 28.9 65 | 32.8 66 | 15.3 94 | 20.7 110 | 29.3 110 | 5.91 12 | 24.5 39 | 37.6 44 | 6.56 1 | 22.5 50 | 28.2 51 | 8.34 82 |
HBM-GC [105] | 67.4 | 9.25 56 | 14.5 61 | 1.81 60 | 9.08 90 | 11.9 123 | 3.75 32 | 17.3 58 | 18.7 52 | 17.9 84 | 14.3 37 | 19.2 48 | 8.85 28 | 30.0 95 | 34.0 95 | 15.5 121 | 21.5 122 | 30.4 122 | 8.27 115 | 27.4 87 | 42.1 89 | 7.15 5 | 20.8 14 | 26.1 14 | 7.05 10 |
NL-TV-NCC [25] | 68.0 | 9.19 52 | 14.3 51 | 2.18 102 | 9.02 83 | 11.6 111 | 4.13 47 | 14.8 25 | 16.7 30 | 10.9 33 | 20.8 125 | 28.1 126 | 8.19 20 | 26.5 20 | 30.0 20 | 13.1 55 | 18.9 70 | 26.7 70 | 6.43 55 | 26.6 80 | 40.4 77 | 15.1 112 | 23.7 97 | 29.7 97 | 8.29 73 |
Nguyen [33] | 68.3 | 9.83 86 | 15.2 79 | 1.73 26 | 9.59 115 | 11.0 74 | 5.65 109 | 15.3 29 | 20.5 74 | 10.3 27 | 14.6 49 | 18.8 40 | 12.1 92 | 28.8 60 | 32.7 60 | 12.2 28 | 19.4 86 | 27.4 87 | 8.01 110 | 29.7 113 | 45.5 113 | 8.29 55 | 21.2 22 | 26.6 23 | 8.34 82 |
Efficient-NL [60] | 68.4 | 8.71 14 | 13.5 14 | 1.68 12 | 8.66 68 | 11.2 88 | 3.65 27 | 22.5 93 | 20.0 70 | 19.9 98 | 14.3 37 | 19.3 50 | 11.0 59 | 30.5 110 | 34.7 113 | 15.0 79 | 20.1 99 | 28.4 99 | 6.27 42 | 28.5 101 | 43.7 103 | 8.92 63 | 23.8 98 | 29.9 100 | 6.66 4 |
Complementary OF [21] | 68.5 | 11.4 111 | 18.1 114 | 1.70 21 | 9.23 99 | 12.1 124 | 4.19 50 | 31.6 123 | 19.0 58 | 23.6 116 | 19.5 124 | 26.5 124 | 6.72 6 | 28.1 49 | 31.8 46 | 14.6 74 | 17.3 25 | 24.4 25 | 6.38 51 | 26.1 73 | 39.0 64 | 8.92 63 | 22.3 47 | 27.9 45 | 7.57 13 |
AggregFlow [97] | 69.4 | 12.9 121 | 20.3 121 | 1.75 38 | 8.34 56 | 10.8 63 | 4.14 48 | 20.0 84 | 24.4 104 | 19.5 97 | 16.5 97 | 22.3 101 | 12.2 95 | 25.2 14 | 28.6 14 | 12.2 28 | 16.9 18 | 23.9 18 | 6.60 64 | 29.0 110 | 43.9 105 | 16.7 115 | 23.0 66 | 28.9 67 | 8.03 21 |
FESL [72] | 69.7 | 9.09 43 | 13.9 29 | 1.74 33 | 7.90 27 | 10.3 32 | 3.35 23 | 16.5 44 | 21.9 83 | 12.0 40 | 15.1 62 | 20.3 71 | 11.4 70 | 30.8 119 | 35.0 119 | 15.4 114 | 19.6 91 | 27.8 93 | 6.48 57 | 27.8 92 | 42.6 93 | 7.75 33 | 23.9 105 | 30.0 107 | 8.39 93 |
ProbFlowFields [128] | 69.8 | 10.1 90 | 16.0 94 | 1.78 49 | 8.04 39 | 10.5 49 | 3.08 20 | 25.8 113 | 28.8 119 | 24.3 120 | 14.5 44 | 19.6 57 | 11.4 70 | 27.2 29 | 30.9 30 | 15.3 94 | 17.4 27 | 24.6 27 | 8.78 119 | 27.3 86 | 42.0 87 | 18.8 121 | 22.4 49 | 28.1 49 | 8.39 93 |
StereoOF-V1MT [119] | 70.8 | 11.1 107 | 17.3 107 | 1.73 26 | 8.61 65 | 10.6 54 | 5.28 89 | 23.4 102 | 17.3 36 | 17.1 80 | 16.6 99 | 19.9 65 | 12.3 97 | 27.4 33 | 31.1 33 | 15.0 79 | 17.0 22 | 23.8 16 | 6.80 71 | 30.2 120 | 46.2 119 | 12.3 102 | 21.6 31 | 26.9 28 | 9.58 118 |
FlowFields+ [130] | 71.0 | 9.67 73 | 15.2 79 | 3.33 121 | 7.86 23 | 10.3 32 | 3.02 17 | 23.3 100 | 24.6 105 | 20.8 105 | 17.0 104 | 23.0 110 | 6.91 7 | 27.3 31 | 31.0 31 | 15.4 114 | 19.0 74 | 26.9 75 | 6.24 38 | 25.3 59 | 38.8 62 | 13.1 104 | 23.2 74 | 29.1 72 | 8.39 93 |
RNLOD-Flow [121] | 71.1 | 8.93 27 | 13.8 21 | 1.65 9 | 8.48 60 | 11.0 74 | 4.06 46 | 16.3 41 | 23.2 95 | 12.8 53 | 14.1 31 | 19.1 47 | 11.1 61 | 29.7 83 | 33.7 83 | 15.6 123 | 20.3 104 | 28.7 103 | 8.92 121 | 25.7 65 | 39.4 71 | 16.4 114 | 24.2 112 | 30.4 114 | 8.20 49 |
FlowNet2 [122] | 71.8 | 15.6 129 | 23.6 130 | 1.96 80 | 9.34 102 | 12.1 124 | 4.72 65 | 17.3 58 | 19.2 63 | 13.0 56 | 17.1 107 | 23.1 112 | 10.1 45 | 28.0 45 | 31.8 46 | 12.3 39 | 18.6 57 | 26.3 60 | 6.35 49 | 26.7 81 | 40.8 84 | 8.04 48 | 21.7 34 | 27.2 35 | 8.30 74 |
TI-DOFE [24] | 72.0 | 9.80 83 | 15.2 79 | 2.80 115 | 9.94 122 | 11.4 99 | 5.62 107 | 15.5 31 | 15.7 17 | 10.5 32 | 17.0 104 | 21.7 96 | 10.6 51 | 27.1 28 | 30.8 28 | 12.1 17 | 20.9 114 | 29.6 115 | 6.99 80 | 24.0 31 | 36.3 27 | 8.92 63 | 24.3 114 | 28.1 49 | 12.5 126 |
Sparse-NonSparse [56] | 72.4 | 9.18 50 | 14.3 51 | 1.73 26 | 8.14 46 | 10.6 54 | 3.31 22 | 16.6 47 | 22.9 93 | 13.8 61 | 14.8 53 | 19.8 64 | 11.3 65 | 30.5 110 | 34.6 108 | 15.0 79 | 20.1 99 | 28.5 101 | 7.48 93 | 28.5 101 | 43.7 103 | 9.49 73 | 23.5 91 | 29.5 91 | 8.24 56 |
LSM [39] | 73.2 | 9.10 46 | 14.2 49 | 1.73 26 | 8.33 55 | 10.9 70 | 3.40 24 | 16.6 47 | 22.7 91 | 12.2 43 | 15.0 61 | 20.3 71 | 11.0 59 | 30.5 110 | 34.7 113 | 15.1 84 | 20.7 110 | 29.4 111 | 6.17 32 | 28.1 97 | 43.0 97 | 11.5 97 | 23.8 98 | 29.9 100 | 8.27 67 |
Occlusion-TV-L1 [63] | 73.5 | 10.1 90 | 15.9 92 | 2.43 109 | 9.36 103 | 11.8 121 | 5.01 79 | 12.7 1 | 14.7 7 | 7.22 6 | 17.0 104 | 22.7 106 | 11.4 70 | 28.6 56 | 32.5 56 | 12.0 14 | 18.7 63 | 26.5 66 | 7.48 93 | 25.2 54 | 37.7 47 | 10.0 85 | 24.3 114 | 30.3 113 | 9.33 115 |
ACK-Prior [27] | 73.5 | 9.81 84 | 15.1 77 | 2.07 96 | 8.01 38 | 10.4 42 | 3.86 37 | 25.1 110 | 19.1 59 | 22.0 111 | 15.1 62 | 20.1 67 | 10.1 45 | 30.4 105 | 34.4 103 | 15.4 114 | 19.1 77 | 26.9 75 | 7.57 97 | 25.8 68 | 39.3 69 | 19.5 122 | 22.3 47 | 27.9 45 | 7.73 15 |
Classic+CPF [83] | 73.6 | 9.07 41 | 14.0 36 | 1.80 56 | 8.09 44 | 10.5 49 | 3.71 31 | 17.0 53 | 21.5 80 | 12.9 55 | 13.9 28 | 18.8 40 | 11.4 70 | 30.7 117 | 34.9 118 | 15.4 114 | 21.2 117 | 30.0 118 | 7.73 104 | 28.2 98 | 43.2 99 | 7.80 36 | 24.7 122 | 31.0 122 | 7.89 19 |
FFV1MT [106] | 73.9 | 11.6 112 | 17.7 111 | 2.19 103 | 9.20 97 | 10.9 70 | 5.96 116 | 22.6 94 | 30.3 122 | 16.3 74 | 15.5 73 | 18.8 40 | 12.4 100 | 27.5 35 | 31.2 36 | 11.6 7 | 18.6 57 | 25.7 45 | 7.42 92 | 27.2 85 | 40.7 82 | 8.88 61 | 21.2 22 | 26.4 20 | 9.73 119 |
TriFlow [95] | 74.5 | 13.1 122 | 20.8 122 | 2.06 93 | 9.53 113 | 12.2 126 | 5.29 94 | 16.5 44 | 18.5 49 | 10.1 24 | 17.2 108 | 22.8 108 | 7.74 14 | 27.9 42 | 31.6 42 | 15.1 84 | 19.4 86 | 27.4 87 | 6.07 26 | 24.5 39 | 37.2 37 | 10.9 94 | 23.8 98 | 29.8 99 | 8.15 37 |
CostFilter [40] | 74.9 | 10.8 103 | 17.0 104 | 1.80 56 | 7.90 27 | 10.3 32 | 2.66 1 | 24.6 109 | 27.7 114 | 21.9 109 | 18.7 118 | 25.4 122 | 13.7 107 | 27.5 35 | 31.1 33 | 12.6 47 | 18.2 48 | 25.8 51 | 5.87 10 | 28.9 108 | 44.2 109 | 9.34 71 | 24.4 117 | 30.7 118 | 8.20 49 |
Ramp [62] | 75.1 | 9.22 53 | 14.3 51 | 1.73 26 | 8.19 50 | 10.7 60 | 4.24 52 | 21.9 92 | 28.8 119 | 21.1 108 | 14.2 33 | 19.2 48 | 11.6 79 | 30.6 115 | 34.8 116 | 14.8 78 | 20.4 107 | 29.0 108 | 7.40 90 | 28.0 96 | 42.9 96 | 7.57 22 | 23.0 66 | 28.9 67 | 8.28 70 |
TF+OM [100] | 75.3 | 11.8 114 | 18.7 116 | 3.19 118 | 8.23 53 | 10.8 63 | 4.54 60 | 15.1 28 | 19.7 67 | 10.4 29 | 16.3 94 | 21.9 98 | 7.87 16 | 28.9 65 | 32.8 66 | 19.1 127 | 18.6 57 | 26.3 60 | 6.68 66 | 26.5 79 | 40.7 82 | 11.5 97 | 23.8 98 | 29.9 100 | 8.23 55 |
IAOF2 [51] | 76.4 | 10.7 102 | 16.6 101 | 2.36 106 | 9.40 104 | 11.6 111 | 5.33 95 | 17.4 61 | 18.0 44 | 12.4 47 | 14.1 31 | 18.2 29 | 9.32 35 | 30.3 103 | 34.4 103 | 14.0 67 | 20.5 109 | 29.1 109 | 8.20 113 | 25.2 54 | 38.3 54 | 8.49 59 | 23.1 69 | 29.1 72 | 8.24 56 |
Heeger++ [104] | 76.7 | 14.5 126 | 21.7 125 | 4.63 125 | 9.50 110 | 11.0 74 | 5.73 111 | 25.4 112 | 23.5 100 | 14.4 64 | 15.5 73 | 18.8 40 | 12.4 100 | 28.7 58 | 32.5 56 | 15.2 91 | 15.8 4 | 22.2 4 | 6.73 68 | 27.6 89 | 39.8 76 | 9.28 70 | 22.1 41 | 27.6 41 | 8.34 82 |
SVFilterOh [111] | 77.5 | 10.5 99 | 16.4 99 | 1.97 83 | 7.65 14 | 9.98 17 | 3.05 19 | 28.0 115 | 30.4 123 | 25.4 122 | 15.6 76 | 21.2 87 | 14.7 114 | 28.9 65 | 32.7 60 | 15.4 114 | 20.1 99 | 28.4 99 | 6.61 65 | 25.8 68 | 39.5 73 | 7.84 41 | 22.5 50 | 28.3 52 | 8.49 106 |
Classic+NL [31] | 77.7 | 8.97 33 | 13.9 29 | 1.79 52 | 8.11 45 | 10.5 49 | 4.01 42 | 20.8 88 | 28.3 116 | 19.9 98 | 14.3 37 | 19.3 50 | 11.5 75 | 30.7 117 | 34.8 116 | 14.6 74 | 20.3 104 | 28.8 105 | 7.40 90 | 28.3 100 | 43.4 100 | 11.9 100 | 23.5 91 | 29.6 93 | 8.25 61 |
TV-L1-improved [17] | 77.8 | 9.53 66 | 14.9 69 | 1.99 85 | 9.46 107 | 11.7 117 | 5.17 85 | 22.6 94 | 14.8 8 | 20.1 101 | 13.4 15 | 17.8 23 | 8.05 19 | 30.2 99 | 34.3 100 | 11.9 9 | 19.6 91 | 27.7 91 | 8.09 111 | 29.9 116 | 45.8 116 | 9.73 80 | 23.4 82 | 29.3 83 | 8.42 100 |
Dynamic MRF [7] | 78.0 | 10.1 90 | 15.9 92 | 1.81 60 | 8.42 59 | 10.8 63 | 4.73 69 | 19.5 79 | 19.1 59 | 12.2 43 | 15.6 76 | 19.3 50 | 12.8 105 | 27.2 29 | 30.8 28 | 15.2 91 | 18.6 57 | 26.3 60 | 7.28 87 | 28.8 107 | 44.1 108 | 12.4 103 | 24.6 119 | 30.7 118 | 9.73 119 |
IAOF [50] | 78.1 | 11.1 107 | 16.6 101 | 5.32 128 | 10.6 128 | 12.3 127 | 5.87 113 | 23.3 100 | 24.2 101 | 19.4 95 | 15.4 69 | 19.7 62 | 12.0 90 | 28.9 65 | 32.8 66 | 12.1 17 | 18.8 66 | 26.6 67 | 7.26 86 | 25.6 64 | 39.1 67 | 7.35 8 | 22.1 41 | 27.8 43 | 8.26 64 |
Adaptive [20] | 78.5 | 11.0 105 | 17.3 107 | 1.89 69 | 9.41 106 | 11.6 111 | 5.19 87 | 14.8 25 | 17.1 34 | 11.1 37 | 15.7 82 | 21.1 84 | 12.1 92 | 31.1 123 | 35.3 123 | 12.0 14 | 18.8 66 | 26.6 67 | 8.00 109 | 27.8 92 | 42.3 90 | 8.01 45 | 22.6 53 | 28.4 54 | 8.63 108 |
ROF-ND [107] | 78.5 | 9.00 36 | 13.9 29 | 1.62 5 | 9.53 113 | 10.8 63 | 10.7 130 | 16.4 43 | 22.9 93 | 12.7 52 | 18.3 116 | 24.1 119 | 11.9 86 | 29.4 81 | 33.3 81 | 15.2 91 | 18.6 57 | 26.2 57 | 7.60 100 | 24.5 39 | 37.3 39 | 13.2 106 | 24.4 117 | 30.5 116 | 9.33 115 |
Steered-L1 [118] | 78.5 | 8.76 16 | 13.7 19 | 1.82 62 | 8.00 37 | 10.3 32 | 4.72 65 | 31.9 124 | 33.2 128 | 29.2 127 | 17.4 109 | 22.8 108 | 14.1 110 | 29.5 82 | 33.5 82 | 14.2 69 | 19.7 94 | 27.9 95 | 6.28 46 | 26.4 76 | 40.4 77 | 18.7 120 | 23.8 98 | 29.9 100 | 7.04 9 |
FOLKI [16] | 78.7 | 10.6 100 | 16.5 100 | 2.43 109 | 9.94 122 | 11.2 88 | 6.70 120 | 19.6 81 | 21.6 81 | 19.9 98 | 18.3 116 | 19.4 53 | 17.3 124 | 28.0 45 | 31.7 44 | 13.6 57 | 19.1 77 | 27.1 83 | 10.9 129 | 24.2 32 | 36.9 34 | 17.3 117 | 21.3 25 | 26.7 24 | 8.10 29 |
TriangleFlow [30] | 78.8 | 9.59 70 | 14.8 66 | 2.06 93 | 9.07 88 | 11.4 99 | 5.47 103 | 19.2 77 | 20.2 72 | 13.9 62 | 13.6 19 | 18.2 29 | 8.31 21 | 30.0 95 | 34.1 97 | 9.31 3 | 17.8 34 | 25.2 37 | 7.56 96 | 30.8 122 | 47.2 122 | 13.9 108 | 25.5 127 | 31.9 128 | 11.3 124 |
SILK [79] | 80.4 | 9.72 79 | 15.1 77 | 2.69 112 | 10.2 126 | 11.4 99 | 7.82 124 | 39.2 130 | 32.9 127 | 28.5 126 | 14.6 49 | 18.4 32 | 9.73 40 | 29.0 74 | 32.9 74 | 10.4 5 | 21.2 117 | 30.0 118 | 7.00 81 | 24.5 39 | 37.5 42 | 8.03 47 | 23.1 69 | 28.9 67 | 8.31 76 |
LocallyOriented [52] | 80.7 | 10.1 90 | 15.7 90 | 1.79 52 | 9.46 107 | 11.6 111 | 5.28 89 | 23.1 98 | 24.2 101 | 20.9 107 | 19.3 123 | 23.2 113 | 7.35 9 | 30.4 105 | 34.6 108 | 12.6 47 | 18.9 70 | 26.8 74 | 6.27 42 | 25.7 65 | 38.6 59 | 7.89 43 | 23.6 93 | 29.6 93 | 8.19 47 |
S2D-Matching [84] | 82.2 | 9.57 69 | 14.9 69 | 1.76 40 | 8.37 57 | 10.8 63 | 4.36 55 | 20.1 86 | 24.9 107 | 18.2 87 | 15.7 82 | 21.3 91 | 15.7 119 | 28.8 60 | 32.7 60 | 14.5 71 | 21.5 122 | 30.4 122 | 11.0 130 | 25.7 65 | 39.3 69 | 11.5 97 | 23.1 69 | 29.1 72 | 8.75 112 |
BriefMatch [124] | 82.8 | 9.89 87 | 15.5 88 | 2.11 99 | 8.05 40 | 10.2 24 | 5.90 114 | 23.5 103 | 18.0 44 | 22.7 115 | 18.2 115 | 18.6 35 | 18.7 127 | 28.1 49 | 31.9 50 | 13.8 62 | 19.5 89 | 27.7 91 | 7.05 82 | 26.7 81 | 39.4 71 | 21.6 127 | 23.4 82 | 29.3 83 | 14.3 130 |
GraphCuts [14] | 83.1 | 11.7 113 | 17.8 112 | 2.02 88 | 8.15 47 | 10.5 49 | 4.65 63 | 25.3 111 | 15.2 11 | 19.4 95 | 14.9 55 | 19.6 57 | 11.9 86 | 29.8 89 | 33.8 88 | 17.8 125 | 19.6 91 | 27.8 93 | 6.50 59 | 28.6 104 | 43.9 105 | 11.1 95 | 24.0 110 | 30.2 111 | 8.15 37 |
RFlow [90] | 83.8 | 9.71 77 | 15.2 79 | 1.91 75 | 9.06 87 | 11.2 88 | 5.42 100 | 22.8 97 | 22.6 90 | 17.9 84 | 15.8 85 | 21.2 87 | 12.7 104 | 29.2 77 | 33.2 78 | 11.9 9 | 19.2 83 | 27.2 85 | 7.63 101 | 28.9 108 | 44.4 110 | 7.73 31 | 23.4 82 | 29.5 91 | 8.46 104 |
ComponentFusion [96] | 84.1 | 12.3 118 | 19.5 118 | 1.66 10 | 8.65 67 | 11.4 99 | 2.88 14 | 19.7 82 | 21.0 79 | 15.1 71 | 15.4 69 | 20.9 81 | 14.4 111 | 29.7 83 | 33.7 83 | 14.5 71 | 18.6 57 | 26.3 60 | 7.67 102 | 31.9 124 | 49.0 125 | 20.5 124 | 24.2 112 | 30.4 114 | 8.18 44 |
Learning Flow [11] | 84.5 | 8.99 35 | 14.1 43 | 1.85 68 | 9.10 92 | 11.3 95 | 4.99 76 | 40.2 131 | 42.5 131 | 31.6 131 | 14.9 55 | 17.2 16 | 12.2 95 | 30.8 119 | 35.0 119 | 15.1 84 | 18.7 63 | 26.4 64 | 7.58 99 | 25.1 53 | 38.3 54 | 11.4 96 | 25.5 127 | 31.7 126 | 8.24 56 |
Adaptive flow [45] | 84.7 | 10.3 94 | 14.8 66 | 2.37 107 | 9.87 121 | 11.5 108 | 5.57 106 | 18.0 68 | 17.9 43 | 17.1 80 | 16.4 96 | 20.0 66 | 14.8 116 | 32.3 126 | 36.7 126 | 16.6 124 | 21.1 116 | 29.8 116 | 8.41 116 | 23.8 27 | 36.4 29 | 13.1 104 | 21.7 34 | 27.1 32 | 7.17 11 |
Shiralkar [42] | 85.5 | 12.0 117 | 18.8 117 | 1.72 25 | 9.11 93 | 11.1 83 | 5.14 84 | 21.2 90 | 16.6 28 | 13.7 60 | 19.2 122 | 24.3 121 | 10.6 51 | 29.7 83 | 33.7 83 | 12.8 51 | 18.0 41 | 25.4 42 | 7.19 85 | 29.4 111 | 44.9 111 | 10.4 89 | 25.1 125 | 31.5 125 | 9.03 114 |
FC-2Layers-FF [74] | 85.5 | 9.71 77 | 14.9 69 | 2.11 99 | 7.51 10 | 9.66 12 | 4.67 64 | 20.5 87 | 25.1 109 | 20.2 102 | 15.6 76 | 21.1 84 | 11.9 86 | 30.5 110 | 34.6 108 | 15.3 94 | 20.8 112 | 29.4 111 | 7.31 88 | 29.7 113 | 45.6 115 | 9.76 81 | 23.6 93 | 29.7 97 | 8.22 54 |
SLK [47] | 85.6 | 9.63 71 | 15.0 73 | 1.90 72 | 9.14 95 | 10.3 32 | 5.63 108 | 34.7 126 | 19.7 67 | 22.4 113 | 18.9 119 | 24.2 120 | 20.4 130 | 29.8 89 | 33.8 88 | 12.2 28 | 18.1 43 | 25.5 43 | 6.93 75 | 31.9 124 | 48.8 124 | 9.12 67 | 22.8 60 | 28.5 59 | 14.2 129 |
EPPM w/o HM [88] | 86.6 | 10.4 96 | 16.2 96 | 2.97 117 | 8.62 66 | 11.3 95 | 2.76 9 | 29.0 119 | 27.4 113 | 22.2 112 | 16.8 101 | 22.6 104 | 10.8 55 | 25.8 18 | 29.2 17 | 12.1 17 | 20.2 102 | 28.6 102 | 6.49 58 | 29.8 115 | 45.8 116 | 18.0 118 | 24.0 110 | 30.2 111 | 8.72 111 |
UnFlow [129] | 87.0 | 13.4 123 | 21.2 124 | 2.71 114 | 8.81 73 | 10.7 60 | 6.35 119 | 18.7 74 | 18.9 55 | 14.8 67 | 14.6 49 | 19.6 57 | 7.77 15 | 31.8 125 | 36.1 125 | 15.0 79 | 22.2 127 | 31.4 127 | 7.79 105 | 24.2 32 | 37.0 35 | 7.49 15 | 28.1 131 | 33.7 131 | 11.6 125 |
Correlation Flow [75] | 89.3 | 9.75 81 | 15.3 86 | 1.84 66 | 9.28 101 | 11.6 111 | 5.17 85 | 17.5 62 | 18.9 55 | 15.2 72 | 16.1 91 | 21.7 96 | 11.3 65 | 30.2 99 | 34.3 100 | 12.5 45 | 21.2 117 | 29.9 117 | 8.24 114 | 31.3 123 | 47.8 123 | 9.82 83 | 24.9 124 | 31.3 124 | 6.61 3 |
HBpMotionGpu [43] | 90.7 | 12.7 119 | 19.5 118 | 2.69 112 | 9.65 117 | 11.7 117 | 5.48 104 | 20.0 84 | 23.3 98 | 17.0 77 | 17.6 113 | 23.4 114 | 10.6 51 | 30.8 119 | 35.0 119 | 25.1 131 | 20.4 107 | 28.9 107 | 7.95 107 | 22.0 4 | 33.7 5 | 7.44 11 | 23.2 74 | 29.1 72 | 8.40 96 |
PGAM+LK [55] | 92.0 | 11.9 116 | 18.0 113 | 7.26 131 | 9.48 109 | 10.8 63 | 7.62 122 | 31.5 122 | 39.9 130 | 31.4 130 | 19.0 120 | 23.6 116 | 16.3 123 | 29.1 75 | 33.0 75 | 12.6 47 | 18.4 51 | 26.0 53 | 6.80 71 | 25.5 62 | 39.0 64 | 14.8 110 | 22.6 53 | 28.4 54 | 8.41 99 |
2bit-BM-tele [98] | 92.8 | 11.1 107 | 17.2 106 | 2.34 105 | 9.40 104 | 11.7 117 | 5.36 97 | 28.5 117 | 37.1 129 | 31.0 129 | 15.7 82 | 20.8 79 | 9.18 33 | 28.6 56 | 32.5 56 | 15.0 79 | 22.0 126 | 31.1 126 | 9.53 125 | 39.1 131 | 59.9 131 | 26.9 131 | 20.8 14 | 26.1 14 | 8.11 32 |
StereoFlow [44] | 93.9 | 14.9 127 | 22.2 127 | 3.28 119 | 10.0 124 | 12.7 130 | 4.99 76 | 16.8 50 | 18.9 55 | 12.1 42 | 15.2 64 | 20.4 74 | 10.4 49 | 33.4 129 | 37.9 129 | 20.8 129 | 23.8 129 | 33.5 129 | 8.41 116 | 25.3 59 | 38.8 62 | 7.81 38 | 23.6 93 | 29.6 93 | 8.67 110 |
Rannacher [23] | 94.1 | 11.1 107 | 17.5 110 | 1.89 69 | 9.59 115 | 11.8 121 | 5.28 89 | 24.3 108 | 18.0 44 | 20.8 105 | 15.9 87 | 21.2 87 | 11.6 79 | 30.4 105 | 34.5 106 | 12.3 39 | 19.7 94 | 27.9 95 | 7.98 108 | 29.6 112 | 45.3 112 | 9.57 76 | 24.7 122 | 31.0 122 | 8.19 47 |
SimpleFlow [49] | 96.3 | 9.15 48 | 14.3 51 | 1.73 26 | 9.05 86 | 11.4 99 | 5.35 96 | 36.0 128 | 32.6 126 | 29.4 128 | 14.9 55 | 20.2 68 | 11.2 64 | 30.6 115 | 34.7 113 | 15.1 84 | 22.6 128 | 32.0 128 | 9.11 122 | 34.7 128 | 53.2 128 | 13.8 107 | 23.9 105 | 29.9 100 | 8.33 79 |
SegOF [10] | 97.5 | 11.8 114 | 18.2 115 | 5.53 129 | 8.88 78 | 11.4 99 | 4.62 62 | 31.1 121 | 20.5 74 | 23.7 117 | 25.8 130 | 34.8 131 | 18.2 126 | 30.2 99 | 34.2 99 | 15.3 94 | 19.1 77 | 27.0 78 | 7.08 83 | 32.5 126 | 49.7 126 | 16.8 116 | 22.8 60 | 28.6 61 | 8.08 25 |
Aniso-Texture [82] | 97.7 | 10.6 100 | 16.7 103 | 1.74 33 | 9.83 120 | 12.4 128 | 5.36 97 | 17.6 65 | 19.9 69 | 13.1 57 | 22.9 127 | 27.8 125 | 19.7 129 | 30.3 103 | 34.4 103 | 15.4 114 | 20.8 112 | 29.4 111 | 8.86 120 | 26.8 83 | 41.0 85 | 8.38 57 | 24.6 119 | 30.9 120 | 8.26 64 |
SPSA-learn [13] | 98.0 | 15.1 128 | 22.9 128 | 1.93 76 | 9.08 90 | 11.0 74 | 5.42 100 | 33.0 125 | 24.8 106 | 23.8 118 | 17.6 113 | 22.4 102 | 12.1 92 | 29.3 80 | 33.2 78 | 15.1 84 | 17.8 34 | 25.1 35 | 6.73 68 | 37.7 129 | 57.7 129 | 25.5 130 | 25.4 126 | 31.8 127 | 8.33 79 |
HCIC-L [99] | 101.9 | 14.3 125 | 20.9 123 | 2.86 116 | 11.2 129 | 13.3 131 | 7.62 122 | 23.9 105 | 31.6 125 | 25.6 123 | 21.0 126 | 28.2 127 | 14.8 116 | 25.6 15 | 29.0 15 | 12.2 28 | 24.0 130 | 33.9 130 | 10.5 128 | 30.5 121 | 46.8 121 | 18.5 119 | 23.3 80 | 29.2 79 | 7.37 12 |
GroupFlow [9] | 109.7 | 15.6 129 | 23.3 129 | 3.31 120 | 9.20 97 | 11.4 99 | 6.26 117 | 30.9 120 | 22.4 87 | 18.9 92 | 25.4 129 | 30.0 129 | 21.2 131 | 32.4 127 | 36.7 126 | 15.3 94 | 20.9 114 | 29.4 111 | 7.71 103 | 29.9 116 | 45.5 113 | 9.50 74 | 23.3 80 | 29.1 72 | 10.6 123 |
Pyramid LK [2] | 116.0 | 14.0 124 | 21.7 125 | 4.34 124 | 13.7 130 | 11.5 108 | 9.94 128 | 37.6 129 | 26.8 112 | 24.6 121 | 25.9 131 | 29.3 128 | 18.7 127 | 35.0 130 | 39.7 130 | 13.3 56 | 19.9 97 | 24.8 29 | 9.57 126 | 33.3 127 | 51.1 127 | 10.7 91 | 26.0 129 | 32.4 129 | 13.0 127 |
Periodicity [78] | 129.1 | 18.1 131 | 27.0 131 | 6.22 130 | 17.4 131 | 12.4 128 | 10.2 129 | 35.2 127 | 30.7 124 | 27.8 125 | 24.1 128 | 31.6 130 | 17.5 125 | 37.6 131 | 42.6 131 | 18.8 126 | 27.7 131 | 39.3 131 | 11.2 131 | 38.6 130 | 58.9 130 | 22.9 128 | 27.3 130 | 33.2 130 | 14.3 130 |
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. |