Optical flow evaluation results |
Statistics:
Average
SD
R0.5
R1.0
R2.0
A90
A95
A99
Error type: endpoint angle interpolation normalized interpolation |
R2.0 normalized interpolation error |
avg. |
Mequon (Hidden texture) im0 GT im1 |
Schefflera (Hidden texture) im0 GT im1 |
Urban (Synthetic) im0 GT im1 |
Teddy (Stereo) im0 GT im1 |
Backyard (High-speed camera) im0 GT im1 |
Basketball (High-speed camera) im0 GT im1 |
Dumptruck (High-speed camera) im0 GT im1 |
Evergreen (High-speed camera) im0 GT im1 | ||||||||||||||||
rank | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | |
PMMST [114] | 9.8 | 0.48 4 | 1.85 2 | 0.32 7 | 1.36 20 | 3.31 13 | 0.45 21 | 1.55 2 | 2.89 2 | 0.44 1 | 3.53 4 | 3.89 7 | 7.51 17 | 2.71 6 | 2.65 5 | 4.22 16 | 2.27 3 | 4.31 2 | 2.43 21 | 0.60 6 | 2.35 8 | 0.51 28 | 1.40 12 | 3.51 13 | 0.20 16 |
MDP-Flow2 [68] | 10.1 | 0.45 3 | 1.84 1 | 0.30 5 | 1.24 11 | 3.23 11 | 0.38 1 | 1.57 4 | 3.09 6 | 0.48 3 | 3.54 5 | 3.96 9 | 7.46 11 | 2.68 2 | 2.62 3 | 4.24 20 | 2.61 30 | 5.64 31 | 2.47 32 | 0.59 3 | 2.33 6 | 0.49 13 | 1.40 12 | 3.51 13 | 0.19 7 |
CombBMOF [113] | 15.0 | 0.58 21 | 2.00 6 | 0.38 34 | 1.21 8 | 3.19 10 | 0.39 3 | 1.78 35 | 3.38 14 | 0.75 67 | 3.70 23 | 4.15 22 | 7.59 29 | 2.79 11 | 2.82 10 | 4.16 9 | 2.35 6 | 4.57 7 | 2.34 11 | 0.62 10 | 2.42 11 | 0.48 7 | 1.26 2 | 3.18 3 | 0.16 1 |
NNF-Local [87] | 15.6 | 0.44 2 | 1.93 4 | 0.28 2 | 1.02 1 | 2.54 2 | 0.40 5 | 1.56 3 | 2.97 4 | 0.47 2 | 3.79 38 | 4.58 59 | 7.48 13 | 2.69 4 | 2.61 2 | 4.16 9 | 2.81 50 | 6.47 54 | 2.56 50 | 0.61 7 | 2.42 11 | 0.52 39 | 1.36 5 | 3.42 7 | 0.17 2 |
NN-field [71] | 17.3 | 0.49 6 | 2.19 12 | 0.31 6 | 1.06 2 | 2.64 4 | 0.38 1 | 1.84 51 | 3.00 5 | 0.53 8 | 3.92 57 | 5.00 82 | 7.52 19 | 2.68 2 | 2.60 1 | 4.18 11 | 2.64 38 | 5.76 36 | 2.46 27 | 0.59 3 | 2.30 5 | 0.51 28 | 1.36 5 | 3.41 5 | 0.17 2 |
PH-Flow [101] | 19.1 | 0.56 17 | 2.14 9 | 0.37 26 | 1.13 6 | 2.87 6 | 0.44 18 | 1.60 6 | 3.16 7 | 0.56 14 | 3.52 3 | 3.84 2 | 7.48 13 | 2.69 4 | 2.67 6 | 4.12 6 | 2.86 53 | 7.09 82 | 2.53 45 | 0.63 16 | 2.57 24 | 0.50 21 | 1.43 17 | 3.56 17 | 0.22 40 |
Layers++ [37] | 21.2 | 0.58 21 | 2.27 18 | 0.40 48 | 1.06 2 | 2.51 1 | 0.42 10 | 1.68 12 | 3.47 18 | 0.61 32 | 3.59 10 | 4.04 16 | 7.44 10 | 2.94 35 | 3.06 39 | 4.28 23 | 2.87 57 | 6.80 69 | 2.51 40 | 0.58 2 | 2.21 1 | 0.44 2 | 1.37 7 | 3.41 5 | 0.21 30 |
nLayers [57] | 23.7 | 0.55 16 | 2.17 11 | 0.37 26 | 1.23 9 | 3.14 8 | 0.41 7 | 1.62 8 | 2.92 3 | 0.51 7 | 3.60 11 | 4.00 12 | 7.49 16 | 2.99 50 | 3.10 48 | 4.42 58 | 2.85 52 | 6.56 57 | 2.66 67 | 0.62 10 | 2.28 3 | 0.55 67 | 1.37 7 | 3.43 8 | 0.19 7 |
FMOF [94] | 24.0 | 0.69 65 | 2.59 39 | 0.43 58 | 1.32 17 | 3.32 14 | 0.41 7 | 1.86 54 | 3.57 23 | 0.74 64 | 3.81 44 | 4.40 41 | 7.39 6 | 2.80 13 | 2.81 9 | 4.23 18 | 2.54 18 | 5.43 23 | 2.40 19 | 0.57 1 | 2.23 2 | 0.43 1 | 1.42 16 | 3.54 16 | 0.19 7 |
IROF++ [58] | 24.1 | 0.53 13 | 2.21 14 | 0.33 14 | 1.42 27 | 3.64 25 | 0.44 18 | 1.81 42 | 3.57 23 | 0.67 50 | 3.56 6 | 3.85 5 | 7.56 24 | 2.77 9 | 2.85 13 | 4.11 5 | 2.47 10 | 5.30 17 | 2.36 15 | 0.69 43 | 2.86 46 | 0.51 28 | 1.52 33 | 3.81 33 | 0.24 65 |
COFM [59] | 27.3 | 0.59 24 | 2.35 24 | 0.41 52 | 1.38 23 | 3.49 18 | 0.45 21 | 1.66 10 | 3.36 12 | 0.61 32 | 3.57 8 | 3.96 9 | 7.39 6 | 2.78 10 | 2.85 13 | 4.06 2 | 2.90 62 | 7.73 106 | 2.39 17 | 0.65 23 | 2.60 26 | 0.58 78 | 1.44 19 | 3.61 20 | 0.22 40 |
SepConv-v1 [127] | 29.2 | 0.38 1 | 1.91 3 | 0.18 1 | 1.69 43 | 3.97 38 | 0.64 57 | 1.23 1 | 3.17 8 | 1.03 97 | 4.39 92 | 4.38 37 | 8.06 88 | 2.73 7 | 2.63 4 | 4.29 24 | 1.90 1 | 3.17 1 | 2.06 1 | 0.71 51 | 2.78 40 | 0.67 101 | 0.83 1 | 2.05 1 | 0.17 2 |
HAST [109] | 30.4 | 0.53 13 | 2.07 8 | 0.36 25 | 1.23 9 | 3.15 9 | 0.43 14 | 1.90 58 | 3.84 50 | 0.63 41 | 3.46 1 | 3.70 1 | 7.34 2 | 2.92 29 | 3.07 42 | 4.14 7 | 2.99 77 | 7.61 100 | 2.46 27 | 0.65 23 | 2.71 36 | 0.47 5 | 1.64 67 | 4.09 70 | 0.20 16 |
NNF-EAC [103] | 30.8 | 0.76 83 | 2.27 18 | 0.54 84 | 1.48 30 | 3.71 29 | 0.48 28 | 1.71 17 | 3.30 10 | 0.58 25 | 3.88 54 | 4.24 27 | 8.30 100 | 2.80 13 | 2.82 10 | 4.20 14 | 2.27 3 | 4.36 3 | 2.35 12 | 0.65 23 | 2.64 30 | 0.57 74 | 1.43 17 | 3.59 18 | 0.20 16 |
TV-L1-MCT [64] | 30.8 | 0.65 51 | 2.58 38 | 0.38 34 | 1.74 47 | 4.52 47 | 0.53 40 | 1.77 31 | 3.83 48 | 0.62 37 | 3.63 16 | 4.04 16 | 7.48 13 | 2.96 42 | 3.12 52 | 4.20 14 | 2.55 20 | 5.40 19 | 2.47 32 | 0.65 23 | 2.68 31 | 0.51 28 | 1.39 10 | 3.48 10 | 0.22 40 |
2DHMM-SAS [92] | 33.5 | 0.65 51 | 2.43 30 | 0.40 48 | 2.14 68 | 5.02 66 | 0.58 49 | 1.73 22 | 3.60 28 | 0.62 37 | 3.69 20 | 4.10 19 | 7.61 34 | 2.86 16 | 2.93 23 | 4.32 34 | 2.50 15 | 5.43 23 | 2.32 10 | 0.66 28 | 2.59 25 | 0.53 44 | 1.60 48 | 3.98 50 | 0.20 16 |
LME [70] | 33.9 | 0.48 4 | 1.98 5 | 0.32 7 | 1.46 28 | 3.68 27 | 0.70 71 | 1.77 31 | 4.14 62 | 0.57 20 | 3.64 18 | 4.24 27 | 7.62 38 | 3.14 96 | 3.32 96 | 4.87 115 | 2.66 40 | 6.03 43 | 2.46 27 | 0.59 3 | 2.28 3 | 0.48 7 | 1.44 19 | 3.62 22 | 0.18 5 |
WLIF-Flow [93] | 34.8 | 0.54 15 | 2.22 15 | 0.35 20 | 1.51 34 | 3.90 36 | 0.52 36 | 1.69 14 | 3.36 12 | 0.54 9 | 3.75 27 | 3.97 11 | 8.06 88 | 2.88 18 | 2.92 21 | 4.45 71 | 3.19 94 | 7.37 89 | 2.93 103 | 0.61 7 | 2.35 8 | 0.49 13 | 1.47 27 | 3.68 28 | 0.22 40 |
Sparse-NonSparse [56] | 36.3 | 0.58 21 | 2.36 26 | 0.37 26 | 1.41 26 | 3.60 21 | 0.46 23 | 1.75 27 | 3.62 33 | 0.58 25 | 3.72 25 | 4.16 23 | 7.61 34 | 2.91 26 | 2.99 30 | 4.29 24 | 2.96 73 | 6.88 72 | 2.65 63 | 0.74 65 | 3.05 66 | 0.50 21 | 1.62 58 | 4.02 56 | 0.19 7 |
ComponentFusion [96] | 36.5 | 0.52 11 | 2.30 21 | 0.33 14 | 1.31 16 | 3.50 19 | 0.42 10 | 1.77 31 | 3.60 28 | 0.63 41 | 3.58 9 | 4.06 18 | 7.39 6 | 2.99 50 | 3.06 39 | 4.31 28 | 2.63 35 | 5.74 35 | 2.50 39 | 0.81 86 | 3.51 91 | 0.60 80 | 1.66 75 | 4.15 79 | 0.20 16 |
FlowFields [110] | 38.5 | 0.61 34 | 2.80 62 | 0.35 20 | 1.33 18 | 3.57 20 | 0.43 14 | 1.75 27 | 3.76 44 | 0.56 14 | 3.83 48 | 4.66 63 | 7.83 71 | 2.91 26 | 2.95 26 | 4.32 34 | 2.87 57 | 6.73 64 | 2.65 63 | 0.63 16 | 2.53 18 | 0.53 44 | 1.59 47 | 4.01 54 | 0.22 40 |
DPOF [18] | 38.8 | 0.71 72 | 3.29 102 | 0.45 63 | 1.14 7 | 2.90 7 | 0.41 7 | 2.18 91 | 3.65 36 | 0.92 88 | 3.83 48 | 4.56 58 | 7.74 60 | 2.79 11 | 2.85 13 | 4.19 12 | 2.60 27 | 5.78 37 | 2.31 8 | 0.66 28 | 2.56 21 | 0.52 39 | 1.52 33 | 3.81 33 | 0.21 30 |
FlowFields+ [130] | 38.9 | 0.59 24 | 2.70 52 | 0.32 7 | 1.28 13 | 3.42 16 | 0.43 14 | 1.75 27 | 3.88 52 | 0.57 20 | 3.77 35 | 4.55 56 | 7.69 52 | 3.00 53 | 3.12 52 | 4.40 55 | 2.93 69 | 6.96 76 | 2.68 72 | 0.62 10 | 2.49 15 | 0.53 44 | 1.60 48 | 4.02 56 | 0.20 16 |
DeepFlow2 [108] | 40.3 | 0.66 56 | 2.63 44 | 0.47 69 | 1.95 62 | 4.94 64 | 0.69 66 | 1.79 37 | 4.36 76 | 0.61 32 | 3.90 55 | 4.43 43 | 7.66 45 | 2.88 18 | 2.87 16 | 4.47 78 | 2.35 6 | 4.36 3 | 2.48 34 | 0.64 19 | 2.61 28 | 0.51 28 | 1.46 24 | 3.64 25 | 0.22 40 |
SRR-TVOF-NL [91] | 40.9 | 0.75 82 | 2.87 75 | 0.50 76 | 1.89 57 | 4.81 58 | 0.70 71 | 1.73 22 | 3.57 23 | 0.61 32 | 3.62 14 | 4.01 13 | 7.35 4 | 2.98 46 | 3.19 73 | 4.32 34 | 2.47 10 | 5.81 40 | 2.21 3 | 0.64 19 | 2.55 20 | 0.50 21 | 1.65 71 | 4.13 77 | 0.22 40 |
S2F-IF [123] | 40.9 | 0.59 24 | 2.71 54 | 0.33 14 | 1.26 12 | 3.35 15 | 0.42 10 | 1.75 27 | 3.88 52 | 0.57 20 | 3.68 19 | 4.42 42 | 7.47 12 | 3.00 53 | 3.11 51 | 4.43 61 | 2.90 62 | 6.83 70 | 2.67 70 | 0.67 33 | 2.79 42 | 0.56 70 | 1.60 48 | 4.02 56 | 0.24 65 |
AGIF+OF [85] | 41.0 | 0.62 39 | 2.49 34 | 0.33 14 | 1.50 33 | 3.86 33 | 0.52 36 | 1.77 31 | 3.49 19 | 0.69 53 | 3.60 11 | 3.87 6 | 7.36 5 | 3.10 90 | 3.27 91 | 4.41 57 | 2.99 77 | 7.51 95 | 2.49 37 | 0.68 38 | 2.53 18 | 0.46 4 | 1.63 64 | 4.07 68 | 0.21 30 |
MDP-Flow [26] | 42.2 | 0.52 11 | 2.40 29 | 0.32 7 | 1.34 19 | 3.60 21 | 0.46 23 | 1.68 12 | 3.30 10 | 0.61 32 | 4.10 76 | 4.91 77 | 7.80 66 | 2.88 18 | 2.93 23 | 4.46 73 | 3.32 104 | 8.61 119 | 2.83 93 | 0.66 28 | 2.68 31 | 0.56 70 | 1.51 31 | 3.80 32 | 0.19 7 |
Aniso. Huber-L1 [22] | 42.2 | 0.73 77 | 2.79 59 | 0.53 82 | 2.74 90 | 5.84 88 | 0.78 77 | 1.92 62 | 3.61 30 | 0.73 61 | 3.80 40 | 4.31 33 | 7.66 45 | 2.87 17 | 2.93 23 | 4.31 28 | 2.40 8 | 5.24 14 | 2.28 5 | 0.64 19 | 2.52 17 | 0.49 13 | 1.45 22 | 3.61 20 | 0.26 83 |
LSM [39] | 43.1 | 0.61 34 | 2.46 32 | 0.38 34 | 1.46 28 | 3.69 28 | 0.48 28 | 1.79 37 | 3.86 51 | 0.60 29 | 3.75 27 | 4.19 24 | 7.64 40 | 2.96 42 | 3.08 45 | 4.34 41 | 2.97 75 | 7.04 80 | 2.62 57 | 0.75 68 | 3.08 68 | 0.51 28 | 1.63 64 | 4.07 68 | 0.19 7 |
FC-2Layers-FF [74] | 43.1 | 0.60 31 | 2.53 35 | 0.37 26 | 1.07 4 | 2.58 3 | 0.46 23 | 1.67 11 | 3.59 26 | 0.56 14 | 3.63 16 | 4.03 15 | 7.58 25 | 3.00 53 | 3.15 64 | 4.44 65 | 3.19 94 | 7.90 110 | 2.78 89 | 0.79 80 | 3.22 76 | 0.52 39 | 1.60 48 | 3.94 48 | 0.22 40 |
Ramp [62] | 43.4 | 0.64 47 | 2.59 39 | 0.41 52 | 1.48 30 | 3.80 31 | 0.53 40 | 1.72 21 | 3.61 30 | 0.54 9 | 3.62 14 | 4.02 14 | 7.52 19 | 2.93 32 | 3.02 34 | 4.36 43 | 3.18 92 | 7.72 105 | 2.79 90 | 0.73 59 | 2.96 58 | 0.48 7 | 1.67 80 | 4.15 79 | 0.20 16 |
PMF [73] | 43.4 | 0.49 6 | 2.14 9 | 0.32 7 | 1.39 24 | 3.66 26 | 0.42 10 | 1.94 65 | 4.81 91 | 0.71 57 | 3.69 20 | 4.21 26 | 7.64 40 | 3.01 60 | 3.12 52 | 4.27 22 | 2.93 69 | 5.59 28 | 3.04 108 | 0.68 38 | 2.81 43 | 0.53 44 | 1.68 82 | 4.22 85 | 0.21 30 |
DeepFlow [86] | 44.0 | 0.64 47 | 2.56 37 | 0.42 54 | 2.01 64 | 4.98 65 | 0.82 79 | 1.82 46 | 4.50 80 | 0.63 41 | 4.02 68 | 4.44 45 | 7.78 65 | 2.91 26 | 2.92 21 | 4.53 88 | 2.55 20 | 4.48 6 | 2.76 83 | 0.62 10 | 2.47 14 | 0.49 13 | 1.45 22 | 3.62 22 | 0.22 40 |
ProbFlowFields [128] | 44.0 | 0.62 39 | 2.96 82 | 0.39 39 | 1.28 13 | 3.44 17 | 0.39 3 | 1.65 9 | 3.45 17 | 0.55 12 | 3.85 52 | 4.63 61 | 7.81 67 | 3.02 65 | 3.13 60 | 4.65 105 | 3.05 83 | 7.41 91 | 2.75 82 | 0.63 16 | 2.56 21 | 0.54 61 | 1.39 10 | 3.49 11 | 0.22 40 |
OFLAF [77] | 44.3 | 0.49 6 | 2.06 7 | 0.32 7 | 1.11 5 | 2.84 5 | 0.40 5 | 1.69 14 | 3.49 19 | 0.48 3 | 3.50 2 | 3.84 2 | 7.28 1 | 3.00 53 | 3.12 52 | 4.49 81 | 3.14 91 | 8.10 114 | 2.67 70 | 1.02 110 | 4.49 111 | 0.64 93 | 1.71 85 | 4.23 87 | 0.22 40 |
Classic+NL [31] | 46.2 | 0.70 70 | 2.67 49 | 0.48 72 | 1.54 35 | 3.84 32 | 0.52 36 | 1.71 17 | 3.71 41 | 0.57 20 | 3.74 26 | 4.29 31 | 7.72 55 | 2.94 35 | 3.03 36 | 4.38 49 | 3.03 82 | 7.04 80 | 2.71 75 | 0.73 59 | 2.99 60 | 0.50 21 | 1.62 58 | 4.04 62 | 0.19 7 |
SVFilterOh [111] | 46.5 | 0.59 24 | 2.19 12 | 0.43 58 | 1.29 15 | 3.30 12 | 0.47 26 | 1.81 42 | 3.40 15 | 0.60 29 | 3.76 32 | 4.12 21 | 8.42 105 | 3.20 102 | 3.30 94 | 4.93 116 | 2.86 53 | 6.39 53 | 2.63 59 | 0.62 10 | 2.34 7 | 0.56 70 | 1.56 40 | 3.87 39 | 0.26 83 |
DF-Auto [115] | 46.8 | 0.67 60 | 2.47 33 | 0.50 76 | 2.20 75 | 5.04 67 | 1.05 95 | 1.71 17 | 3.43 16 | 0.54 9 | 3.93 59 | 4.55 56 | 7.67 50 | 2.88 18 | 2.88 17 | 4.46 73 | 2.47 10 | 5.15 12 | 2.52 42 | 0.71 51 | 2.95 57 | 0.54 61 | 1.57 43 | 3.91 44 | 0.26 83 |
AggregFlow [97] | 47.0 | 0.79 89 | 3.07 91 | 0.50 76 | 1.73 46 | 4.44 44 | 0.65 60 | 1.60 6 | 3.19 9 | 0.49 5 | 3.82 46 | 4.53 54 | 7.65 43 | 2.90 24 | 2.88 17 | 4.51 85 | 2.74 46 | 5.17 13 | 2.85 96 | 0.73 59 | 3.01 63 | 0.55 67 | 1.49 29 | 3.73 29 | 0.21 30 |
ALD-Flow [66] | 47.0 | 0.71 72 | 2.79 59 | 0.48 72 | 1.77 48 | 4.67 54 | 0.62 55 | 1.83 49 | 4.35 73 | 0.60 29 | 3.76 32 | 4.38 37 | 7.90 78 | 2.95 39 | 3.02 34 | 4.55 93 | 2.55 20 | 4.75 9 | 2.68 72 | 0.62 10 | 2.42 11 | 0.49 13 | 1.66 75 | 4.14 78 | 0.20 16 |
SuperFlow [81] | 47.2 | 0.68 63 | 2.69 51 | 0.55 86 | 2.21 76 | 5.06 68 | 1.12 100 | 1.95 67 | 4.06 58 | 0.75 67 | 4.01 65 | 4.39 39 | 7.85 73 | 2.89 22 | 2.90 20 | 4.51 85 | 2.22 2 | 4.39 5 | 2.27 4 | 0.69 43 | 2.88 49 | 0.53 44 | 1.37 7 | 3.44 9 | 0.21 30 |
PGM-C [120] | 47.3 | 0.62 39 | 2.83 68 | 0.37 26 | 1.37 21 | 3.62 24 | 0.49 31 | 1.92 62 | 4.18 63 | 0.56 14 | 3.80 40 | 4.71 66 | 7.58 25 | 2.98 46 | 3.07 42 | 4.48 80 | 2.68 41 | 5.97 42 | 2.56 50 | 0.71 51 | 2.99 60 | 0.51 28 | 1.66 75 | 4.18 84 | 0.23 57 |
EPPM w/o HM [88] | 47.4 | 0.50 9 | 2.26 17 | 0.29 3 | 1.55 37 | 4.22 41 | 0.44 18 | 2.24 95 | 5.31 106 | 0.90 84 | 3.77 35 | 4.54 55 | 7.51 17 | 2.85 15 | 2.89 19 | 4.32 34 | 2.79 48 | 6.25 49 | 2.57 52 | 0.77 75 | 3.34 85 | 0.63 88 | 1.61 55 | 4.03 60 | 0.22 40 |
RNLOD-Flow [121] | 47.5 | 0.56 17 | 2.33 22 | 0.38 34 | 1.89 57 | 4.91 63 | 0.52 36 | 1.85 53 | 4.05 57 | 0.78 70 | 3.61 13 | 3.95 8 | 7.54 22 | 3.02 65 | 3.21 78 | 4.32 34 | 2.95 71 | 6.99 78 | 2.63 59 | 0.67 33 | 2.56 21 | 0.54 61 | 1.67 80 | 4.15 79 | 0.21 30 |
Brox et al. [5] | 47.9 | 0.69 65 | 2.80 62 | 0.40 48 | 1.91 61 | 4.80 57 | 0.65 60 | 2.01 76 | 4.92 97 | 0.76 69 | 3.86 53 | 4.19 24 | 7.60 33 | 2.94 35 | 2.99 30 | 4.44 65 | 2.72 45 | 6.17 47 | 2.48 34 | 0.72 57 | 3.02 64 | 0.51 28 | 1.40 12 | 3.49 11 | 0.20 16 |
Efficient-NL [60] | 49.0 | 0.57 20 | 2.25 16 | 0.35 20 | 1.78 50 | 4.51 46 | 0.53 40 | 2.29 96 | 3.78 45 | 1.08 98 | 3.75 27 | 4.25 29 | 7.53 21 | 2.89 22 | 2.98 28 | 4.30 26 | 2.92 67 | 7.43 92 | 2.46 27 | 0.77 75 | 3.18 72 | 0.53 44 | 1.81 102 | 4.38 97 | 0.20 16 |
FESL [72] | 49.6 | 0.64 47 | 2.38 28 | 0.39 39 | 1.48 30 | 3.86 33 | 0.48 28 | 1.81 42 | 3.73 42 | 0.72 58 | 3.76 32 | 4.25 29 | 7.61 34 | 3.03 72 | 3.15 64 | 4.46 73 | 3.10 87 | 7.65 103 | 2.72 78 | 0.71 51 | 2.96 58 | 0.47 5 | 1.62 58 | 4.03 60 | 0.22 40 |
Second-order prior [8] | 49.7 | 0.73 77 | 2.73 55 | 0.55 86 | 2.51 85 | 5.74 86 | 0.66 63 | 2.31 97 | 5.29 104 | 0.94 91 | 3.80 40 | 4.39 39 | 7.41 9 | 2.90 24 | 3.00 32 | 4.31 28 | 2.49 14 | 5.42 21 | 2.45 25 | 0.66 28 | 2.68 31 | 0.48 7 | 1.57 43 | 3.90 43 | 0.24 65 |
CPM-Flow [116] | 50.1 | 0.63 43 | 2.83 68 | 0.39 39 | 1.40 25 | 3.71 29 | 0.51 34 | 1.84 51 | 3.99 56 | 0.56 14 | 4.01 65 | 5.11 92 | 7.73 57 | 2.97 44 | 3.05 38 | 4.52 87 | 2.61 30 | 5.41 20 | 2.62 57 | 0.72 57 | 3.02 64 | 0.54 61 | 1.57 43 | 3.93 46 | 0.26 83 |
Kuang [131] | 50.5 | 0.65 51 | 3.03 86 | 0.39 39 | 1.54 35 | 4.01 39 | 0.47 26 | 1.95 67 | 4.29 69 | 0.62 37 | 3.90 55 | 4.77 69 | 7.82 68 | 2.98 46 | 3.09 46 | 4.43 61 | 2.60 27 | 5.91 41 | 2.39 17 | 0.79 80 | 3.41 88 | 0.63 88 | 1.53 35 | 3.84 36 | 0.19 7 |
S2D-Matching [84] | 51.3 | 0.67 60 | 2.70 52 | 0.42 54 | 2.02 65 | 5.10 71 | 0.59 50 | 1.73 22 | 3.59 26 | 0.64 47 | 3.75 27 | 4.10 19 | 7.93 81 | 3.02 65 | 3.18 69 | 4.31 28 | 3.19 94 | 7.90 110 | 2.76 83 | 0.68 38 | 2.60 26 | 0.48 7 | 1.60 48 | 3.98 50 | 0.22 40 |
IROF-TV [53] | 52.0 | 0.69 65 | 2.91 77 | 0.46 67 | 1.58 39 | 3.95 37 | 0.49 31 | 1.90 58 | 4.86 94 | 0.69 53 | 3.71 24 | 4.34 34 | 7.82 68 | 3.07 80 | 3.19 73 | 4.66 107 | 2.63 35 | 6.35 52 | 2.31 8 | 0.68 38 | 2.81 43 | 0.52 39 | 1.46 24 | 3.64 25 | 0.25 76 |
SIOF [67] | 52.1 | 0.82 93 | 2.94 80 | 0.57 91 | 2.85 96 | 6.23 106 | 1.15 102 | 1.83 49 | 4.11 61 | 0.72 58 | 3.84 50 | 4.51 50 | 7.59 29 | 2.74 8 | 2.74 8 | 4.22 16 | 2.52 16 | 5.33 18 | 2.49 37 | 0.66 28 | 2.68 31 | 0.53 44 | 1.62 58 | 4.02 56 | 0.24 65 |
TC/T-Flow [76] | 52.9 | 0.71 72 | 2.62 43 | 0.40 48 | 1.77 48 | 4.61 52 | 0.54 44 | 1.74 25 | 3.65 36 | 0.57 20 | 3.78 37 | 4.37 36 | 7.73 57 | 3.04 76 | 3.18 69 | 4.54 91 | 2.61 30 | 5.54 26 | 2.51 40 | 0.97 107 | 3.95 103 | 0.63 88 | 1.61 55 | 4.04 62 | 0.18 5 |
CLG-TV [48] | 53.1 | 0.70 70 | 2.82 64 | 0.52 81 | 2.57 87 | 5.88 90 | 0.76 76 | 2.02 78 | 4.55 84 | 0.92 88 | 3.94 60 | 4.49 47 | 7.82 68 | 2.94 35 | 3.01 33 | 4.35 42 | 2.43 9 | 4.87 10 | 2.46 27 | 0.64 19 | 2.62 29 | 0.50 21 | 1.55 37 | 3.84 36 | 0.26 83 |
Classic+CPF [83] | 54.5 | 0.61 34 | 2.36 26 | 0.35 20 | 1.58 39 | 4.01 39 | 0.51 34 | 1.79 37 | 3.80 47 | 0.66 49 | 3.56 6 | 3.84 2 | 7.34 2 | 3.25 106 | 3.54 113 | 4.43 61 | 3.20 97 | 8.18 115 | 2.64 61 | 0.83 89 | 3.31 83 | 0.49 13 | 1.77 96 | 4.41 99 | 0.22 40 |
p-harmonic [29] | 56.0 | 0.60 31 | 2.63 44 | 0.39 39 | 2.63 89 | 5.89 91 | 0.82 79 | 1.98 73 | 4.94 98 | 0.89 81 | 4.13 81 | 4.86 74 | 7.68 51 | 2.95 39 | 3.06 39 | 4.38 49 | 2.58 24 | 5.68 33 | 2.52 42 | 0.70 47 | 2.92 54 | 0.54 61 | 1.49 29 | 3.74 30 | 0.24 65 |
CostFilter [40] | 57.1 | 0.51 10 | 2.34 23 | 0.29 3 | 1.37 21 | 3.61 23 | 0.43 14 | 2.06 83 | 5.38 108 | 0.74 64 | 3.79 38 | 4.49 47 | 7.55 23 | 3.12 92 | 3.29 93 | 4.43 61 | 3.28 102 | 5.79 38 | 3.61 123 | 0.73 59 | 3.10 70 | 0.53 44 | 1.78 98 | 4.48 104 | 0.21 30 |
CBF [12] | 57.2 | 0.60 31 | 2.64 47 | 0.45 63 | 2.17 71 | 5.10 71 | 0.75 74 | 1.91 61 | 3.73 42 | 0.72 58 | 4.20 87 | 4.50 49 | 9.33 113 | 2.92 29 | 2.84 12 | 5.05 119 | 2.57 23 | 5.58 27 | 2.53 45 | 0.71 51 | 2.87 47 | 0.60 80 | 1.48 28 | 3.66 27 | 0.32 118 |
TCOF [69] | 57.9 | 0.64 47 | 2.61 42 | 0.39 39 | 2.92 97 | 6.38 108 | 0.86 84 | 1.69 14 | 3.65 36 | 0.50 6 | 3.82 46 | 4.51 50 | 7.88 77 | 2.97 44 | 3.09 46 | 4.26 21 | 2.91 65 | 7.16 85 | 2.48 34 | 0.79 80 | 3.39 87 | 0.49 13 | 1.76 94 | 4.40 98 | 0.25 76 |
ComplOF-FED-GPU [35] | 59.1 | 0.65 51 | 2.82 64 | 0.39 39 | 1.69 43 | 4.52 47 | 0.57 47 | 2.33 98 | 4.35 73 | 0.98 94 | 3.80 40 | 4.69 65 | 7.75 62 | 2.95 39 | 3.07 42 | 4.37 46 | 2.70 44 | 6.17 47 | 2.54 48 | 0.76 72 | 3.06 67 | 0.53 44 | 1.74 90 | 4.31 92 | 0.24 65 |
OAR-Flow [125] | 59.5 | 0.71 72 | 2.83 68 | 0.44 62 | 1.87 55 | 4.84 59 | 0.67 65 | 1.81 42 | 4.28 68 | 0.59 27 | 3.69 20 | 4.46 46 | 7.63 39 | 3.03 72 | 3.14 62 | 4.56 94 | 2.90 62 | 6.65 59 | 2.73 79 | 0.90 98 | 3.63 95 | 0.57 74 | 1.56 40 | 3.88 41 | 0.21 30 |
LDOF [28] | 60.8 | 0.90 101 | 2.91 77 | 0.69 109 | 2.17 71 | 4.58 51 | 1.40 108 | 2.14 88 | 5.04 100 | 0.91 86 | 4.10 76 | 5.00 82 | 7.97 82 | 2.92 29 | 2.95 26 | 4.44 65 | 2.48 13 | 5.10 11 | 2.42 20 | 0.70 47 | 2.93 56 | 0.53 44 | 1.55 37 | 3.88 41 | 0.22 40 |
Sparse Occlusion [54] | 62.1 | 0.66 56 | 2.83 68 | 0.45 63 | 2.18 73 | 5.57 81 | 0.59 50 | 1.78 35 | 3.53 22 | 0.73 61 | 3.84 50 | 4.52 52 | 7.65 43 | 3.04 76 | 3.18 69 | 4.42 58 | 3.06 84 | 7.39 90 | 2.73 79 | 0.76 72 | 3.21 75 | 0.45 3 | 1.66 75 | 4.15 79 | 0.25 76 |
TC-Flow [46] | 63.0 | 0.59 24 | 2.60 41 | 0.39 39 | 1.83 53 | 4.90 62 | 0.61 52 | 1.95 67 | 4.37 77 | 0.62 37 | 4.11 78 | 5.04 86 | 8.05 87 | 3.07 80 | 3.22 79 | 4.50 84 | 2.96 73 | 6.72 63 | 2.82 92 | 0.67 33 | 2.70 35 | 0.51 28 | 1.71 85 | 4.30 91 | 0.24 65 |
HBM-GC [105] | 63.5 | 0.69 65 | 2.73 55 | 0.53 82 | 1.86 54 | 4.86 60 | 0.61 52 | 1.58 5 | 2.85 1 | 0.56 14 | 4.09 74 | 4.59 60 | 8.15 93 | 3.32 111 | 3.43 109 | 5.14 122 | 3.62 116 | 9.31 124 | 3.00 106 | 0.65 23 | 2.50 16 | 0.53 44 | 1.51 31 | 3.74 30 | 0.25 76 |
MLDP_OF [89] | 63.5 | 0.56 17 | 2.27 18 | 0.35 20 | 1.90 60 | 4.87 61 | 0.53 40 | 1.71 17 | 3.62 33 | 0.55 12 | 4.07 70 | 4.34 34 | 8.25 96 | 3.07 80 | 3.20 77 | 4.68 108 | 3.80 122 | 7.63 101 | 3.75 124 | 0.73 59 | 2.88 49 | 0.61 82 | 1.63 64 | 4.05 67 | 0.30 113 |
TF+OM [100] | 64.0 | 0.66 56 | 2.76 57 | 0.45 63 | 1.55 37 | 3.89 35 | 0.82 79 | 1.95 67 | 4.46 79 | 0.67 50 | 4.07 70 | 4.87 75 | 7.86 75 | 3.05 79 | 3.18 69 | 4.54 91 | 2.74 46 | 5.80 39 | 2.69 74 | 0.77 75 | 3.30 80 | 0.58 78 | 1.56 40 | 3.87 39 | 0.26 83 |
EpicFlow [102] | 64.6 | 0.62 39 | 2.83 68 | 0.38 34 | 1.80 52 | 4.77 56 | 0.57 47 | 1.89 57 | 4.29 69 | 0.59 27 | 3.95 61 | 5.00 82 | 7.84 72 | 3.01 60 | 3.12 52 | 4.45 71 | 2.80 49 | 6.73 64 | 2.58 54 | 0.86 93 | 3.57 92 | 0.62 86 | 1.82 103 | 4.58 106 | 0.23 57 |
RFlow [90] | 67.8 | 0.65 51 | 2.85 73 | 0.46 67 | 2.58 88 | 6.04 99 | 0.69 66 | 1.92 62 | 4.55 84 | 0.73 61 | 4.08 73 | 5.11 92 | 7.77 63 | 3.02 65 | 3.23 80 | 4.32 34 | 2.61 30 | 6.28 50 | 2.35 12 | 0.75 68 | 3.26 79 | 0.54 61 | 1.75 92 | 4.34 94 | 0.26 83 |
FlowNet2 [122] | 67.8 | 1.16 115 | 3.96 119 | 0.64 102 | 1.88 56 | 4.53 49 | 0.84 83 | 2.04 80 | 4.35 73 | 0.80 74 | 3.92 57 | 5.12 94 | 7.70 54 | 3.08 85 | 3.25 86 | 4.44 65 | 2.61 30 | 5.71 34 | 2.44 23 | 0.76 72 | 3.22 76 | 0.53 44 | 1.54 36 | 3.85 38 | 0.26 83 |
ROF-ND [107] | 69.0 | 0.66 56 | 2.35 24 | 0.37 26 | 2.10 66 | 5.37 76 | 0.61 52 | 1.74 25 | 3.66 39 | 0.63 41 | 4.70 105 | 6.43 118 | 7.86 75 | 2.93 32 | 2.98 28 | 4.46 73 | 3.06 84 | 7.88 109 | 2.58 54 | 0.88 95 | 3.58 93 | 0.64 93 | 2.06 118 | 5.02 117 | 0.23 57 |
Fusion [6] | 70.2 | 0.68 63 | 3.27 100 | 0.39 39 | 1.67 42 | 4.22 41 | 0.54 44 | 1.82 46 | 3.68 40 | 0.78 70 | 4.24 89 | 5.23 99 | 7.66 45 | 3.07 80 | 3.44 111 | 4.14 7 | 2.99 77 | 8.18 115 | 2.44 23 | 0.84 91 | 3.75 98 | 0.56 70 | 1.78 98 | 4.46 102 | 0.27 96 |
IAOF [50] | 70.8 | 1.07 112 | 3.34 106 | 0.69 109 | 4.58 128 | 7.97 130 | 1.63 114 | 2.16 90 | 4.59 88 | 0.87 79 | 4.35 90 | 4.52 52 | 7.69 52 | 2.98 46 | 3.12 52 | 4.38 49 | 2.68 41 | 6.31 51 | 2.43 21 | 0.69 43 | 2.92 54 | 0.51 28 | 1.60 48 | 3.99 52 | 0.24 65 |
Modified CLG [34] | 71.6 | 0.63 43 | 2.53 35 | 0.49 74 | 3.29 114 | 6.18 103 | 1.69 115 | 2.21 92 | 6.06 113 | 0.96 93 | 4.13 81 | 5.03 85 | 7.73 57 | 3.02 65 | 3.13 60 | 4.44 65 | 2.86 53 | 6.65 59 | 2.66 67 | 0.69 43 | 2.88 49 | 0.53 44 | 1.61 55 | 3.99 52 | 0.28 102 |
FlowNetS+ft+v [112] | 71.6 | 0.83 94 | 2.85 73 | 0.66 106 | 2.95 100 | 6.18 103 | 1.43 109 | 1.98 73 | 4.82 92 | 0.74 64 | 3.96 63 | 4.66 63 | 7.91 79 | 3.09 86 | 3.25 86 | 4.59 100 | 2.58 24 | 5.61 29 | 2.53 45 | 0.81 86 | 3.50 90 | 0.53 44 | 1.58 46 | 3.94 48 | 0.20 16 |
TriFlow [95] | 71.9 | 0.78 87 | 3.43 107 | 0.49 74 | 2.42 80 | 5.49 79 | 1.16 103 | 1.88 55 | 4.51 81 | 0.67 50 | 3.95 61 | 4.82 73 | 7.58 25 | 3.19 101 | 3.43 109 | 4.58 98 | 2.89 61 | 6.47 54 | 2.52 42 | 0.74 65 | 3.08 68 | 0.55 67 | 1.64 67 | 4.01 54 | 0.24 65 |
Local-TV-L1 [65] | 72.2 | 0.98 107 | 3.10 93 | 0.80 116 | 3.02 110 | 5.97 95 | 1.46 110 | 1.82 46 | 3.79 46 | 0.63 41 | 4.45 96 | 4.74 68 | 9.64 117 | 3.00 53 | 3.10 48 | 4.60 101 | 3.42 110 | 5.46 25 | 3.82 125 | 0.67 33 | 2.78 40 | 0.51 28 | 1.41 15 | 3.51 13 | 0.27 96 |
F-TV-L1 [15] | 73.1 | 0.94 104 | 3.18 96 | 0.74 112 | 2.81 95 | 6.05 100 | 0.96 90 | 2.11 87 | 4.90 96 | 1.02 96 | 4.12 80 | 4.96 80 | 8.14 92 | 3.01 60 | 3.26 88 | 4.09 4 | 2.63 35 | 5.42 21 | 2.65 63 | 0.79 80 | 3.30 80 | 0.61 82 | 1.44 19 | 3.59 18 | 0.25 76 |
OFH [38] | 74.7 | 0.72 76 | 2.82 64 | 0.47 69 | 2.15 69 | 5.09 70 | 0.65 60 | 2.09 85 | 5.22 103 | 0.70 55 | 3.81 44 | 4.73 67 | 7.66 45 | 3.02 65 | 3.19 73 | 4.31 28 | 2.92 67 | 6.97 77 | 2.74 81 | 1.05 112 | 4.43 110 | 0.64 93 | 1.88 111 | 4.71 112 | 0.23 57 |
Occlusion-TV-L1 [63] | 75.6 | 0.69 65 | 2.82 64 | 0.57 91 | 2.80 94 | 6.49 111 | 0.82 79 | 1.94 65 | 4.87 95 | 0.85 77 | 4.36 91 | 5.61 102 | 8.07 91 | 2.93 32 | 3.03 36 | 4.36 43 | 3.02 81 | 6.74 66 | 2.91 100 | 0.90 98 | 2.91 52 | 0.79 114 | 1.66 75 | 4.12 76 | 0.20 16 |
Complementary OF [21] | 75.7 | 0.63 43 | 3.05 88 | 0.33 14 | 1.69 43 | 4.57 50 | 0.55 46 | 2.83 115 | 4.32 72 | 1.35 111 | 3.96 63 | 4.94 79 | 7.85 73 | 3.09 86 | 3.32 96 | 4.31 28 | 2.88 59 | 6.87 71 | 2.66 67 | 1.04 111 | 4.35 108 | 0.61 82 | 2.26 122 | 5.68 124 | 0.24 65 |
Classic++ [32] | 75.9 | 0.73 77 | 2.95 81 | 0.54 84 | 2.23 77 | 5.44 77 | 0.69 66 | 2.00 75 | 4.52 83 | 0.78 70 | 4.21 88 | 5.06 88 | 7.97 82 | 3.02 65 | 3.14 62 | 4.39 53 | 3.20 97 | 6.95 75 | 3.15 114 | 0.75 68 | 3.14 71 | 0.53 44 | 1.65 71 | 4.09 70 | 0.26 83 |
AdaConv-v1 [126] | 76.9 | 1.36 122 | 3.88 117 | 1.03 121 | 2.98 104 | 5.28 74 | 2.45 127 | 3.19 120 | 6.26 115 | 2.19 127 | 6.11 124 | 7.09 124 | 9.70 119 | 2.66 1 | 2.68 7 | 4.08 3 | 2.31 5 | 4.65 8 | 2.29 6 | 0.91 100 | 3.87 100 | 0.90 121 | 1.27 3 | 3.17 2 | 0.27 96 |
CRTflow [80] | 77.3 | 0.85 96 | 3.17 95 | 0.63 101 | 2.56 86 | 5.92 92 | 0.80 78 | 2.15 89 | 5.29 104 | 1.01 95 | 4.07 70 | 4.64 62 | 8.61 107 | 3.09 86 | 3.24 82 | 4.53 88 | 2.60 27 | 5.29 15 | 2.65 63 | 0.75 68 | 3.20 74 | 0.57 74 | 1.62 58 | 4.04 62 | 0.26 83 |
Steered-L1 [118] | 78.7 | 0.61 34 | 2.96 82 | 0.37 26 | 1.78 50 | 4.71 55 | 0.66 63 | 2.46 102 | 4.22 64 | 1.21 105 | 4.53 100 | 5.21 98 | 8.54 106 | 3.15 97 | 3.35 104 | 4.42 58 | 2.95 71 | 6.77 67 | 2.85 96 | 0.83 89 | 3.63 95 | 0.66 98 | 1.65 71 | 4.11 75 | 0.26 83 |
GraphCuts [14] | 79.0 | 1.07 112 | 3.93 118 | 0.60 96 | 1.97 63 | 4.46 45 | 1.07 97 | 3.51 123 | 3.64 35 | 1.45 117 | 4.47 97 | 5.20 97 | 8.18 94 | 2.99 50 | 3.12 52 | 4.19 12 | 2.59 26 | 6.58 58 | 2.16 2 | 0.89 97 | 3.92 102 | 0.70 106 | 1.77 96 | 4.41 99 | 0.28 102 |
SimpleFlow [49] | 79.6 | 0.67 60 | 2.79 59 | 0.42 54 | 2.16 70 | 5.07 69 | 0.63 56 | 2.79 112 | 4.51 81 | 1.32 109 | 3.75 27 | 4.29 31 | 7.74 60 | 3.01 60 | 3.15 64 | 4.39 53 | 3.36 109 | 8.79 121 | 2.77 87 | 1.46 126 | 7.29 128 | 1.11 128 | 2.11 120 | 5.29 120 | 0.19 7 |
BlockOverlap [61] | 79.9 | 0.96 105 | 3.02 85 | 0.85 117 | 2.94 99 | 5.79 87 | 1.60 113 | 1.90 58 | 3.50 21 | 0.90 84 | 4.65 103 | 4.80 71 | 10.2 122 | 3.22 104 | 3.16 67 | 5.42 126 | 3.34 105 | 6.09 45 | 3.55 122 | 0.70 47 | 2.73 38 | 0.63 88 | 1.34 4 | 3.33 4 | 0.28 102 |
Black & Anandan [4] | 81.5 | 1.02 110 | 3.18 96 | 0.73 111 | 3.65 119 | 6.67 119 | 1.36 106 | 2.81 113 | 5.56 110 | 1.38 115 | 4.43 94 | 5.07 89 | 7.61 34 | 3.13 93 | 3.30 94 | 4.57 95 | 2.54 18 | 5.63 30 | 2.38 16 | 0.78 78 | 3.24 78 | 0.53 44 | 1.60 48 | 3.92 45 | 0.28 102 |
Adaptive [20] | 81.8 | 0.80 90 | 3.21 99 | 0.60 96 | 2.98 104 | 6.54 116 | 0.92 87 | 2.03 79 | 4.56 86 | 0.89 81 | 4.03 69 | 4.79 70 | 7.91 79 | 3.09 86 | 3.26 88 | 4.37 46 | 2.99 77 | 6.78 68 | 2.80 91 | 0.79 80 | 3.38 86 | 0.49 13 | 1.72 87 | 4.27 89 | 0.27 96 |
Aniso-Texture [82] | 81.8 | 0.59 24 | 2.66 48 | 0.37 26 | 2.78 91 | 6.44 110 | 0.89 85 | 2.45 100 | 4.08 60 | 0.84 76 | 4.43 94 | 5.74 108 | 8.41 103 | 3.20 102 | 3.41 107 | 4.64 103 | 3.73 118 | 10.1 128 | 3.01 107 | 0.67 33 | 2.73 38 | 0.50 21 | 1.82 103 | 4.46 102 | 0.25 76 |
Nguyen [33] | 82.0 | 1.00 109 | 3.06 90 | 0.79 114 | 4.07 124 | 6.92 122 | 1.78 118 | 2.22 93 | 6.36 116 | 0.95 92 | 4.71 106 | 5.33 101 | 7.72 55 | 3.01 60 | 3.19 73 | 4.30 26 | 2.68 41 | 6.52 56 | 2.35 12 | 0.99 108 | 4.61 112 | 0.68 103 | 1.62 58 | 4.04 62 | 0.20 16 |
2D-CLG [1] | 82.0 | 0.81 91 | 2.89 76 | 0.59 94 | 3.59 117 | 6.36 107 | 1.88 121 | 2.65 108 | 5.32 107 | 1.30 106 | 4.54 101 | 5.09 90 | 7.59 29 | 3.00 53 | 3.10 48 | 4.44 65 | 2.84 51 | 6.94 73 | 2.64 61 | 0.94 103 | 4.26 107 | 0.62 86 | 1.65 71 | 3.93 46 | 0.23 57 |
HBpMotionGpu [43] | 82.3 | 1.20 116 | 3.99 120 | 0.94 118 | 3.63 118 | 7.12 124 | 1.71 117 | 1.79 37 | 3.83 48 | 0.63 41 | 4.63 102 | 5.92 112 | 8.41 103 | 3.03 72 | 3.24 82 | 4.46 73 | 3.10 87 | 7.03 79 | 2.87 98 | 0.61 7 | 2.39 10 | 0.48 7 | 1.79 100 | 4.37 95 | 0.29 109 |
Correlation Flow [75] | 82.4 | 0.61 34 | 2.63 44 | 0.34 19 | 2.48 83 | 6.02 97 | 0.64 57 | 1.79 37 | 3.61 30 | 0.64 47 | 4.01 65 | 4.43 43 | 8.36 101 | 3.33 112 | 3.34 99 | 5.77 127 | 3.76 119 | 9.14 122 | 2.99 105 | 1.07 113 | 4.88 113 | 0.77 111 | 1.87 109 | 4.62 108 | 0.26 83 |
ACK-Prior [27] | 82.8 | 0.59 24 | 2.67 49 | 0.32 7 | 1.61 41 | 4.38 43 | 0.50 33 | 2.71 110 | 4.06 58 | 1.37 112 | 4.11 78 | 4.89 76 | 8.03 86 | 3.30 109 | 3.36 105 | 5.18 123 | 3.48 112 | 7.75 107 | 3.17 117 | 0.82 88 | 3.30 80 | 0.67 101 | 1.86 108 | 4.62 108 | 0.30 113 |
IAOF2 [51] | 83.4 | 0.99 108 | 3.48 109 | 0.65 105 | 3.05 112 | 6.85 120 | 1.13 101 | 1.88 55 | 4.31 71 | 0.70 55 | 4.39 92 | 5.13 95 | 8.06 88 | 3.44 116 | 3.87 119 | 4.49 81 | 3.10 87 | 7.70 104 | 2.57 52 | 0.70 47 | 2.87 47 | 0.50 21 | 1.72 87 | 4.27 89 | 0.22 40 |
Ad-TV-NDC [36] | 84.7 | 1.54 124 | 3.33 105 | 1.43 125 | 3.73 120 | 6.52 113 | 1.78 118 | 1.96 72 | 4.57 87 | 0.85 77 | 4.78 109 | 4.93 78 | 8.94 110 | 3.25 106 | 3.34 99 | 4.64 103 | 2.91 65 | 5.29 15 | 3.06 109 | 0.71 51 | 2.91 52 | 0.52 39 | 1.46 24 | 3.62 22 | 0.29 109 |
TriangleFlow [30] | 84.9 | 0.84 95 | 3.27 100 | 0.56 90 | 2.34 79 | 5.47 78 | 0.69 66 | 2.05 81 | 4.27 67 | 0.88 80 | 4.19 86 | 5.13 95 | 8.28 99 | 3.00 53 | 3.17 68 | 4.23 18 | 3.06 84 | 7.45 93 | 2.61 56 | 1.09 115 | 5.06 115 | 0.84 118 | 2.27 123 | 5.56 122 | 0.23 57 |
CNN-flow-warp+ref [117] | 85.7 | 0.63 43 | 2.43 30 | 0.50 76 | 2.42 80 | 5.68 83 | 0.97 91 | 2.47 104 | 5.75 111 | 1.08 98 | 5.30 116 | 6.00 115 | 9.01 112 | 3.10 90 | 3.23 80 | 4.73 110 | 2.86 53 | 6.69 62 | 2.71 75 | 1.08 114 | 4.97 114 | 0.70 106 | 1.64 67 | 4.09 70 | 0.23 57 |
BriefMatch [124] | 87.9 | 0.77 85 | 2.78 58 | 0.55 86 | 1.89 57 | 4.66 53 | 1.10 99 | 2.46 102 | 3.90 54 | 1.34 110 | 5.30 116 | 5.70 106 | 10.5 124 | 3.04 76 | 3.12 52 | 4.79 111 | 4.37 128 | 7.13 83 | 4.74 128 | 0.74 65 | 3.00 62 | 0.64 93 | 1.69 83 | 4.16 83 | 0.27 96 |
Shiralkar [42] | 90.7 | 0.81 91 | 3.32 104 | 0.47 69 | 2.78 91 | 5.92 92 | 0.74 73 | 2.56 105 | 6.86 119 | 1.11 100 | 4.89 111 | 6.32 117 | 7.64 40 | 3.03 72 | 3.34 99 | 4.00 1 | 3.22 99 | 7.57 99 | 2.91 100 | 1.21 119 | 5.44 120 | 0.70 106 | 2.03 115 | 5.06 118 | 0.20 16 |
Filter Flow [19] | 91.0 | 0.88 99 | 2.99 84 | 0.67 107 | 3.21 113 | 6.22 105 | 1.70 116 | 2.06 83 | 4.41 78 | 0.89 81 | 4.69 104 | 4.81 72 | 8.97 111 | 3.15 97 | 3.24 82 | 4.94 117 | 2.88 59 | 6.13 46 | 2.77 87 | 0.79 80 | 3.33 84 | 0.61 82 | 1.72 87 | 4.23 87 | 0.35 122 |
TV-L1-improved [17] | 92.1 | 0.74 81 | 3.05 88 | 0.55 86 | 2.97 103 | 6.52 113 | 0.95 89 | 2.45 100 | 4.22 64 | 1.18 104 | 4.09 74 | 4.99 81 | 8.00 85 | 3.16 99 | 3.34 99 | 4.38 49 | 3.12 90 | 7.30 88 | 2.76 83 | 1.12 117 | 5.32 119 | 0.78 113 | 1.74 90 | 4.32 93 | 0.28 102 |
LocallyOriented [52] | 92.6 | 0.85 96 | 3.04 87 | 0.64 102 | 3.00 108 | 6.41 109 | 1.01 93 | 2.23 94 | 4.84 93 | 0.80 74 | 4.47 97 | 5.61 102 | 8.25 96 | 3.07 80 | 3.26 88 | 4.32 34 | 3.55 114 | 7.16 85 | 3.51 121 | 0.88 95 | 3.63 95 | 0.57 74 | 1.76 94 | 4.37 95 | 0.27 96 |
Bartels [41] | 95.5 | 0.90 101 | 3.30 103 | 0.74 112 | 2.13 67 | 5.53 80 | 1.01 93 | 1.95 67 | 4.26 66 | 0.91 86 | 4.98 112 | 5.87 110 | 10.9 127 | 3.59 119 | 3.28 92 | 6.74 131 | 5.60 130 | 7.64 102 | 6.55 131 | 0.73 59 | 2.71 36 | 0.79 114 | 1.64 67 | 4.04 62 | 0.39 125 |
Horn & Schunck [3] | 96.0 | 0.92 103 | 3.16 94 | 0.61 98 | 3.79 121 | 6.91 121 | 1.51 111 | 2.98 116 | 6.57 118 | 1.59 119 | 5.15 114 | 5.92 112 | 7.97 82 | 3.25 106 | 3.50 112 | 4.58 98 | 2.64 38 | 6.07 44 | 2.45 25 | 0.91 100 | 3.87 100 | 0.64 93 | 1.75 92 | 4.22 85 | 0.28 102 |
TI-DOFE [24] | 96.3 | 1.46 123 | 3.65 114 | 1.28 124 | 4.57 126 | 7.55 127 | 2.30 126 | 2.65 108 | 6.94 120 | 1.30 106 | 5.46 119 | 5.88 111 | 8.37 102 | 3.13 93 | 3.42 108 | 4.47 78 | 2.52 16 | 5.64 31 | 2.30 7 | 0.84 91 | 3.58 93 | 0.66 98 | 1.79 100 | 4.10 73 | 0.31 117 |
NL-TV-NCC [25] | 96.8 | 0.77 85 | 2.93 79 | 0.43 58 | 2.18 73 | 5.65 82 | 0.64 57 | 2.09 85 | 4.68 90 | 0.93 90 | 4.72 107 | 5.82 109 | 9.36 114 | 3.56 117 | 3.34 99 | 6.58 130 | 3.26 100 | 8.02 113 | 2.83 93 | 0.96 105 | 3.86 99 | 0.72 109 | 1.83 105 | 4.53 105 | 0.33 120 |
SegOF [10] | 99.0 | 0.76 83 | 3.08 92 | 0.51 80 | 2.42 80 | 5.35 75 | 0.97 91 | 3.17 119 | 5.83 112 | 1.57 118 | 4.52 99 | 6.66 121 | 7.66 45 | 3.13 93 | 3.33 98 | 4.57 95 | 3.31 103 | 8.43 118 | 2.87 98 | 1.38 123 | 6.82 127 | 1.03 124 | 1.90 112 | 4.75 113 | 0.23 57 |
StereoOF-V1MT [119] | 99.1 | 0.86 98 | 3.62 113 | 0.43 58 | 2.50 84 | 5.73 84 | 0.75 74 | 2.99 117 | 6.55 117 | 1.41 116 | 5.71 122 | 6.70 122 | 8.82 108 | 3.37 115 | 3.73 117 | 4.37 46 | 3.68 117 | 7.84 108 | 3.42 119 | 1.18 118 | 5.29 117 | 0.96 122 | 1.69 83 | 4.10 73 | 0.21 30 |
StereoFlow [44] | 99.8 | 2.07 126 | 5.46 130 | 1.08 122 | 3.85 123 | 7.22 125 | 1.51 111 | 2.01 76 | 5.12 102 | 0.79 73 | 4.13 81 | 5.04 86 | 7.77 63 | 4.97 129 | 6.41 129 | 4.80 113 | 3.89 126 | 11.3 131 | 2.76 83 | 0.68 38 | 2.85 45 | 0.53 44 | 2.09 119 | 5.25 119 | 0.28 102 |
Rannacher [23] | 100.4 | 0.78 87 | 3.18 96 | 0.59 94 | 3.00 108 | 6.62 117 | 0.90 86 | 2.56 105 | 4.96 99 | 1.37 112 | 4.13 81 | 5.27 100 | 8.18 94 | 3.18 100 | 3.37 106 | 4.49 81 | 3.18 92 | 7.49 94 | 2.83 93 | 1.10 116 | 5.20 116 | 0.77 111 | 1.84 106 | 4.59 107 | 0.29 109 |
SPSA-learn [13] | 102.4 | 0.97 106 | 3.54 110 | 0.64 102 | 3.02 110 | 6.01 96 | 1.36 106 | 3.01 118 | 5.38 108 | 1.62 120 | 4.74 108 | 5.09 90 | 7.59 29 | 3.34 113 | 3.71 116 | 4.53 88 | 2.98 76 | 7.52 96 | 2.55 49 | 2.11 131 | 11.0 131 | 1.87 131 | 3.18 129 | 7.90 130 | 0.24 65 |
UnFlow [129] | 103.5 | 1.04 111 | 4.08 121 | 0.62 99 | 2.95 100 | 6.08 101 | 1.05 95 | 2.57 107 | 7.05 121 | 1.16 103 | 4.15 85 | 5.62 104 | 7.58 25 | 3.57 118 | 4.08 122 | 4.57 95 | 3.55 114 | 9.38 125 | 2.71 75 | 0.95 104 | 4.07 105 | 0.63 88 | 2.64 127 | 6.00 125 | 0.30 113 |
Dynamic MRF [7] | 104.8 | 0.73 77 | 3.46 108 | 0.42 54 | 2.27 78 | 5.93 94 | 0.69 66 | 2.82 114 | 7.28 124 | 1.37 112 | 5.71 122 | 7.07 123 | 9.50 115 | 3.24 105 | 3.59 114 | 4.36 43 | 3.79 121 | 9.86 127 | 3.13 112 | 1.29 122 | 5.96 122 | 0.89 120 | 2.04 116 | 4.77 114 | 0.30 113 |
2bit-BM-tele [98] | 107.9 | 1.12 114 | 3.60 112 | 0.94 118 | 2.95 100 | 6.65 118 | 1.22 104 | 2.05 81 | 3.90 54 | 1.13 101 | 5.13 113 | 5.96 114 | 10.8 126 | 3.74 124 | 3.76 118 | 6.21 128 | 4.96 129 | 9.26 123 | 5.02 129 | 1.93 130 | 9.67 130 | 1.56 130 | 1.55 37 | 3.82 35 | 0.35 122 |
HCIC-L [99] | 109.5 | 3.01 130 | 5.27 129 | 3.31 131 | 2.98 104 | 5.18 73 | 2.23 125 | 2.76 111 | 5.07 101 | 1.13 101 | 5.22 115 | 5.73 107 | 8.87 109 | 3.30 109 | 3.24 82 | 5.33 125 | 3.45 111 | 7.55 97 | 3.09 110 | 0.86 93 | 3.48 89 | 0.68 103 | 2.61 126 | 6.30 127 | 0.34 121 |
Learning Flow [11] | 112.4 | 0.89 100 | 3.59 111 | 0.62 99 | 2.93 98 | 6.52 113 | 0.92 87 | 3.33 122 | 7.23 122 | 1.64 121 | 5.36 118 | 6.54 120 | 9.51 116 | 3.75 126 | 4.10 123 | 5.32 124 | 3.34 105 | 7.55 97 | 3.12 111 | 1.01 109 | 4.41 109 | 0.74 110 | 2.04 116 | 4.85 116 | 0.37 124 |
Adaptive flow [45] | 112.9 | 1.98 125 | 4.23 122 | 1.80 126 | 4.57 126 | 7.46 126 | 3.10 129 | 2.35 99 | 4.62 89 | 1.30 106 | 5.62 121 | 5.69 105 | 10.7 125 | 3.74 124 | 4.07 121 | 5.06 120 | 3.81 123 | 9.47 126 | 3.13 112 | 0.78 78 | 3.18 72 | 0.69 105 | 1.87 109 | 4.67 111 | 0.29 109 |
GroupFlow [9] | 113.9 | 1.35 121 | 4.96 127 | 0.79 114 | 2.79 93 | 5.84 88 | 1.31 105 | 3.68 126 | 7.72 125 | 2.05 124 | 4.78 109 | 6.45 119 | 8.26 98 | 3.88 128 | 4.56 128 | 4.65 105 | 3.77 120 | 10.3 129 | 2.91 100 | 1.24 121 | 5.48 121 | 0.66 98 | 2.47 125 | 6.17 126 | 0.26 83 |
SILK [79] | 114.8 | 1.24 117 | 3.81 115 | 1.00 120 | 4.12 125 | 6.98 123 | 1.91 122 | 3.56 124 | 7.27 123 | 1.82 122 | 5.59 120 | 6.31 116 | 9.67 118 | 3.35 114 | 3.67 115 | 4.72 109 | 3.82 124 | 6.94 73 | 4.15 127 | 0.93 102 | 3.99 104 | 0.82 116 | 1.85 107 | 4.43 101 | 0.32 118 |
Heeger++ [104] | 116.4 | 1.32 119 | 4.80 126 | 0.58 93 | 2.99 107 | 5.73 84 | 1.09 98 | 4.83 128 | 10.4 129 | 2.28 128 | 6.73 125 | 7.16 125 | 9.92 120 | 3.72 122 | 4.28 126 | 4.79 111 | 3.83 125 | 8.26 117 | 3.31 118 | 1.52 127 | 6.36 123 | 0.85 119 | 2.32 124 | 5.62 123 | 0.25 76 |
SLK [47] | 118.0 | 1.33 120 | 3.87 116 | 1.14 123 | 3.84 122 | 6.17 102 | 2.07 123 | 3.83 127 | 7.77 126 | 2.10 126 | 7.08 127 | 7.94 127 | 10.2 122 | 3.75 126 | 4.39 127 | 4.40 55 | 3.34 105 | 7.98 112 | 2.96 104 | 1.39 124 | 6.69 125 | 1.03 124 | 2.23 121 | 5.47 121 | 0.43 126 |
FFV1MT [106] | 120.0 | 1.24 117 | 4.63 124 | 0.68 108 | 3.40 115 | 6.02 97 | 1.79 120 | 4.86 129 | 11.9 130 | 2.45 129 | 6.73 125 | 7.16 125 | 9.92 120 | 3.73 123 | 4.12 124 | 4.97 118 | 3.54 113 | 7.22 87 | 3.16 116 | 1.59 128 | 6.61 124 | 1.01 123 | 2.70 128 | 6.37 128 | 0.51 130 |
FOLKI [16] | 120.2 | 2.60 128 | 4.64 125 | 2.99 128 | 4.63 129 | 7.59 128 | 2.71 128 | 3.30 121 | 8.65 128 | 2.07 125 | 7.97 129 | 7.95 128 | 13.7 130 | 3.69 121 | 4.22 125 | 4.86 114 | 3.34 105 | 6.66 61 | 3.47 120 | 1.22 120 | 5.31 118 | 1.07 127 | 1.96 113 | 4.62 108 | 0.46 127 |
PGAM+LK [55] | 121.4 | 2.08 127 | 5.24 128 | 1.91 127 | 3.50 116 | 6.51 112 | 2.09 124 | 3.61 125 | 7.85 127 | 2.01 123 | 7.70 128 | 8.44 129 | 13.1 129 | 3.61 120 | 3.95 120 | 5.07 121 | 4.07 127 | 8.65 120 | 3.90 126 | 0.96 105 | 4.24 106 | 0.83 117 | 1.99 114 | 4.84 115 | 0.46 127 |
Pyramid LK [2] | 123.5 | 2.73 129 | 4.32 123 | 3.06 129 | 5.47 130 | 7.60 129 | 3.83 130 | 6.84 130 | 6.09 114 | 3.76 130 | 12.2 131 | 16.6 131 | 16.4 131 | 5.05 130 | 6.65 130 | 4.62 102 | 3.26 100 | 7.14 84 | 3.15 114 | 1.40 125 | 6.78 126 | 1.04 126 | 3.71 131 | 9.32 131 | 0.46 127 |
Periodicity [78] | 130.2 | 3.07 131 | 6.73 131 | 3.09 130 | 7.28 131 | 8.40 131 | 5.04 131 | 7.68 131 | 13.2 131 | 5.81 131 | 9.03 130 | 16.0 130 | 12.5 128 | 6.13 131 | 7.92 131 | 6.46 129 | 6.09 131 | 10.5 130 | 6.23 130 | 1.72 129 | 8.02 129 | 1.33 129 | 3.56 130 | 7.62 129 | 1.32 131 |
Method | time* | frames | color | Reference and notes | |
[1] 2D-CLG | 844 | 2 | gray | The 2D-CLG method by Bruhn et al. as implemented by Stefan Roth. [A. Bruhn, J. Weickert, and C. Schnörr. Lucas/Kanade meets Horn/Schunck: combining local and global optic flow methods. IJCV 63(3), 2005.] Parameters were set to match the published performance on Yosemite sequence, which may not be optimal for other sequences. | |
[2] Pyramid LK | 12 | 2 | color | A modification of Bouguet's pyramidal implementation of Lucas-Kanade. | |
[3] Horn & Schunck | 49 | 2 | gray | A modern Matlab implementation of the Horn & Schunck method by Deqing Sun. Parameters set to optimize AAE on all training data. | |
[4] Black & Anandan | 328 | 2 | gray | A modern Matlab implementation of the Black & Anandan method by Deqing Sun. | |
[5] Brox et al. | 18 | 2 | color | T. Brox, A. Bruhn, N. Papenberg, and J. Weickert. High accuracy optical flow estimation based on a theory for warping. ECCV 2004. (Improved using separate robust functions as proposed in A. Bruhn and J. Weickert, Towards ultimate motion estimation, ICCV 2005; improved by training on the training set.) | |
[6] Fusion | 2,666 | 2 | color | V. Lempitsky, S. Roth, and C. Rother. Discrete-continuous optimization for optical flow estimation. CVPR 2008. | |
[7] Dynamic MRF | 366 | 2 | gray | B. Glocker, N. Paragios, N. Komodakis, G. Tziritas, and N. Navab. Optical flow estimation with uncertainties through dynamic MRFs. CVPR 2008. (Method improved since publication.) | |
[8] Second-order prior | 14 | 2 | gray | W. Trobin, T. Pock, D. Cremers, and H. Bischof. An unbiased second-order prior for high-accuracy motion estimation. DAGM 2008. (Method improved since publication; for details see W. Trobin, Ph.D. thesis, 2009.) | |
[9] GroupFlow | 600 | 2 | gray | X. Ren. Local Grouping for Optical Flow. CVPR 2008. | |
[10] SegOF | 60 | 2 | color | L. Xu, J. Chen, and J. Jia. Segmentation based variational model for accurate optical flow estimation. ECCV 2008. Code available. | |
[11] Learning Flow | 825 | 2 | gray | D. Sun, S. Roth, J.P. Lewis, and M. Black. Learning optical flow (SRF-LFC). ECCV 2008. | |
[12] CBF | 69 | 2 | color | W. Trobin, T. Pock, D. Cremers, and H. Bischof. Continuous energy minimization via repeated binary fusion. ECCV 2008. (Method improved since publication; for details see W. Trobin, Ph.D. thesis, 2009.) | |
[13] SPSA-learn | 200 | 2 | color | Y. Li and D. Huttenlocher. Learning for optical flow using stochastic optimization. ECCV 2008. | |
[14] GraphCuts | 1,200 | 2 | color | T. Cooke. Two applications of graph-cuts to image processing. DICTA 2008. | |
[15] F-TV-L1 | 8 | 2 | gray | A. Wedel, T. Pock, J. Braun, U. Franke, and D. Cremers. Duality TV-L1 flow with fundamental matrix prior. IVCNZ 2008. | |
[16] FOLKI | 1.4 | 2 | gray | G. Le Besnerais and F. Champagnat. Dense optical flow by iterative local window registration. ICIP 2005. | |
[17] TV-L1-improved | 2.9 | 2 | gray | A. Wedel, T. Pock, C. Zach, H. Bischof, and D. Cremers. An improved algorithm for TV-L1 optical flow computation. Proceedings of the Dagstuhl Visual Motion Analysis Workshop 2008. Code at GPU4Vision. | |
[18] DPOF | 287 | 2 | color | C. Lei and Y.-H. Yang. Optical flow estimation on coarse-to-fine region-trees using discrete optimization. ICCV 2009. (Method improved since publication.) | |
[19] Filter Flow | 34,000 | 2 | color | S. Seitz and S. Baker. Filter flow. ICCV 2009. | |
[20] Adaptive | 9.2 | 2 | gray | A. Wedel, D. Cremers, T. Pock, and H. Bischof. Structure- and motion-adaptive regularization for high accuracy optic flow. ICCV 2009. | |
[21] Complementary OF | 44 | 2 | color | H. Zimmer, A. Bruhn, J. Weickert, L. Valgaerts, A. Salgado, B. Rosenhahn, and H.-P. Seidel. Complementary optic flow. EMMCVPR 2009. | |
[22] Aniso. Huber-L1 | 2 | 2 | gray | M. Werlberger, W. Trobin, T. Pock, A. Wedel, D. Cremers, and H. Bischof. Anisotropic Huber-L1 optical flow. BMVC 2009. Code at GPU4Vision. | |
[23] Rannacher | 0.12 | 2 | gray | J. Rannacher. Realtime 3D motion estimation on graphics hardware. Bachelor thesis, Heidelberg University, 2009. | |
[24] TI-DOFE | 260 | 2 | gray | C. Cassisa, S. Simoens, and V. Prinet. Two-frame optical flow formulation in an unwarped multiresolution scheme. CIARP 2009. | |
[25] NL-TV-NCC | 20 | 2 | color | M. Werlberger, T. Pock, and H. Bischof. Motion estimation with non-local total variation regularization. CVPR 2010. | |
[26] MDP-Flow | 188 | 2 | color | L. Xu, J. Jia, and Y. Matsushita. Motion detail preserving optical flow estimation. CVPR 2010. | |
[27] ACK-Prior | 5872 | 2 | color | K. Lee, D. Kwon, I. Yun, and S. Lee. Optical flow estimation with adaptive convolution kernel prior on discrete framework. CVPR 2010. | |
[28] LDOF | 122 | 2 | color | T. Brox and J. Malik. Large displacement optical flow: descriptor matching in variational motion estimation. PAMI 33(3):500-513, 2011. | |
[29] p-harmonic | 565 | 2 | gray | J. Gai and R. Stevenson. Optical flow estimation with p-harmonic regularization. ICIP 2010. | |
[30] TriangleFlow | 4200 | 2 | gray | B. Glocker, H. Heibel, N. Navab, P. Kohli, and C. Rother. TriangleFlow: Optical flow with triangulation-based higher-order likelihoods. ECCV 2010. | |
[31] Classic+NL | 972 | 2 | color | D. Sun, S. Roth, and M. Black. Secrets of optical flow estimation and their principles. CVPR 2010. Matlab code. | |
[32] Classic++ | 486 | 2 | gray | A modern implementation of the classical formulation descended from Horn & Schunck and Black & Anandan; see D. Sun, S. Roth, and M. Black, Secrets of optical flow estimation and their principles, CVPR 2010. | |
[33] Nguyen | 33 | 2 | gray | D. Nguyen. Tuning optical flow estimation with image-driven functions. ICRA 2011. | |
[34] Modified CLG | 133 | 2 | gray | R. Fezzani, F. Champagnat, and G. Le Besnerais. Combined local global method for optic flow computation. EUSIPCO 2010. | |
[35] ComplOF-FED-GPU | 0.97 | 2 | color | P. Gwosdek, H. Zimmer, S. Grewenig, A. Bruhn, and J. Weickert. A highly efficient GPU implementation for variational optic flow based on the Euler-Lagrange framework. CVGPU Workshop 2010. | |
[36] Ad-TV-NDC | 35 | 2 | gray | M. Nawaz. Motion estimation with adaptive regularization and neighborhood dependent constraint. DICTA 2010. | |
[37] Layers++ | 18206 | 2 | color | D. Sun, E. Sudderth, and M. Black. Layered image motion with explicit occlusions, temporal consistency, and depth ordering. NIPS 2010. | |
[38] OFH | 620 | 3 | color | H. Zimmer, A. Bruhn, J. Weickert. Optic flow in harmony. IJCV 93(3) 2011. | |
[39] LSM | 1615 | 2 | color | K. Jia, X. Wang, and X. Tang. Optical flow estimation using learned sparse model. ICCV 2011. | |
[40] CostFilter | 55 | 2 | color | C. Rhemann, A. Hosni, M. Bleyer, C. Rother, and M. Gelautz. Fast cost-volume filtering for visual correspondence and beyond. CVPR 2011. | |
[41] Bartels | 0.15 | 2 | gray | C. Bartels and G. de Haan. Smoothness constraints in recursive search motion estimation for picture rate conversion. IEEE TCSVT 2010. Version improved since publication: mapped on GPU. | |
[42] Shiralkar | 600 | 2 | gray | M. Shiralkar and R. Schalkoff. A self organization-based optical flow estimator with GPU implementation. MVA 23(6):1229-1242. | |
[43] HBpMotionGpu | 1000 | 5 | gray | S. Grauer-Gray and C. Kambhamettu. Hierarchical belief propagation to reduce search space using CUDA for stereo and motion estimation. WACV 2009. (Method improved since publication.) | |
[44] StereoFlow | 7200 | 2 | color | G. Rosman, S. Shem-Tov, D. Bitton, T. Nir, G. Adiv, R. Kimmel, A. Feuer, and A. Bruckstein. Over-parameterized optical flow using a stereoscopic constraint. SSVM 2011:761-772. | |
[45] Adaptive flow | 121 | 2 | gray | T. Arici. Energy minimization based motion estimation using adaptive smoothness priors. Submitted to IEEE TIP 2011. | |
[46] TC-Flow | 2500 | 5 | color | S. Volz, A. Bruhn, L. Valgaerts, and H. Zimmer. Modeling temporal coherence for optical flow. ICCV 2011. | |
[47] SLK | 300 | 2 | gray | T. Corpetti and E. Mémin. Stochastic uncertainty models for the luminance consistency assumption. IEEE TIP 2011. | |
[48] CLG-TV | 29 | 2 | gray | M. Drulea. Total variation regularization of local-global optical flow. ITSC 2011. Matlab code. | |
[49] SimpleFlow | 1.7 | 2 | color | M. Tao, J. Bai, P. Kohli, S. Paris. SimpleFlow: a non-iterative, sublinear optical flow algorithm. EUROGRAPHICS 2012. | |
[50] IAOF | 57 | 2 | gray | D. Nguyen. Improving motion estimation using image-driven functions and hybrid scheme. PSIVT 2011. | |
[51] IAOF2 | 56 | 2 | gray | D. Nguyen. Enhancing the sharpness of flow field using image-driven functions with occlusion-aware filter. Submitted to TIP 2011. | |
[52] LocallyOriented | 9541 | 2 | gray | Y.Niu, A. Dick, and M. Brooks. Locally oriented optical flow computation. To appear in TIP 2012. | |
[53] IROF-TV | 261 | 2 | color | H. Rashwan, D. Puig, and M. Garcia. On improving the robustness of differential optical flow. ICCV 2011 Artemis workshop. | |
[54] Sparse Occlusion | 2312 | 2 | color | A. Ayvaci, M. Raptis, and S. Soatto. Sparse occlusion detection with optical flow. Submitted to IJCV 2011. | |
[55] PGAM+LK | 0.37 | 2 | gray | A. Alba, E. Arce-Santana, and M. Rivera. Optical flow estimation with prior models obtained from phase correlation. ISVC 2010. | |
[56] Sparse-NonSparse | 713 | 2 | color | L. Chen, J. Wang, and Y. Wu. Decomposing and regularizing sparse/non-sparse components for motion field estimation. Submitted to PAMI 2013. | |
[57] nLayers | 36150 | 4 | color | D. Sun, E. Sudderth, and M. Black. Layered segmentation and optical flow estimation over time. CVPR 2012. | |
[58] IROF++ | 187 | 2 | color | H. Rashwan, D. Puig, and M. Garcia. Variational optical flow estimation based on stick tensor voting. IEEE TIP 2013. | |
[59] COFM | 600 | 3 | color | M. Mozerov. Constrained optical flow estimation as a matching problem. IEEE TIP 2013. | |
[60] Efficient-NL | 400 | 2 | color | P. Krähenbühl and V. Koltun. Efficient nonlocal regularization for optical flow. ECCV 2012. | |
[61] BlockOverlap | 2 | 2 | gray | M. Santoro, G. AlRegib, and Y. Altunbasak. Motion estimation using block overlap minimization. Submitted to MMSP 2012. | |
[62] Ramp | 1200 | 2 | color | A. Singh and N. Ahuja. Exploiting ramp structures for improving optical flow estimation. ICPR 2012. | |
[63] Occlusion-TV-L1 | 538 | 3 | gray | C. Ballester, L. Garrido, V. Lazcano, and V. Caselles. A TV-L1 optical flow method with occlusion detection. DAGM-OAGM 2012. | |
[64] TV-L1-MCT | 90 | 2 | color | M. Mohamed and B. Mertsching. TV-L1 optical flow estimation with image details recovering based on modified census transform. ISVC 2012. | |
[65] Local-TV-L1 | 500 | 2 | gray | L. Raket. Local smoothness for global optical flow. ICIP 2012. | |
[66] ALD-Flow | 61 | 2 | color | M. Stoll, A. Bruhn, and S. Volz. Adaptive integration of feature matches into variational optic flow methods. ACCV 2012. | |
[67] SIOF | 234 | 2 | color | L. Xu, Z. Dai, and J. Jia. Scale invariant optical flow. ECCV 2012. | |
[68] MDP-Flow2 | 342 | 2 | color | L. Xu, J. Jia, and Y. Matsushita. Motion detail preserving optical flow estimation. PAMI 34(9):1744-1757, 2012. Code available. | |
[69] TCOF | 1421 | all | gray | J. Sanchez, A. Salgado, and N. Monzon. Optical flow estimation with consistent spatio-temporal coherence models. VISAPP 2013. | |
[70] LME | 476 | 2 | color | W. Li, D. Cosker, M. Brown, and R. Tang. Optical flow estimation using Laplacian mesh energy. CVPR 2013. | |
[71] NN-field | 362 | 2 | color | L. Chen, H. Jin, Z. Lin, S. Cohen, and Y. Wu. Large displacement optical flow from nearest neighbor fields. CVPR 2013. | |
[72] FESL | 3310 | 2 | color | W. Dong, G. Shi, X. Hu, and Y. Ma. Nonlocal sparse and low-rank regularization for optical flow estimation. Submitted to IEEE TIP 2013. | |
[73] PMF | 35 | 2 | color | J. Lu, H. Yang, D. Min, and M. Do. PatchMatch filter: efficient edge-aware filtering meets randomized search for fast correspondence field estimation. CVPR 2013. | |
[74] FC-2Layers-FF | 2662 | 4 | color | D. Sun, J. Wulff, E. Sudderth, H. Pfister, and M. Black. A fully-connected layered model of foreground and background flow. CVPR 2013. | |
[75] Correlation Flow | 290 | 2 | color | M. Drulea and S. Nedevschi. Motion estimation using the correlation transform. TIP 2013. Matlab code. | |
[76] TC/T-Flow | 341 | 5 | color | M. Stoll, S. Volz, and A. Bruhn. Joint trilateral filtering for multiframe optical flow. ICIP 2013. | |
[77] OFLAF | 1530 | 2 | color | T. Kim, H. Lee, and K. Lee. Optical flow via locally adaptive fusion of complementary data costs. ICCV 2013. | |
[78] Periodicity | 8000 | 4 | color | G. Khachaturov, S. Gonzalez-Brambila, and J. Gonzalez-Trejo. Periodicity-based computation of optical flow. Submitted to Computacion y Sistemas (CyS) 2013. | |
[79] SILK | 572 | 2 | gray | P. Zille, C. Xu, T. Corpetti, L. Shao. Observation models based on scale interactions for optical flow estimation. Submitted to IEEE TIP. | |
[80] CRTflow | 13 | 3 | color | O. Demetz, D. Hafner, and J. Weickert. The complete rank transform: a tool for accurate and morphologically invariant matching of structures. BMVC 2013. | |
[81] SuperFlow | 178 | 2 | color | Anonymous. Superpixel based optical flow estimation. ICCV 2013 submission 507. | |
[82] Aniso-Texture | 300 | 2 | color | Anonymous. Texture information-based optical flow estimation using an incremental multi-resolution approach. ITC-CSCC 2013 submission 267. | |
[83] Classic+CPF | 640 | 2 | gray | Z. Tu, R. Veltkamp, and N. van der Aa. A combined post-filtering method to improve accuracy of variational optical flow estimation. Submitted to Pattern Recognition 2013. | |
[84] S2D-Matching | 1200 | 2 | color | Anonymous. Locally affine sparse-to-dense matching for motion and occlusion estimation. ICCV 2013 submission 1479. | |
[85] AGIF+OF | 438 | 2 | gray | Z. Tu, R. Poppe, and R. Veltkamp. Adaptive guided image filter to warped interpolation image for variational optical flow computation. Submitted to Signal Processing 2015. | |
[86] DeepFlow | 13 | 2 | color | P. Weinzaepfel, J. Revaud, Z. Harchaoui, and C. Schmid. DeepFlow: large displacement optical flow with deep matching. ICCV 2013. | |
[87] NNF-Local | 673 | 2 | color | Z. Chen, H. Jin, Z. Lin, S. Cohen, and Y. Wu. Large displacement optical flow with nearest neighbor field. Submitted to PAMI 2014. | |
[88] EPPM w/o HM | 2.5 | 2 | color | L. Bao, Q. Yang, and H. Jin. Fast edge-preserving PatchMatch for large displacement optical flow. CVPR 2014. | |
[89] MLDP_OF | 165 | 2 | gray | M. Mohamed, H. Rashwan, B. Mertsching, M. Garcia, and D. Puig. Illumination-robust optical flow approach using local directional pattern. IEEE TCSVT 24(9):1499-1508, 2014. | |
[90] RFlow | 20 | 2 | gray | S. Ali, C. Daul, and W. Blondel. Robust and accurate optical flow estimation for weak texture and varying illumination condition: Application to cystoscopy. IPTA 2014. | |
[91] SRR-TVOF-NL | 32 | all | color | P. Pohl, M. Sirotenko, E. Tolstaya, and V. Bucha. Edge preserving motion estimation with occlusions correction for assisted 2D to 3D conversion. IS&T/SPIE Electronic Imaging 2014. | |
[92] 2DHMM-SAS | 157 | 2 | color | M.-C. Shih, R. Shenoy, and K. Rose. A two-dimensional hidden Markov model with spatially-adaptive states with application of optical flow. ICIP 2014 submission. | |
[93] WLIF-Flow | 700 | 2 | color | Z. Tu, R. Veltkamp, N. van der Aa, and C. Van Gemeren. Weighted local intensity fusion method for variational optical flow estimation. Submitted to TIP 2014. | |
[94] FMOF | 215 | 2 | color | N. Jith, A. Ramakanth, and V. Babu. Optical flow estimation using approximate nearest neighbor field fusion. ICASSP 2014. | |
[95] TriFlow | 150 | 2 | color | TriFlow. Optical flow with geometric occlusion estimation and fusion of multiple frames. ECCV 2014 submission 914. | |
[96] ComponentFusion | 6.5 | 2 | color | Anonymous. Fast optical flow by component fusion. ECCV 2014 submission 941. | |
[97] AggregFlow | 1642 | 2 | color | D. Fortun, P. Bouthemy, and C. Kervrann. Aggregation of local parametric candidates and exemplar-based occlusion handling for optical flow. Preprint arXiv:1407.5759. | |
[98] 2bit-BM-tele | 124 | 2 | gray | R. Xu and D. Taubman. Robust dense block-based motion estimation using a two-bit transform on a Laplacian pyramid. ICIP 2013. | |
[99] HCIC-L | 330 | 2 | color | Anonymous. Globally-optimal image correspondence using a hierarchical graphical model. NIPS 2014 submission 114. | |
[100] TF+OM | 600 | 2 | color | R. Kennedy and C. Taylor. Optical flow with geometric occlusion estimation and fusion of multiple frames. EMMCVPR 2015. | |
[101] PH-Flow | 800 | 2 | color | J. Yang and H. Li. Dense, accurate optical flow estimation with piecewise parametric model. CVPR 2015. | |
[102] EpicFlow | 16 | 2 | color | J. Revaud, P. Weinzaepfel, Z. Harchaoui, and C. Schmid. EpicFlow: edge-preserving interpolation of correspondences for optical flow. CVPR 2015. | |
[103] NNF-EAC | 380 | 2 | color | Anonymous. Variational method for joint optical flow estimation and edge-aware image restoration. CVPR 2015 submission 2336. | |
[104] Heeger++ | 6600 | 5 | gray | Anonymous. A context aware biologically inspired algorithm for optical flow (updated results). CVPR 2015 submission 2238. | |
[105] HBM-GC | 330 | 2 | color | A. Zheng and Y. Yuan. Motion estimation via hierarchical block matching and graph cut. Submitted to ICIP 2015. | |
[106] FFV1MT | 358 | 5 | gray | F. Solari, M. Chessa, N. Medathati, and P. Kornprobst. What can we expect from a V1-MT feedforward architecture for optical flow estimation? Submitted to Signal Processing: Image Communication 2015. | |
[107] ROF-ND | 4 | 2 | color | S. Ali, C. Daul, E. Galbrun, and W. Blondel. Illumination invariant large displacement optical flow using robust neighbourhood descriptors. Submitted to CVIU 2015. | |
[108] DeepFlow2 | 16 | 2 | color | J. Revaud, P. Weinzaepfel, Z. Harchaoui, and C. Schmid. Deep convolutional matching. Submitted to IJCV, 2015. | |
[109] HAST | 2667 | 2 | color | Anonymous. Highly accurate optical flow estimation on superpixel tree. ICCV 2015 submission 2221. | |
[110] FlowFields | 15 | 2 | color | C. Bailer, B. Taetz, and D. Stricker. Flow Fields: Dense unregularized correspondence fields for highly accurate large displacement optical flow estimation. ICCV 2015. | |
[111] SVFilterOh | 1.56 | 2 | color | Anonymous. Fast estimation of large displacement optical flow using PatchMatch and dominant motion patterns. CVPR 2016 submission 1788. | |
[112] FlowNetS+ft+v | 0.5 | 2 | color | Anonymous. Learning optical flow with convolutional neural networks. ICCV 2015 submission 235. | |
[113] CombBMOF | 51 | 2 | color | M. Brüggemann, R. Kays, P. Springer, and O. Erdler. Combined block-matching and adaptive differential motion estimation in a hierarchical multi-scale framework. ICGIP 2014. (Method improved since publication.) | |
[114] PMMST | 182 | 2 | color | F. Zhang, S. Xu, and X. Zhang. High accuracy correspondence field estimation via MST based patch matching. Submitted to TIP 2015. | |
[115] DF-Auto | 70 | 2 | color | N. Monzon, A. Salgado, and J. Sanchez. Regularization strategies for discontinuity-preserving optical flow methods. Submitted to TIP 2015. | |
[116] CPM-Flow | 3 | 2 | color | Anonymous. Efficient coarse-to-fine PatchMatch for large displacement optical flow. CVPR 2016 submission 241. | |
[117] CNN-flow-warp+ref | 1.4 | 3 | color | D. Teney and M. Hebert. Learning to extract motion from videos in convolutional neural networks. ArXiv 1601.07532, 2016. | |
[118] Steered-L1 | 804 | 2 | color | Anonymous. Optical flow estimation via steered-L1 norm. Submitted to WSCG 2016. | |
[119] StereoOF-V1MT | 343 | 2 | gray | Anonymous. Visual features for action-oriented tasks: a cortical-like model for disparity and optic flow computation. BMVC 2016 submission 132. | |
[120] PGM-C | 5 | 2 | color | Y. Li. Pyramidal gradient matching for optical flow estimation. Submitted to PAMI 2016. | |
[121] RNLOD-Flow | 1040 | 2 | gray | C. Zhang, Z. Chen, M. Wang, M. Li, and S. Jiang. Robust non-local TV-L1 optical flow estimation with occlusion detection. Submitted to TIP 2016. | |
[122] FlowNet2 | 0.091 | 2 | color | Anonymous. FlowNet 2.0: Evolution of optical flow estimation with deep networks. CVPR 2017 submission 900. | |
[123] S2F-IF | 20 | 2 | color | Anonymous. S2F-IF: Slow-to-fast interpolator flow. CVPR 2017 submission 765. | |
[124] BriefMatch | 0.068 | 2 | gray | G. Eilertsen, P.-E. Forssen, and J. Unger. Dense binary feature matching for real-time optical flow estimation. SCIA 2017 submission 62. | |
[125] OAR-Flow | 60 | 2 | color | Anonymous. Order-adaptive regularisation for variational optical flow: global, local and in between. SSVM 2017 submission 20. | |
[126] AdaConv-v1 | 2.8 | 2 | color | S. Niklaus, L. Mai, and F. Liu. (Interpolation results only.) Video frame interpolation via adaptive convolution. CVPR 2017. | |
[127] SepConv-v1 | 0.2 | 2 | color | S. Niklaus, L. Mai, and F. Liu. (Interpolation results only.) Video frame interpolation via adaptive separable convolution. ICCV 2017. | |
[128] ProbFlowFields | 37 | 2 | color | A. Wannenwetsch, M. Keuper, and S. Roth. ProbFlow: joint optical flow and uncertainty estimation. ICCV 2017. | |
[129] UnFlow | 0.12 | 2 | color | Anonymous. UnFlow: Unsupervised learning of optical flow with a bidirectional census loss. Submitted to AAAI 2018. | |
[130] FlowFields+ | 10.5 | 2 | color | C. Bailer, B. Taetz, and D. Stricker. Flow fields: Dense correspondence fields for highly accurate large displacement optical flow estimation. Submitted to PAMI 2017. | |
[131] Kuang | 9.9 | 2 | gray | F. Kuang. PatchMatch algorithms for motion estimation and stereo reconstruction. Master thesis, University of Stuttgart, 2017. |