Optical flow evaluation results |
Statistics:
Average
SD
R0.5
R1.0
R2.0
A90
A95
A99
Error type: endpoint angle interpolation normalized interpolation |
Average normalized interpolation error |
avg. |
Mequon (Hidden texture) im0 GT im1 |
Schefflera (Hidden texture) im0 GT im1 |
Urban (Synthetic) im0 GT im1 |
Teddy (Stereo) im0 GT im1 |
Backyard (High-speed camera) im0 GT im1 |
Basketball (High-speed camera) im0 GT im1 |
Dumptruck (High-speed camera) im0 GT im1 |
Evergreen (High-speed camera) im0 GT im1 | ||||||||||||||||
rank | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | |
PMMST [114] | 17.3 | 0.59 3 | 0.73 4 | 0.64 4 | 0.64 18 | 0.85 11 | 0.59 4 | 0.99 21 | 1.69 45 | 1.05 36 | 0.97 25 | 1.14 65 | 1.23 32 | 0.99 5 | 0.96 4 | 1.14 2 | 1.03 7 | 1.29 8 | 1.04 7 | 0.71 21 | 1.33 24 | 0.67 10 | 0.77 5 | 1.10 6 | 0.64 48 |
NN-field [71] | 20.8 | 0.59 3 | 0.77 14 | 0.64 4 | 0.59 1 | 0.77 3 | 0.58 1 | 1.09 38 | 1.77 53 | 1.16 47 | 1.00 55 | 1.18 79 | 1.26 58 | 0.98 1 | 0.95 1 | 1.14 2 | 1.08 24 | 1.46 31 | 1.05 18 | 0.68 9 | 1.26 13 | 0.70 23 | 0.78 8 | 1.12 9 | 0.63 4 |
MDP-Flow2 [68] | 21.8 | 0.59 3 | 0.72 2 | 0.63 2 | 0.62 8 | 0.85 11 | 0.58 1 | 1.24 63 | 2.52 94 | 1.61 73 | 0.94 5 | 1.05 25 | 1.24 38 | 0.98 1 | 0.96 4 | 1.16 32 | 1.09 32 | 1.49 36 | 1.05 18 | 0.70 18 | 1.32 22 | 0.68 15 | 0.78 8 | 1.12 9 | 0.63 4 |
SepConv-v1 [127] | 25.2 | 0.54 1 | 0.81 42 | 0.54 1 | 0.67 35 | 0.91 31 | 0.63 51 | 1.07 35 | 1.18 3 | 1.07 40 | 1.03 75 | 1.04 20 | 1.28 79 | 0.99 5 | 0.96 4 | 1.14 2 | 0.96 1 | 1.01 1 | 1.02 1 | 0.68 9 | 1.15 7 | 0.76 63 | 0.70 1 | 0.96 1 | 0.66 97 |
CombBMOF [113] | 28.6 | 0.62 40 | 0.80 34 | 0.65 41 | 0.63 11 | 0.86 13 | 0.59 4 | 1.11 41 | 2.06 74 | 1.19 50 | 1.00 55 | 1.14 65 | 1.28 79 | 1.02 16 | 1.02 20 | 1.15 12 | 1.08 24 | 1.43 24 | 1.04 7 | 0.71 21 | 1.33 24 | 0.70 23 | 0.75 2 | 1.06 3 | 0.63 4 |
SuperFlow [81] | 29.5 | 0.62 40 | 0.84 60 | 0.66 72 | 0.76 66 | 1.04 66 | 0.69 84 | 0.90 8 | 1.17 2 | 0.74 8 | 1.05 85 | 1.04 20 | 1.25 45 | 0.99 5 | 0.96 4 | 1.16 32 | 1.05 10 | 1.36 13 | 1.03 2 | 0.69 15 | 1.26 13 | 0.70 23 | 0.82 17 | 1.18 18 | 0.62 1 |
NNF-Local [87] | 29.5 | 0.58 2 | 0.71 1 | 0.63 2 | 0.59 1 | 0.76 1 | 0.58 1 | 1.21 54 | 2.31 85 | 1.51 70 | 0.98 37 | 1.13 60 | 1.25 45 | 0.98 1 | 0.95 1 | 1.14 2 | 1.13 53 | 1.61 59 | 1.07 36 | 0.75 40 | 1.45 45 | 0.88 98 | 0.77 5 | 1.10 6 | 0.63 4 |
ALD-Flow [66] | 29.7 | 0.62 40 | 0.81 42 | 0.66 72 | 0.70 45 | 0.99 47 | 0.62 40 | 0.87 4 | 1.28 8 | 0.65 4 | 0.94 5 | 1.01 7 | 1.21 8 | 1.09 52 | 1.12 52 | 1.54 77 | 1.03 7 | 1.24 5 | 1.07 36 | 0.64 3 | 1.12 3 | 0.65 2 | 0.97 74 | 1.44 75 | 0.63 4 |
DeepFlow [86] | 31.4 | 0.62 40 | 0.84 60 | 0.65 41 | 0.74 63 | 1.04 66 | 0.66 72 | 0.86 3 | 1.33 11 | 0.65 4 | 0.99 48 | 1.02 10 | 1.23 32 | 1.04 27 | 1.05 27 | 1.16 32 | 1.02 4 | 1.23 4 | 1.05 18 | 0.63 1 | 1.07 1 | 0.65 2 | 0.96 69 | 1.43 70 | 0.64 48 |
LME [70] | 32.8 | 0.59 3 | 0.72 2 | 0.64 4 | 0.66 28 | 0.90 28 | 0.62 40 | 0.99 21 | 1.78 54 | 0.92 27 | 0.96 19 | 1.09 40 | 1.24 38 | 1.20 102 | 1.30 103 | 1.70 116 | 1.12 44 | 1.57 50 | 1.05 18 | 0.64 3 | 1.12 3 | 0.68 15 | 0.79 11 | 1.14 13 | 0.63 4 |
FMOF [94] | 34.4 | 0.62 40 | 0.79 28 | 0.65 41 | 0.63 11 | 0.84 8 | 0.59 4 | 1.32 72 | 2.09 76 | 1.41 64 | 0.99 48 | 1.08 33 | 1.26 58 | 1.00 11 | 0.98 10 | 1.14 2 | 1.08 24 | 1.44 25 | 1.04 7 | 0.68 9 | 1.26 13 | 0.67 10 | 1.07 114 | 1.61 114 | 0.63 4 |
IROF++ [58] | 34.8 | 0.59 3 | 0.74 6 | 0.64 4 | 0.65 26 | 0.89 23 | 0.59 4 | 1.15 46 | 1.71 47 | 1.17 48 | 0.92 1 | 0.96 1 | 1.21 8 | 1.17 79 | 1.26 79 | 1.69 99 | 1.11 37 | 1.54 39 | 1.04 7 | 0.68 9 | 1.23 10 | 0.70 23 | 1.07 114 | 1.62 117 | 0.63 4 |
DeepFlow2 [108] | 35.8 | 0.63 69 | 0.86 67 | 0.65 41 | 0.73 61 | 1.03 61 | 0.63 51 | 0.85 2 | 1.35 15 | 0.64 2 | 0.99 48 | 1.04 20 | 1.22 22 | 1.05 33 | 1.07 36 | 1.18 42 | 0.99 2 | 1.14 2 | 1.03 2 | 0.66 8 | 1.17 9 | 0.68 15 | 0.98 77 | 1.45 77 | 0.66 97 |
PH-Flow [101] | 36.5 | 0.60 12 | 0.77 14 | 0.64 4 | 0.61 6 | 0.82 7 | 0.59 4 | 1.26 67 | 2.55 97 | 1.55 71 | 0.94 5 | 1.03 14 | 1.21 8 | 0.99 5 | 0.98 10 | 1.17 40 | 1.22 95 | 1.86 103 | 1.13 89 | 0.74 37 | 1.42 41 | 0.75 60 | 0.82 17 | 1.19 21 | 0.64 48 |
CLG-TV [48] | 36.8 | 0.63 69 | 0.86 67 | 0.66 72 | 0.81 85 | 1.12 88 | 0.66 72 | 0.96 18 | 1.43 22 | 0.96 30 | 0.97 25 | 1.03 14 | 1.25 45 | 1.06 41 | 1.08 43 | 1.15 12 | 1.02 4 | 1.25 6 | 1.04 7 | 0.63 1 | 1.09 2 | 0.66 5 | 0.97 74 | 1.45 77 | 0.63 4 |
Aniso. Huber-L1 [22] | 38.1 | 0.62 40 | 0.80 34 | 0.66 72 | 0.84 92 | 1.13 91 | 0.66 72 | 1.03 27 | 1.44 23 | 0.93 28 | 0.97 25 | 1.03 14 | 1.26 58 | 1.06 41 | 1.09 44 | 1.15 12 | 1.08 24 | 1.46 31 | 1.03 2 | 0.64 3 | 1.12 3 | 0.66 5 | 0.99 82 | 1.48 87 | 0.63 4 |
SIOF [67] | 38.5 | 0.63 69 | 0.81 42 | 0.66 72 | 0.84 92 | 1.16 103 | 0.70 89 | 1.14 43 | 2.04 72 | 1.00 33 | 0.99 48 | 1.11 46 | 1.25 45 | 0.98 1 | 0.95 1 | 1.15 12 | 1.07 19 | 1.40 20 | 1.04 7 | 0.68 9 | 1.24 12 | 0.72 38 | 0.83 22 | 1.20 24 | 0.63 4 |
Second-order prior [8] | 39.5 | 0.61 21 | 0.78 23 | 0.66 72 | 0.80 83 | 1.11 84 | 0.64 60 | 1.05 30 | 1.85 59 | 0.99 32 | 0.96 19 | 1.04 20 | 1.21 8 | 1.05 33 | 1.07 36 | 1.15 12 | 1.05 10 | 1.38 16 | 1.05 18 | 0.69 15 | 1.28 17 | 0.65 2 | 1.00 89 | 1.50 92 | 0.66 97 |
TV-L1-MCT [64] | 40.1 | 0.62 40 | 0.81 42 | 0.65 41 | 0.71 51 | 1.00 50 | 0.63 51 | 1.21 54 | 2.34 87 | 1.25 53 | 0.95 14 | 1.04 20 | 1.22 22 | 1.19 99 | 1.29 99 | 1.61 85 | 1.07 19 | 1.39 17 | 1.05 18 | 0.71 21 | 1.32 22 | 0.69 18 | 0.82 17 | 1.18 18 | 0.63 4 |
IROF-TV [53] | 40.3 | 0.62 40 | 0.84 60 | 0.65 41 | 0.67 35 | 0.92 34 | 0.60 21 | 0.92 14 | 1.49 28 | 0.79 14 | 0.94 5 | 1.02 10 | 1.22 22 | 1.18 85 | 1.28 91 | 1.70 116 | 1.12 44 | 1.58 52 | 1.05 18 | 0.79 54 | 1.57 56 | 0.70 23 | 0.85 28 | 1.24 29 | 0.64 48 |
WLIF-Flow [93] | 40.3 | 0.59 3 | 0.73 4 | 0.64 4 | 0.66 28 | 0.92 34 | 0.61 33 | 1.34 75 | 2.50 92 | 1.59 72 | 0.98 37 | 1.07 30 | 1.28 79 | 1.03 20 | 1.04 25 | 1.22 53 | 1.19 83 | 1.76 86 | 1.12 80 | 0.72 26 | 1.37 31 | 0.70 23 | 0.83 22 | 1.20 24 | 0.63 4 |
OAR-Flow [125] | 40.5 | 0.63 69 | 0.84 60 | 0.65 41 | 0.71 51 | 1.01 54 | 0.63 51 | 0.88 6 | 1.38 18 | 0.64 2 | 0.93 3 | 1.00 5 | 1.21 8 | 1.18 85 | 1.27 84 | 1.70 116 | 1.12 44 | 1.56 46 | 1.10 63 | 0.73 30 | 1.33 24 | 0.70 23 | 0.89 40 | 1.31 44 | 0.63 4 |
NNF-EAC [103] | 41.0 | 0.63 69 | 0.78 23 | 0.66 72 | 0.66 28 | 0.91 31 | 0.60 21 | 1.45 83 | 3.35 114 | 2.05 91 | 1.05 85 | 1.26 99 | 1.27 68 | 1.06 41 | 1.09 44 | 1.14 2 | 1.02 4 | 1.26 7 | 1.03 2 | 0.72 26 | 1.36 29 | 0.69 18 | 0.79 11 | 1.13 11 | 0.63 4 |
p-harmonic [29] | 41.5 | 0.61 21 | 0.83 53 | 0.64 4 | 0.82 88 | 1.14 95 | 0.68 80 | 0.91 12 | 1.49 28 | 0.77 10 | 1.04 82 | 1.11 46 | 1.28 79 | 1.05 33 | 1.07 36 | 1.15 12 | 1.06 16 | 1.39 17 | 1.07 36 | 0.70 18 | 1.31 19 | 0.76 63 | 0.96 69 | 1.44 75 | 0.63 4 |
ComplOF-FED-GPU [35] | 41.9 | 0.62 40 | 0.86 67 | 0.65 41 | 0.69 42 | 0.98 42 | 0.61 33 | 1.63 93 | 1.15 1 | 2.12 93 | 0.94 5 | 1.03 14 | 1.21 8 | 1.14 69 | 1.21 69 | 1.52 75 | 1.07 19 | 1.41 22 | 1.06 30 | 0.74 37 | 1.36 29 | 0.71 33 | 0.96 69 | 1.43 70 | 0.63 4 |
TC/T-Flow [76] | 42.2 | 0.62 40 | 0.80 34 | 0.65 41 | 0.70 45 | 1.00 50 | 0.62 40 | 0.90 8 | 1.41 21 | 0.84 18 | 0.95 14 | 1.01 7 | 1.21 8 | 1.18 85 | 1.27 84 | 1.69 99 | 1.07 19 | 1.42 23 | 1.04 7 | 0.86 72 | 1.68 69 | 0.88 98 | 0.95 63 | 1.41 66 | 0.62 1 |
CBF [12] | 42.4 | 0.61 21 | 0.79 28 | 0.66 72 | 0.77 70 | 1.07 73 | 0.66 72 | 1.00 23 | 1.50 31 | 0.90 23 | 0.98 37 | 1.02 10 | 1.31 97 | 0.99 5 | 0.96 4 | 1.18 42 | 1.05 10 | 1.33 10 | 1.06 30 | 0.80 57 | 1.59 59 | 0.74 50 | 0.89 40 | 1.29 39 | 0.67 114 |
TC-Flow [46] | 44.0 | 0.60 12 | 0.77 14 | 0.65 41 | 0.70 45 | 1.01 54 | 0.62 40 | 0.82 1 | 1.21 6 | 0.62 1 | 0.98 37 | 1.11 46 | 1.25 45 | 1.17 79 | 1.26 79 | 1.65 88 | 1.12 44 | 1.56 46 | 1.10 63 | 0.70 18 | 1.29 18 | 0.69 18 | 1.00 89 | 1.50 92 | 0.65 80 |
nLayers [57] | 44.1 | 0.60 12 | 0.76 11 | 0.65 41 | 0.62 8 | 0.84 8 | 0.60 21 | 2.15 112 | 4.10 123 | 2.76 113 | 0.97 25 | 1.11 46 | 1.21 8 | 1.18 85 | 1.28 91 | 1.61 85 | 1.14 56 | 1.64 65 | 1.10 63 | 0.68 9 | 1.23 10 | 0.67 10 | 0.76 4 | 1.07 4 | 0.64 48 |
OFLAF [77] | 45.6 | 0.59 3 | 0.75 9 | 0.64 4 | 0.60 4 | 0.79 5 | 0.59 4 | 0.92 14 | 1.34 12 | 0.77 10 | 0.93 3 | 0.99 3 | 1.20 4 | 1.21 115 | 1.32 114 | 1.69 99 | 1.18 75 | 1.75 83 | 1.13 89 | 1.00 105 | 2.14 108 | 0.81 87 | 0.91 48 | 1.33 49 | 0.64 48 |
FlowFields [110] | 46.5 | 0.62 40 | 0.88 81 | 0.64 4 | 0.63 11 | 0.87 16 | 0.59 4 | 1.64 94 | 3.24 110 | 2.16 95 | 0.98 37 | 1.14 65 | 1.22 22 | 1.10 58 | 1.14 56 | 1.54 77 | 1.11 37 | 1.56 46 | 1.08 46 | 0.77 47 | 1.51 49 | 0.67 10 | 0.93 51 | 1.38 55 | 0.63 4 |
MLDP_OF [89] | 46.7 | 0.60 12 | 0.77 14 | 0.64 4 | 0.73 61 | 1.03 61 | 0.62 40 | 0.90 8 | 1.38 18 | 0.70 6 | 1.03 75 | 1.10 41 | 1.31 97 | 1.10 58 | 1.15 60 | 1.33 66 | 1.16 68 | 1.58 52 | 1.15 99 | 0.73 30 | 1.38 34 | 0.78 74 | 0.86 31 | 1.26 32 | 0.65 80 |
PMF [73] | 47.1 | 0.59 3 | 0.75 9 | 0.64 4 | 0.64 18 | 0.89 23 | 0.59 4 | 1.85 100 | 3.91 121 | 2.44 106 | 0.98 37 | 1.12 55 | 1.25 45 | 1.03 20 | 1.03 22 | 1.15 12 | 1.09 32 | 1.44 25 | 1.08 46 | 0.95 96 | 2.02 98 | 0.79 79 | 0.88 37 | 1.30 42 | 0.66 97 |
COFM [59] | 47.8 | 0.61 21 | 0.77 14 | 0.65 41 | 0.64 18 | 0.88 20 | 0.60 21 | 1.32 72 | 2.95 105 | 1.79 84 | 0.97 25 | 1.12 55 | 1.19 1 | 1.01 15 | 1.00 15 | 1.16 32 | 1.18 75 | 1.76 86 | 1.09 53 | 0.89 79 | 1.85 82 | 1.03 113 | 0.79 11 | 1.14 13 | 0.66 97 |
MDP-Flow [26] | 48.0 | 0.59 3 | 0.74 6 | 0.64 4 | 0.64 18 | 0.90 28 | 0.60 21 | 1.16 48 | 1.18 3 | 1.43 67 | 1.03 75 | 1.17 75 | 1.27 68 | 1.18 85 | 1.28 91 | 1.69 99 | 1.26 111 | 1.97 114 | 1.18 111 | 0.73 30 | 1.39 36 | 0.71 33 | 0.79 11 | 1.13 11 | 0.63 4 |
2DHMM-SAS [92] | 48.1 | 0.61 21 | 0.77 14 | 0.64 4 | 0.77 70 | 1.07 73 | 0.65 67 | 1.16 48 | 2.02 68 | 1.12 44 | 0.98 37 | 1.10 41 | 1.22 22 | 1.18 85 | 1.28 91 | 1.65 88 | 1.05 10 | 1.37 14 | 1.03 2 | 0.76 45 | 1.48 46 | 0.77 68 | 1.01 95 | 1.51 98 | 0.63 4 |
HAST [109] | 48.7 | 0.60 12 | 0.74 6 | 0.64 4 | 0.62 8 | 0.84 8 | 0.59 4 | 2.15 112 | 3.90 120 | 2.59 109 | 0.92 1 | 0.98 2 | 1.19 1 | 1.05 33 | 1.07 36 | 1.14 2 | 1.22 95 | 1.87 105 | 1.15 99 | 0.94 95 | 1.98 96 | 0.74 50 | 0.90 44 | 1.32 46 | 0.65 80 |
Layers++ [37] | 51.1 | 0.60 12 | 0.76 11 | 0.65 41 | 0.59 1 | 0.76 1 | 0.59 4 | 1.43 82 | 3.28 111 | 1.95 87 | 0.97 25 | 1.13 60 | 1.23 32 | 1.31 127 | 1.48 127 | 1.79 130 | 1.26 111 | 1.97 114 | 1.11 75 | 0.72 26 | 1.35 28 | 0.64 1 | 0.78 8 | 1.11 8 | 0.63 4 |
TCOF [69] | 51.1 | 0.61 21 | 0.78 23 | 0.64 4 | 0.88 107 | 1.22 117 | 0.72 101 | 1.08 37 | 1.90 63 | 1.09 41 | 0.98 37 | 1.11 46 | 1.24 38 | 1.07 45 | 1.10 48 | 1.15 12 | 1.12 44 | 1.58 52 | 1.07 36 | 0.95 96 | 2.02 98 | 0.73 43 | 0.87 33 | 1.27 36 | 0.64 48 |
CPM-Flow [116] | 52.3 | 0.62 40 | 0.86 67 | 0.64 4 | 0.64 18 | 0.88 20 | 0.60 21 | 1.04 29 | 1.39 20 | 0.82 15 | 1.01 64 | 1.19 84 | 1.27 68 | 1.17 79 | 1.26 79 | 1.69 99 | 1.11 37 | 1.52 38 | 1.07 36 | 0.82 61 | 1.66 67 | 0.76 63 | 0.99 82 | 1.47 84 | 0.65 80 |
LSM [39] | 52.9 | 0.61 21 | 0.78 23 | 0.64 4 | 0.66 28 | 0.89 23 | 0.61 33 | 1.16 48 | 2.21 79 | 1.17 48 | 0.94 5 | 1.01 7 | 1.21 8 | 1.20 102 | 1.30 103 | 1.65 88 | 1.18 75 | 1.73 79 | 1.08 46 | 0.92 88 | 1.94 92 | 0.80 84 | 1.00 89 | 1.50 92 | 0.63 4 |
CostFilter [40] | 53.0 | 0.60 12 | 0.79 28 | 0.64 4 | 0.63 11 | 0.87 16 | 0.59 4 | 1.89 102 | 3.95 122 | 2.39 104 | 0.96 19 | 1.07 30 | 1.20 4 | 1.07 45 | 1.09 44 | 1.32 64 | 1.14 56 | 1.55 43 | 1.10 63 | 1.02 109 | 2.20 112 | 0.85 94 | 0.93 51 | 1.38 55 | 0.65 80 |
ComponentFusion [96] | 53.2 | 0.60 12 | 0.80 34 | 0.64 4 | 0.64 18 | 0.88 20 | 0.59 4 | 1.41 80 | 2.74 100 | 1.63 75 | 0.95 14 | 1.08 33 | 1.20 4 | 1.13 67 | 1.19 66 | 1.35 67 | 1.11 37 | 1.55 43 | 1.08 46 | 1.23 125 | 2.74 125 | 1.51 125 | 0.95 63 | 1.41 66 | 0.64 48 |
PGM-C [120] | 53.2 | 0.62 40 | 0.86 67 | 0.64 4 | 0.64 18 | 0.89 23 | 0.60 21 | 1.21 54 | 1.51 32 | 0.93 28 | 1.00 55 | 1.18 79 | 1.27 68 | 1.18 85 | 1.27 84 | 1.69 99 | 1.06 16 | 1.40 20 | 1.07 36 | 0.96 99 | 2.04 101 | 0.78 74 | 0.99 82 | 1.48 87 | 0.63 4 |
EPPM w/o HM [88] | 53.5 | 0.60 12 | 0.80 34 | 0.64 4 | 0.67 35 | 0.95 41 | 0.59 4 | 2.36 117 | 3.43 116 | 2.13 94 | 1.01 64 | 1.22 89 | 1.23 32 | 1.00 11 | 0.99 13 | 1.15 12 | 1.14 56 | 1.61 59 | 1.09 53 | 1.18 122 | 2.63 122 | 1.25 121 | 0.87 33 | 1.27 36 | 0.63 4 |
Sparse-NonSparse [56] | 53.5 | 0.61 21 | 0.79 28 | 0.64 4 | 0.65 26 | 0.89 23 | 0.61 33 | 1.23 60 | 2.49 91 | 1.38 63 | 0.94 5 | 1.03 14 | 1.20 4 | 1.18 85 | 1.28 91 | 1.58 81 | 1.18 75 | 1.73 79 | 1.09 53 | 0.95 96 | 2.00 97 | 0.79 79 | 0.99 82 | 1.49 91 | 0.63 4 |
Brox et al. [5] | 54.0 | 0.67 100 | 1.04 114 | 0.65 41 | 0.72 56 | 1.02 58 | 0.63 51 | 0.96 18 | 1.34 12 | 0.83 17 | 0.98 37 | 0.99 3 | 1.24 38 | 1.02 16 | 1.02 20 | 1.15 12 | 1.20 87 | 1.78 92 | 1.11 75 | 1.67 128 | 3.86 128 | 2.48 130 | 0.86 31 | 1.26 32 | 0.62 1 |
DF-Auto [115] | 54.7 | 0.65 86 | 0.96 99 | 0.66 72 | 0.78 74 | 1.08 75 | 0.70 89 | 1.02 24 | 1.67 43 | 0.87 21 | 1.00 55 | 1.12 55 | 1.25 45 | 0.99 5 | 0.97 9 | 1.15 12 | 1.08 24 | 1.45 28 | 1.04 7 | 0.90 86 | 1.89 88 | 1.09 117 | 0.94 56 | 1.40 62 | 0.65 80 |
OFH [38] | 55.1 | 0.62 40 | 0.83 53 | 0.65 41 | 0.76 66 | 1.05 69 | 0.63 51 | 1.14 43 | 1.95 64 | 0.89 22 | 0.95 14 | 1.06 27 | 1.21 8 | 1.15 71 | 1.24 73 | 1.54 77 | 1.12 44 | 1.56 46 | 1.10 63 | 1.00 105 | 1.97 95 | 1.11 118 | 0.95 63 | 1.41 66 | 0.63 4 |
DPOF [18] | 55.2 | 0.66 93 | 1.05 119 | 0.68 101 | 0.61 6 | 0.80 6 | 0.59 4 | 1.60 92 | 1.55 36 | 2.16 95 | 1.05 85 | 1.33 111 | 1.28 79 | 1.05 33 | 1.07 36 | 1.14 2 | 1.08 24 | 1.47 33 | 1.04 7 | 0.77 47 | 1.49 47 | 0.69 18 | 1.04 101 | 1.56 102 | 0.64 48 |
LDOF [28] | 55.2 | 0.66 93 | 0.94 92 | 0.67 92 | 0.79 77 | 0.99 47 | 0.82 119 | 1.15 46 | 1.37 17 | 1.14 46 | 0.98 37 | 1.08 33 | 1.24 38 | 1.00 11 | 0.98 10 | 1.15 12 | 1.06 16 | 1.39 17 | 1.04 7 | 1.14 118 | 2.51 119 | 1.27 122 | 0.83 22 | 1.19 21 | 0.67 114 |
Modified CLG [34] | 55.3 | 0.61 21 | 0.77 14 | 0.66 72 | 0.90 115 | 1.16 103 | 0.80 114 | 1.26 67 | 1.67 43 | 1.61 73 | 1.01 64 | 1.10 41 | 1.27 68 | 1.03 20 | 1.03 22 | 1.15 12 | 1.14 56 | 1.61 59 | 1.09 53 | 0.65 6 | 1.13 6 | 0.67 10 | 1.09 121 | 1.64 120 | 0.64 48 |
Kuang [131] | 55.3 | 0.62 40 | 0.88 81 | 0.64 4 | 0.66 28 | 0.93 38 | 0.60 21 | 1.54 88 | 2.19 78 | 1.41 64 | 1.00 55 | 1.18 79 | 1.25 45 | 1.12 65 | 1.18 65 | 1.52 75 | 1.08 24 | 1.45 28 | 1.06 30 | 0.96 99 | 2.03 100 | 1.05 115 | 0.92 49 | 1.37 53 | 0.63 4 |
RNLOD-Flow [121] | 55.8 | 0.61 21 | 0.79 28 | 0.64 4 | 0.72 56 | 1.02 58 | 0.62 40 | 1.25 65 | 2.31 85 | 1.37 59 | 0.94 5 | 1.00 5 | 1.23 32 | 1.20 102 | 1.30 103 | 1.68 96 | 1.19 83 | 1.77 88 | 1.10 63 | 0.72 26 | 1.33 24 | 0.74 50 | 1.09 121 | 1.65 122 | 0.63 4 |
AGIF+OF [85] | 56.5 | 0.61 21 | 0.78 23 | 0.64 4 | 0.67 35 | 0.92 34 | 0.61 33 | 1.17 51 | 2.07 75 | 1.37 59 | 0.97 25 | 1.06 27 | 1.25 45 | 1.20 102 | 1.30 103 | 1.69 99 | 1.18 75 | 1.75 83 | 1.07 36 | 0.75 40 | 1.44 44 | 0.72 38 | 1.13 127 | 1.72 128 | 0.64 48 |
F-TV-L1 [15] | 57.5 | 0.67 100 | 0.99 103 | 0.68 101 | 0.85 96 | 1.15 97 | 0.70 89 | 0.97 20 | 1.51 32 | 0.86 19 | 1.01 64 | 1.08 33 | 1.28 79 | 1.03 20 | 1.04 25 | 1.14 2 | 1.04 9 | 1.31 9 | 1.06 30 | 0.85 69 | 1.73 72 | 0.79 79 | 1.07 114 | 1.61 114 | 0.63 4 |
Ad-TV-NDC [36] | 57.5 | 0.75 121 | 1.01 106 | 0.76 124 | 0.95 121 | 1.19 111 | 0.82 119 | 0.90 8 | 1.44 23 | 0.78 13 | 1.09 96 | 1.13 60 | 1.32 102 | 1.03 20 | 1.03 22 | 1.16 32 | 1.10 35 | 1.45 28 | 1.10 63 | 0.65 6 | 1.15 7 | 0.66 5 | 0.94 56 | 1.38 55 | 0.64 48 |
SVFilterOh [111] | 57.8 | 0.62 40 | 0.81 42 | 0.65 41 | 0.64 18 | 0.87 16 | 0.60 21 | 2.19 114 | 4.17 124 | 2.76 113 | 0.97 25 | 1.08 33 | 1.26 58 | 1.20 102 | 1.29 99 | 1.71 124 | 1.15 64 | 1.64 65 | 1.09 53 | 0.73 30 | 1.38 34 | 0.66 5 | 0.84 26 | 1.21 26 | 0.67 114 |
SRR-TVOF-NL [91] | 58.1 | 0.63 69 | 0.83 53 | 0.65 41 | 0.72 56 | 1.01 54 | 0.64 60 | 1.84 99 | 3.78 118 | 2.38 103 | 0.96 19 | 1.08 33 | 1.21 8 | 1.20 102 | 1.30 103 | 1.69 99 | 1.14 56 | 1.65 69 | 1.05 18 | 0.73 30 | 1.40 37 | 0.73 43 | 0.88 37 | 1.29 39 | 0.64 48 |
Classic++ [32] | 58.2 | 0.62 40 | 0.80 34 | 0.66 72 | 0.78 74 | 1.10 79 | 0.66 72 | 0.93 17 | 1.36 16 | 0.75 9 | 1.04 82 | 1.12 55 | 1.28 79 | 1.08 50 | 1.11 50 | 1.18 42 | 1.18 75 | 1.69 74 | 1.10 63 | 0.89 79 | 1.86 84 | 0.72 38 | 0.99 82 | 1.47 84 | 0.64 48 |
FlowNet2 [122] | 58.4 | 0.71 112 | 0.99 103 | 0.68 101 | 0.71 51 | 0.98 42 | 0.67 77 | 1.27 69 | 2.29 84 | 1.41 64 | 0.99 48 | 1.18 79 | 1.26 58 | 1.05 33 | 1.07 36 | 1.22 53 | 1.09 32 | 1.47 33 | 1.06 30 | 0.81 58 | 1.61 60 | 0.77 68 | 0.80 15 | 1.15 15 | 0.65 80 |
FlowNetS+ft+v [112] | 58.5 | 0.67 100 | 1.01 106 | 0.67 92 | 0.89 111 | 1.17 106 | 0.82 119 | 0.88 6 | 1.18 3 | 0.77 10 | 0.96 19 | 1.03 14 | 1.26 58 | 1.20 102 | 1.30 103 | 1.70 116 | 1.05 10 | 1.35 12 | 1.06 30 | 0.82 61 | 1.65 66 | 0.70 23 | 0.95 63 | 1.42 69 | 0.63 4 |
S2F-IF [123] | 58.6 | 0.62 40 | 0.86 67 | 0.64 4 | 0.63 11 | 0.86 13 | 0.59 4 | 1.21 54 | 2.53 96 | 1.50 69 | 0.97 25 | 1.13 60 | 1.22 22 | 1.18 85 | 1.28 91 | 1.70 116 | 1.11 37 | 1.54 39 | 1.10 63 | 0.82 61 | 1.66 67 | 0.76 63 | 1.08 119 | 1.64 120 | 0.65 80 |
Ramp [62] | 59.1 | 0.62 40 | 0.82 49 | 0.65 41 | 0.66 28 | 0.90 28 | 0.61 33 | 1.59 90 | 3.35 114 | 2.06 92 | 0.95 14 | 1.05 25 | 1.21 8 | 1.16 72 | 1.25 76 | 1.58 81 | 1.22 95 | 1.85 101 | 1.12 80 | 0.85 69 | 1.73 72 | 0.70 23 | 0.96 69 | 1.43 70 | 0.64 48 |
Sparse Occlusion [54] | 59.3 | 0.63 69 | 0.87 75 | 0.65 41 | 0.77 70 | 1.11 84 | 0.63 51 | 0.91 12 | 1.45 26 | 0.86 19 | 0.96 19 | 1.08 33 | 1.21 8 | 1.21 115 | 1.32 114 | 1.69 99 | 1.14 56 | 1.63 64 | 1.12 80 | 0.86 72 | 1.77 76 | 0.71 33 | 1.04 101 | 1.56 102 | 0.63 4 |
CRTflow [80] | 60.8 | 0.64 82 | 0.89 85 | 0.67 92 | 0.83 89 | 1.16 103 | 0.69 84 | 1.12 42 | 1.66 42 | 1.02 34 | 1.00 55 | 1.10 41 | 1.28 79 | 1.18 85 | 1.27 84 | 1.69 99 | 1.05 10 | 1.33 10 | 1.05 18 | 0.93 92 | 1.95 93 | 0.69 18 | 0.94 56 | 1.40 62 | 0.63 4 |
Local-TV-L1 [65] | 61.0 | 0.65 86 | 0.83 53 | 0.69 110 | 0.88 107 | 1.15 97 | 0.76 107 | 0.87 4 | 1.27 7 | 0.71 7 | 1.01 64 | 1.06 27 | 1.31 97 | 1.13 67 | 1.19 66 | 1.35 67 | 1.14 56 | 1.48 35 | 1.13 89 | 0.81 58 | 1.63 63 | 0.73 43 | 0.85 28 | 1.23 28 | 0.66 97 |
FlowFields+ [130] | 62.1 | 0.62 40 | 0.88 81 | 0.65 41 | 0.63 11 | 0.86 13 | 0.59 4 | 1.66 95 | 3.28 111 | 2.17 97 | 1.02 71 | 1.25 97 | 1.24 38 | 1.18 85 | 1.27 84 | 1.69 99 | 1.16 68 | 1.68 72 | 1.10 63 | 0.88 78 | 1.83 79 | 0.80 84 | 0.88 37 | 1.28 38 | 0.63 4 |
Fusion [6] | 62.9 | 0.64 82 | 0.94 92 | 0.65 41 | 0.70 45 | 0.98 42 | 0.61 33 | 1.35 78 | 1.48 27 | 1.70 80 | 1.06 91 | 1.26 99 | 1.22 22 | 1.12 65 | 1.20 68 | 1.22 53 | 1.29 114 | 2.07 119 | 1.19 112 | 0.78 51 | 1.54 53 | 0.72 38 | 0.85 28 | 1.24 29 | 0.64 48 |
Classic+NL [31] | 63.0 | 0.62 40 | 0.82 49 | 0.65 41 | 0.67 35 | 0.92 34 | 0.62 40 | 1.56 89 | 3.23 108 | 1.95 87 | 0.97 25 | 1.11 46 | 1.25 45 | 1.17 79 | 1.27 84 | 1.54 77 | 1.17 72 | 1.71 77 | 1.09 53 | 0.92 88 | 1.92 89 | 0.77 68 | 1.00 89 | 1.50 92 | 0.63 4 |
FESL [72] | 64.0 | 0.61 21 | 0.77 14 | 0.64 4 | 0.66 28 | 0.91 31 | 0.60 21 | 1.18 53 | 2.09 76 | 1.11 43 | 0.99 48 | 1.10 41 | 1.27 68 | 1.21 115 | 1.32 114 | 1.69 99 | 1.20 87 | 1.81 99 | 1.12 80 | 0.90 86 | 1.87 86 | 0.74 50 | 1.06 111 | 1.60 112 | 0.64 48 |
AdaConv-v1 [126] | 64.4 | 0.73 119 | 1.12 126 | 0.72 121 | 0.85 96 | 1.03 61 | 0.91 128 | 1.31 70 | 1.34 12 | 1.31 55 | 1.22 117 | 1.28 104 | 1.52 118 | 1.00 11 | 1.00 15 | 1.13 1 | 1.01 3 | 1.19 3 | 1.04 7 | 0.86 72 | 1.57 56 | 1.01 112 | 0.77 5 | 1.08 5 | 0.72 129 |
ACK-Prior [27] | 64.7 | 0.61 21 | 0.83 53 | 0.64 4 | 0.69 42 | 0.98 42 | 0.60 21 | 2.41 118 | 1.84 57 | 3.18 119 | 1.02 71 | 1.12 55 | 1.27 68 | 1.22 120 | 1.32 114 | 1.72 126 | 1.17 72 | 1.64 65 | 1.12 80 | 0.79 54 | 1.54 53 | 0.78 74 | 0.83 22 | 1.19 21 | 0.65 80 |
EpicFlow [102] | 65.1 | 0.62 40 | 0.86 67 | 0.64 4 | 0.70 45 | 1.00 50 | 0.62 40 | 1.06 32 | 1.31 9 | 0.82 15 | 1.09 96 | 1.42 118 | 1.31 97 | 1.18 85 | 1.27 84 | 1.69 99 | 1.10 35 | 1.51 37 | 1.09 53 | 1.00 105 | 2.13 106 | 0.85 94 | 1.04 101 | 1.56 102 | 0.64 48 |
Classic+CPF [83] | 65.2 | 0.61 21 | 0.79 28 | 0.64 4 | 0.68 41 | 0.93 38 | 0.62 40 | 1.21 54 | 2.25 82 | 1.22 51 | 0.94 5 | 1.02 10 | 1.19 1 | 1.22 120 | 1.34 122 | 1.69 99 | 1.23 101 | 1.88 109 | 1.11 75 | 0.92 88 | 1.92 89 | 0.74 50 | 1.14 128 | 1.73 129 | 0.65 80 |
Black & Anandan [4] | 65.9 | 0.68 107 | 0.96 99 | 0.69 110 | 0.94 119 | 1.21 114 | 0.76 107 | 2.33 116 | 1.75 50 | 2.52 108 | 1.08 93 | 1.15 70 | 1.25 45 | 1.04 27 | 1.05 27 | 1.16 32 | 1.11 37 | 1.54 39 | 1.07 36 | 0.73 30 | 1.37 31 | 0.70 23 | 0.87 33 | 1.26 32 | 0.66 97 |
Efficient-NL [60] | 66.6 | 0.61 21 | 0.77 14 | 0.64 4 | 0.71 51 | 0.99 47 | 0.62 40 | 2.03 107 | 1.80 56 | 2.75 112 | 0.97 25 | 1.11 46 | 1.22 22 | 1.18 85 | 1.28 91 | 1.64 87 | 1.19 83 | 1.77 88 | 1.09 53 | 0.96 99 | 2.04 101 | 0.78 74 | 1.10 123 | 1.65 122 | 0.64 48 |
Adaptive [20] | 67.4 | 0.64 82 | 0.91 87 | 0.66 72 | 0.88 107 | 1.22 117 | 0.71 97 | 1.06 32 | 1.76 52 | 1.05 36 | 1.03 75 | 1.17 75 | 1.33 106 | 1.09 52 | 1.12 52 | 1.15 12 | 1.20 87 | 1.78 92 | 1.14 94 | 0.86 72 | 1.76 75 | 0.71 33 | 0.93 51 | 1.38 55 | 0.63 4 |
NL-TV-NCC [25] | 67.5 | 0.63 69 | 0.84 60 | 0.65 41 | 0.77 70 | 1.10 79 | 0.64 60 | 1.02 24 | 1.71 47 | 0.90 23 | 1.07 92 | 1.30 108 | 1.32 102 | 1.07 45 | 1.06 30 | 1.38 70 | 1.25 108 | 1.91 111 | 1.14 94 | 0.75 40 | 1.40 37 | 0.74 50 | 0.99 82 | 1.46 81 | 0.66 97 |
Occlusion-TV-L1 [63] | 68.2 | 0.62 40 | 0.86 67 | 0.66 72 | 0.85 96 | 1.20 112 | 0.68 80 | 0.92 14 | 1.58 39 | 0.90 23 | 1.13 104 | 1.43 119 | 1.30 94 | 1.04 27 | 1.06 30 | 1.15 12 | 1.20 87 | 1.78 92 | 1.15 99 | 0.89 79 | 1.54 53 | 0.83 92 | 1.04 101 | 1.56 102 | 0.63 4 |
S2D-Matching [84] | 68.3 | 0.62 40 | 0.83 53 | 0.65 41 | 0.74 63 | 1.05 69 | 0.64 60 | 1.42 81 | 2.75 101 | 1.64 76 | 0.97 25 | 1.11 46 | 1.23 32 | 1.14 69 | 1.21 69 | 1.43 72 | 1.29 114 | 2.04 118 | 1.17 108 | 0.78 51 | 1.52 51 | 0.74 50 | 1.04 101 | 1.56 102 | 0.64 48 |
Nguyen [33] | 69.1 | 0.71 112 | 1.01 106 | 0.71 115 | 0.96 123 | 1.20 112 | 0.79 113 | 1.05 30 | 1.75 50 | 0.91 26 | 1.09 96 | 1.16 73 | 1.31 97 | 1.04 27 | 1.06 30 | 1.15 12 | 1.15 64 | 1.67 71 | 1.08 46 | 0.93 92 | 1.96 94 | 0.80 84 | 0.89 40 | 1.30 42 | 0.63 4 |
TF+OM [100] | 69.5 | 0.63 69 | 0.87 75 | 0.66 72 | 0.67 35 | 0.93 38 | 0.65 67 | 1.06 32 | 1.95 64 | 1.05 36 | 1.04 82 | 1.22 89 | 1.26 58 | 1.16 72 | 1.24 73 | 1.58 81 | 1.16 68 | 1.68 72 | 1.09 53 | 0.92 88 | 1.92 89 | 0.86 96 | 0.97 74 | 1.43 70 | 0.67 114 |
ROF-ND [107] | 69.5 | 0.62 40 | 0.76 11 | 0.64 4 | 0.79 77 | 1.08 75 | 0.75 105 | 1.22 59 | 2.44 90 | 1.37 59 | 1.10 100 | 1.38 114 | 1.26 58 | 1.17 79 | 1.26 79 | 1.69 99 | 1.23 101 | 1.89 110 | 1.19 112 | 0.77 47 | 1.49 47 | 0.73 43 | 0.98 77 | 1.45 77 | 0.63 4 |
AggregFlow [97] | 69.6 | 0.67 100 | 1.03 111 | 0.65 41 | 0.69 42 | 0.98 42 | 0.63 51 | 1.39 79 | 2.69 99 | 1.74 83 | 1.01 64 | 1.21 87 | 1.24 38 | 1.02 16 | 1.01 17 | 1.19 45 | 1.07 19 | 1.37 14 | 1.07 36 | 1.30 126 | 2.90 126 | 1.75 127 | 1.04 101 | 1.57 109 | 0.66 97 |
Complementary OF [21] | 69.7 | 0.66 93 | 1.03 111 | 0.64 4 | 0.70 45 | 1.01 54 | 0.63 51 | 3.10 123 | 2.52 94 | 3.34 122 | 0.98 37 | 1.13 60 | 1.22 22 | 1.16 72 | 1.25 76 | 1.59 84 | 1.13 53 | 1.59 55 | 1.10 63 | 0.93 92 | 1.87 86 | 0.97 109 | 0.94 56 | 1.40 62 | 0.64 48 |
Bartels [41] | 70.0 | 0.66 93 | 0.94 92 | 0.68 101 | 0.76 66 | 1.10 79 | 0.70 89 | 1.03 27 | 1.58 39 | 1.03 35 | 1.09 96 | 1.24 95 | 1.39 112 | 1.03 20 | 0.99 13 | 1.25 58 | 1.33 120 | 1.87 105 | 1.25 121 | 0.71 21 | 1.31 19 | 0.78 74 | 0.90 44 | 1.32 46 | 0.67 114 |
Filter Flow [19] | 70.2 | 0.67 100 | 0.97 102 | 0.68 101 | 0.89 111 | 1.17 106 | 0.76 107 | 1.14 43 | 2.02 68 | 1.24 52 | 1.10 100 | 1.16 73 | 1.34 109 | 1.02 16 | 1.01 17 | 1.17 40 | 1.14 56 | 1.59 55 | 1.09 53 | 0.77 47 | 1.51 49 | 0.77 68 | 0.94 56 | 1.39 59 | 0.66 97 |
ProbFlowFields [128] | 70.4 | 0.62 40 | 0.88 81 | 0.64 4 | 0.63 11 | 0.87 16 | 0.59 4 | 1.95 105 | 3.65 117 | 2.45 107 | 0.99 48 | 1.17 75 | 1.26 58 | 1.17 79 | 1.26 79 | 1.70 116 | 1.23 101 | 1.87 105 | 1.17 108 | 0.99 104 | 2.13 106 | 1.39 124 | 0.84 26 | 1.22 27 | 0.64 48 |
FC-2Layers-FF [74] | 70.7 | 0.62 40 | 0.81 42 | 0.65 41 | 0.60 4 | 0.77 3 | 0.60 21 | 1.51 85 | 3.18 106 | 2.01 89 | 1.00 55 | 1.20 85 | 1.21 8 | 1.21 115 | 1.32 114 | 1.68 96 | 1.22 95 | 1.85 101 | 1.12 80 | 1.00 105 | 2.14 108 | 0.81 87 | 0.99 82 | 1.48 87 | 0.64 48 |
TriFlow [95] | 71.4 | 0.67 100 | 1.04 114 | 0.66 72 | 0.79 77 | 1.09 77 | 0.70 89 | 1.07 35 | 2.04 72 | 1.13 45 | 1.02 71 | 1.14 65 | 1.26 58 | 1.20 102 | 1.30 103 | 1.70 116 | 1.12 44 | 1.55 43 | 1.05 18 | 0.78 51 | 1.53 52 | 0.75 60 | 1.04 101 | 1.55 101 | 0.64 48 |
CNN-flow-warp+ref [117] | 71.4 | 0.62 40 | 0.82 49 | 0.66 72 | 0.79 77 | 1.10 79 | 0.69 84 | 1.24 63 | 1.69 45 | 1.31 55 | 1.15 110 | 1.23 93 | 1.38 111 | 1.19 99 | 1.29 99 | 1.70 116 | 1.12 44 | 1.57 50 | 1.12 80 | 0.89 79 | 1.84 80 | 0.75 60 | 0.95 63 | 1.40 62 | 0.63 4 |
Horn & Schunck [3] | 71.5 | 0.66 93 | 0.93 89 | 0.67 92 | 0.96 123 | 1.22 117 | 0.82 119 | 1.91 103 | 1.72 49 | 2.27 100 | 1.14 108 | 1.24 95 | 1.30 94 | 1.04 27 | 1.06 30 | 1.16 32 | 1.08 24 | 1.44 25 | 1.05 18 | 0.75 40 | 1.43 42 | 0.74 50 | 1.03 99 | 1.53 99 | 0.64 48 |
RFlow [90] | 71.6 | 0.61 21 | 0.80 34 | 0.65 41 | 0.81 85 | 1.13 91 | 0.65 67 | 1.59 90 | 3.28 111 | 2.02 90 | 1.05 85 | 1.25 97 | 1.28 79 | 1.06 41 | 1.09 44 | 1.15 12 | 1.17 72 | 1.73 79 | 1.08 46 | 0.89 79 | 1.86 84 | 0.74 50 | 1.06 111 | 1.60 112 | 0.66 97 |
TI-DOFE [24] | 72.3 | 0.74 120 | 0.99 103 | 0.76 124 | 1.03 127 | 1.27 127 | 0.86 125 | 1.02 24 | 1.57 38 | 0.96 30 | 1.20 115 | 1.29 107 | 1.32 102 | 1.04 27 | 1.06 30 | 1.15 12 | 1.15 64 | 1.64 65 | 1.08 46 | 0.71 21 | 1.31 19 | 0.74 50 | 1.01 95 | 1.47 84 | 0.65 80 |
Steered-L1 [118] | 72.9 | 0.61 21 | 0.87 75 | 0.64 4 | 0.71 51 | 1.02 58 | 0.64 60 | 3.16 124 | 4.58 127 | 4.16 127 | 1.16 111 | 1.36 113 | 1.40 113 | 1.09 52 | 1.13 55 | 1.24 57 | 1.13 53 | 1.59 55 | 1.10 63 | 0.89 79 | 1.84 80 | 0.81 87 | 1.00 89 | 1.50 92 | 0.63 4 |
BlockOverlap [61] | 73.2 | 0.66 93 | 0.87 75 | 0.70 113 | 0.86 101 | 1.13 91 | 0.77 111 | 1.34 75 | 1.49 28 | 1.70 80 | 1.13 104 | 1.15 70 | 1.57 122 | 1.07 45 | 1.07 36 | 1.20 50 | 1.20 87 | 1.72 78 | 1.16 105 | 0.76 45 | 1.40 37 | 0.83 92 | 0.75 2 | 1.05 2 | 0.67 114 |
TriangleFlow [30] | 75.2 | 0.64 82 | 0.87 75 | 0.66 72 | 0.79 77 | 1.11 84 | 0.64 60 | 1.23 60 | 2.03 71 | 1.36 58 | 1.03 75 | 1.22 89 | 1.29 90 | 1.07 45 | 1.10 48 | 1.14 2 | 1.19 83 | 1.77 88 | 1.11 75 | 1.12 116 | 2.47 117 | 0.93 104 | 1.01 95 | 1.50 92 | 0.64 48 |
HBM-GC [105] | 75.4 | 0.63 69 | 0.81 42 | 0.67 92 | 0.72 56 | 1.04 66 | 0.62 40 | 1.31 70 | 1.54 35 | 1.65 77 | 1.02 71 | 1.18 79 | 1.27 68 | 1.23 123 | 1.34 122 | 1.73 129 | 1.39 124 | 2.27 124 | 1.25 121 | 0.81 58 | 1.62 62 | 0.72 38 | 0.80 15 | 1.15 15 | 0.67 114 |
2D-CLG [1] | 78.8 | 0.65 86 | 0.87 75 | 0.68 101 | 0.91 116 | 1.15 97 | 0.80 114 | 1.53 87 | 1.32 10 | 1.83 85 | 1.08 93 | 1.11 46 | 1.32 102 | 1.24 124 | 1.37 124 | 1.72 126 | 1.11 37 | 1.54 39 | 1.12 80 | 0.86 72 | 1.77 76 | 0.73 43 | 0.98 77 | 1.45 77 | 0.63 4 |
IAOF [50] | 79.9 | 0.72 116 | 1.03 111 | 0.71 115 | 1.06 129 | 1.33 130 | 0.80 114 | 1.94 104 | 3.23 108 | 2.43 105 | 1.12 102 | 1.14 65 | 1.36 110 | 1.05 33 | 1.06 30 | 1.15 12 | 1.15 64 | 1.66 70 | 1.07 36 | 0.85 69 | 1.74 74 | 0.71 33 | 0.96 69 | 1.43 70 | 0.64 48 |
Correlation Flow [75] | 80.7 | 0.61 21 | 0.82 49 | 0.64 4 | 0.81 85 | 1.15 97 | 0.65 67 | 1.17 51 | 2.22 80 | 1.37 59 | 1.00 55 | 1.17 75 | 1.25 45 | 1.11 61 | 1.14 56 | 1.30 62 | 1.46 128 | 2.44 128 | 1.37 128 | 1.06 114 | 2.29 115 | 0.93 104 | 1.07 114 | 1.61 114 | 0.68 125 |
BriefMatch [124] | 81.2 | 0.63 69 | 0.83 53 | 0.66 72 | 0.72 56 | 1.03 61 | 0.70 89 | 2.06 109 | 1.62 41 | 2.76 113 | 1.26 119 | 1.28 104 | 1.70 125 | 1.05 33 | 1.05 27 | 1.20 50 | 1.25 108 | 1.78 92 | 1.19 112 | 0.84 67 | 1.68 69 | 1.13 120 | 1.03 99 | 1.24 29 | 1.49 131 |
Shiralkar [42] | 81.2 | 0.65 86 | 0.94 92 | 0.65 41 | 0.85 96 | 1.14 95 | 0.67 77 | 1.52 86 | 1.84 57 | 1.71 82 | 1.23 118 | 1.54 125 | 1.29 90 | 1.09 52 | 1.14 56 | 1.21 52 | 1.20 87 | 1.77 88 | 1.11 75 | 0.97 102 | 2.04 101 | 0.77 68 | 1.05 110 | 1.58 110 | 0.63 4 |
IAOF2 [51] | 81.9 | 0.68 107 | 0.96 99 | 0.68 101 | 0.87 104 | 1.21 114 | 0.71 97 | 1.09 38 | 1.95 64 | 1.10 42 | 1.03 75 | 1.15 70 | 1.27 68 | 1.18 85 | 1.28 91 | 1.49 74 | 1.22 95 | 1.86 103 | 1.13 89 | 0.79 54 | 1.57 56 | 0.76 63 | 1.02 98 | 1.53 99 | 0.65 80 |
GraphCuts [14] | 82.3 | 0.70 110 | 1.04 114 | 0.67 92 | 0.74 63 | 1.00 50 | 0.70 89 | 2.29 115 | 1.44 23 | 2.80 117 | 1.08 93 | 1.21 87 | 1.30 94 | 1.16 72 | 1.24 73 | 1.46 73 | 1.12 44 | 1.59 55 | 1.05 18 | 0.97 102 | 2.07 104 | 0.97 109 | 1.07 114 | 1.62 117 | 0.64 48 |
LocallyOriented [52] | 83.1 | 0.65 86 | 0.89 85 | 0.67 92 | 0.86 101 | 1.17 106 | 0.69 84 | 1.85 100 | 2.79 102 | 2.37 102 | 1.19 113 | 1.50 122 | 1.25 45 | 1.08 50 | 1.12 52 | 1.19 45 | 1.16 68 | 1.62 62 | 1.12 80 | 0.82 61 | 1.61 60 | 0.79 79 | 1.12 125 | 1.70 127 | 0.64 48 |
TV-L1-improved [17] | 83.9 | 0.63 69 | 0.85 66 | 0.66 72 | 0.88 107 | 1.22 117 | 0.72 101 | 1.98 106 | 1.55 36 | 2.68 110 | 1.00 55 | 1.07 30 | 1.27 68 | 1.11 61 | 1.16 63 | 1.15 12 | 1.23 101 | 1.87 105 | 1.14 94 | 1.05 112 | 2.28 114 | 0.87 97 | 1.04 101 | 1.56 102 | 0.67 114 |
Aniso-Texture [82] | 84.2 | 0.61 21 | 0.80 34 | 0.64 4 | 0.86 101 | 1.21 114 | 0.71 97 | 1.23 60 | 2.02 68 | 1.25 53 | 1.61 127 | 2.05 127 | 2.27 128 | 1.21 115 | 1.32 114 | 1.71 124 | 1.42 126 | 2.37 126 | 1.31 127 | 0.87 77 | 1.79 78 | 0.73 43 | 0.94 56 | 1.37 53 | 0.64 48 |
StereoOF-V1MT [119] | 85.2 | 0.65 86 | 0.93 89 | 0.65 41 | 0.80 83 | 1.11 84 | 0.65 67 | 1.80 97 | 1.51 32 | 2.20 99 | 1.34 123 | 1.49 121 | 1.58 124 | 1.20 102 | 1.30 103 | 1.65 88 | 1.24 106 | 1.69 74 | 1.19 112 | 1.02 109 | 2.19 111 | 0.89 101 | 0.90 44 | 1.31 44 | 0.63 4 |
SimpleFlow [49] | 88.0 | 0.62 40 | 0.84 60 | 0.65 41 | 0.76 66 | 1.06 71 | 0.64 60 | 3.87 128 | 4.61 128 | 4.32 128 | 1.03 75 | 1.20 85 | 1.29 90 | 1.20 102 | 1.30 103 | 1.65 88 | 1.34 121 | 2.18 121 | 1.16 105 | 1.45 127 | 3.30 127 | 1.60 126 | 0.94 56 | 1.39 59 | 0.63 4 |
HBpMotionGpu [43] | 90.1 | 0.71 112 | 1.04 114 | 0.71 115 | 0.94 119 | 1.25 125 | 0.80 114 | 1.33 74 | 2.51 93 | 1.48 68 | 1.12 102 | 1.39 117 | 1.28 79 | 1.34 128 | 1.52 128 | 2.35 131 | 1.29 114 | 2.02 117 | 1.20 117 | 0.69 15 | 1.27 16 | 0.66 5 | 0.89 40 | 1.29 39 | 0.65 80 |
Rannacher [23] | 91.6 | 0.65 86 | 0.95 97 | 0.66 72 | 0.89 111 | 1.24 122 | 0.71 97 | 2.10 110 | 1.78 54 | 2.78 116 | 1.05 85 | 1.27 102 | 1.28 79 | 1.09 52 | 1.14 56 | 1.16 32 | 1.26 111 | 1.95 113 | 1.15 99 | 1.03 111 | 2.22 113 | 0.88 98 | 1.04 101 | 1.56 102 | 0.65 80 |
UnFlow [129] | 91.7 | 0.71 112 | 1.11 123 | 0.68 101 | 0.83 89 | 1.10 79 | 0.69 84 | 1.25 65 | 2.42 89 | 1.34 57 | 1.01 64 | 1.22 89 | 1.22 22 | 1.20 102 | 1.31 113 | 1.65 88 | 1.47 129 | 2.48 129 | 1.30 126 | 0.84 67 | 1.70 71 | 0.73 43 | 1.29 131 | 1.94 131 | 0.66 97 |
Learning Flow [11] | 94.1 | 0.66 93 | 0.94 92 | 0.67 92 | 0.85 96 | 1.18 110 | 0.68 80 | 4.24 131 | 5.56 130 | 4.33 129 | 1.14 108 | 1.26 99 | 1.33 106 | 1.16 72 | 1.22 72 | 1.32 64 | 1.18 75 | 1.70 76 | 1.13 89 | 0.82 61 | 1.63 63 | 0.81 87 | 1.08 119 | 1.62 117 | 0.66 97 |
PGAM+LK [55] | 94.5 | 0.78 124 | 1.09 121 | 0.80 126 | 0.91 116 | 1.17 106 | 0.82 119 | 3.39 126 | 6.37 131 | 4.52 130 | 1.44 125 | 1.47 120 | 1.75 126 | 1.09 52 | 1.11 50 | 1.19 45 | 1.23 101 | 1.78 92 | 1.16 105 | 0.73 30 | 1.37 31 | 0.79 79 | 0.92 49 | 1.35 50 | 0.67 114 |
SegOF [10] | 95.0 | 0.67 100 | 1.01 106 | 0.67 92 | 0.78 74 | 1.06 71 | 0.68 80 | 3.01 122 | 2.80 103 | 3.24 120 | 1.63 128 | 2.62 129 | 1.57 122 | 1.20 102 | 1.30 103 | 1.69 99 | 1.18 75 | 1.74 82 | 1.14 94 | 1.21 123 | 2.70 123 | 1.11 118 | 0.87 33 | 1.26 32 | 0.64 48 |
Adaptive flow [45] | 95.1 | 0.80 125 | 1.06 120 | 0.81 127 | 1.02 126 | 1.27 127 | 0.91 128 | 1.34 75 | 2.01 67 | 1.68 79 | 1.21 116 | 1.30 108 | 1.52 118 | 1.25 125 | 1.37 124 | 1.35 67 | 1.37 123 | 2.22 123 | 1.25 121 | 0.75 40 | 1.43 42 | 0.81 87 | 0.82 17 | 1.18 18 | 0.65 80 |
StereoFlow [44] | 95.9 | 0.85 128 | 1.29 129 | 0.75 122 | 0.95 121 | 1.24 122 | 0.76 107 | 1.10 40 | 1.85 59 | 1.06 39 | 1.05 85 | 1.23 93 | 1.27 68 | 1.44 129 | 1.67 129 | 1.65 88 | 1.43 127 | 2.40 127 | 1.25 121 | 0.89 79 | 1.85 82 | 0.77 68 | 0.98 77 | 1.46 81 | 0.65 80 |
FFV1MT [106] | 97.4 | 0.72 116 | 1.11 123 | 0.69 110 | 0.89 111 | 1.12 88 | 0.77 111 | 1.83 98 | 2.67 98 | 1.88 86 | 1.31 121 | 1.38 114 | 1.51 116 | 1.11 61 | 1.15 60 | 1.25 58 | 1.24 106 | 1.81 99 | 1.15 99 | 1.05 112 | 2.17 110 | 1.04 114 | 0.93 51 | 1.35 50 | 0.69 126 |
HCIC-L [99] | 97.8 | 0.88 130 | 1.10 122 | 0.94 130 | 0.84 92 | 1.03 61 | 0.84 124 | 2.11 111 | 4.36 126 | 2.83 118 | 1.13 104 | 1.35 112 | 1.29 90 | 1.03 20 | 1.01 17 | 1.19 45 | 1.30 118 | 2.01 116 | 1.21 118 | 1.17 121 | 2.57 121 | 1.34 123 | 0.93 51 | 1.35 50 | 0.70 127 |
Dynamic MRF [7] | 97.9 | 0.63 69 | 0.92 88 | 0.65 41 | 0.79 77 | 1.15 97 | 0.67 77 | 1.49 84 | 1.88 61 | 1.67 78 | 1.26 119 | 1.53 124 | 1.56 121 | 1.20 102 | 1.32 114 | 1.69 99 | 1.31 119 | 2.08 120 | 1.23 120 | 1.09 115 | 2.38 116 | 0.94 107 | 1.06 111 | 1.58 110 | 0.65 80 |
Heeger++ [104] | 99.5 | 0.80 125 | 1.35 130 | 0.70 113 | 0.84 92 | 1.09 77 | 0.70 89 | 2.05 108 | 1.88 61 | 2.18 98 | 1.31 121 | 1.38 114 | 1.51 116 | 1.22 120 | 1.33 121 | 1.67 95 | 1.25 108 | 1.75 83 | 1.21 118 | 1.12 116 | 2.09 105 | 0.97 109 | 0.95 63 | 1.39 59 | 0.64 48 |
SILK [79] | 100.7 | 0.69 109 | 0.93 89 | 0.71 115 | 1.01 125 | 1.24 122 | 0.89 126 | 3.96 129 | 3.80 119 | 3.85 125 | 1.16 111 | 1.27 102 | 1.40 113 | 1.11 61 | 1.16 63 | 1.19 45 | 1.29 114 | 1.93 112 | 1.19 112 | 0.74 37 | 1.41 40 | 0.89 101 | 1.12 125 | 1.68 125 | 0.66 97 |
SLK [47] | 101.1 | 0.72 116 | 0.95 97 | 0.75 122 | 0.93 118 | 1.13 91 | 0.81 118 | 2.97 121 | 2.41 88 | 3.25 121 | 1.38 124 | 1.61 126 | 1.53 120 | 1.19 99 | 1.29 99 | 1.26 60 | 1.21 94 | 1.78 92 | 1.14 94 | 1.14 118 | 2.51 119 | 0.91 103 | 0.90 44 | 1.32 46 | 0.66 97 |
FOLKI [16] | 103.7 | 0.82 127 | 1.04 114 | 0.88 128 | 1.03 127 | 1.26 126 | 0.90 127 | 1.74 96 | 2.22 80 | 2.29 101 | 1.48 126 | 1.50 122 | 1.85 127 | 1.10 58 | 1.15 60 | 1.22 53 | 1.41 125 | 2.30 125 | 1.55 130 | 0.83 66 | 1.64 65 | 1.07 116 | 1.00 89 | 1.48 87 | 0.67 114 |
2bit-BM-tele [98] | 104.9 | 0.70 110 | 1.02 110 | 0.71 115 | 0.87 104 | 1.23 121 | 0.75 105 | 2.82 120 | 5.34 129 | 4.12 126 | 1.13 104 | 1.30 108 | 1.43 115 | 1.16 72 | 1.21 69 | 1.42 71 | 1.59 130 | 2.66 130 | 1.52 129 | 1.90 129 | 4.42 130 | 2.38 129 | 0.82 17 | 1.16 17 | 0.71 128 |
SPSA-learn [13] | 107.1 | 0.75 121 | 1.24 128 | 0.68 101 | 0.87 104 | 1.15 97 | 0.74 104 | 3.22 125 | 3.18 106 | 3.46 123 | 1.19 113 | 1.28 104 | 1.33 106 | 1.16 72 | 1.25 76 | 1.28 61 | 1.20 87 | 1.79 98 | 1.17 108 | 2.04 131 | 4.77 131 | 2.66 131 | 1.10 123 | 1.66 124 | 0.66 97 |
GroupFlow [9] | 113.3 | 0.76 123 | 1.20 127 | 0.71 115 | 0.83 89 | 1.12 88 | 0.73 103 | 2.67 119 | 2.82 104 | 2.74 111 | 1.77 129 | 2.21 128 | 2.39 129 | 1.29 126 | 1.43 126 | 1.72 126 | 1.36 122 | 2.21 122 | 1.25 121 | 1.14 118 | 2.49 118 | 0.93 104 | 0.98 77 | 1.46 81 | 0.67 114 |
Pyramid LK [2] | 116.1 | 0.86 129 | 1.11 123 | 0.90 129 | 1.15 130 | 1.29 129 | 0.99 130 | 3.86 127 | 2.26 83 | 3.64 124 | 2.42 131 | 3.60 130 | 2.78 131 | 1.45 130 | 1.68 130 | 1.30 62 | 1.22 95 | 1.62 62 | 1.15 99 | 1.22 124 | 2.72 124 | 0.95 108 | 1.16 129 | 1.76 130 | 0.66 97 |
Periodicity [78] | 128.5 | 1.11 131 | 2.00 131 | 0.94 130 | 1.39 131 | 1.34 131 | 1.12 131 | 4.06 130 | 4.26 125 | 4.55 131 | 2.25 130 | 3.71 131 | 2.59 130 | 1.53 131 | 1.77 131 | 1.68 96 | 1.69 131 | 2.82 131 | 1.60 131 | 1.93 130 | 4.41 129 | 2.10 128 | 1.17 130 | 1.68 125 | 0.87 130 |
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. |