Optical flow evaluation results |
Statistics:
Average
SD
R2.5
R5.0
R10.0
A50
A75
A95
Error type: endpoint angle interpolation normalized interpolation |
A50 angle error |
avg. |
Army (Hidden texture) GT im0 im1 |
Mequon (Hidden texture) GT im0 im1 |
Schefflera (Hidden texture) GT im0 im1 |
Wooden (Hidden texture) GT im0 im1 |
Grove (Synthetic) GT im0 im1 |
Urban (Synthetic) GT im0 im1 |
Yosemite (Synthetic) GT im0 im1 |
Teddy (Stereo) GT im0 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 | |
TC/T-Flow [76] | 13.9 | 1.17 1 | 2.19 7 | 1.07 1 | 0.63 3 | 2.28 41 | 0.68 3 | 0.81 4 | 1.52 12 | 0.91 5 | 0.42 1 | 1.13 16 | 0.37 1 | 0.71 6 | 1.17 20 | 0.61 5 | 0.44 3 | 2.23 18 | 0.52 4 | 1.21 3 | 1.83 1 | 2.10 74 | 0.63 31 | 0.94 40 | 0.67 34 |
ALD-Flow [66] | 14.1 | 1.35 7 | 2.24 8 | 1.41 10 | 0.64 4 | 1.62 8 | 0.76 5 | 0.81 4 | 1.45 8 | 0.97 11 | 0.48 6 | 1.00 8 | 0.44 8 | 0.67 4 | 1.13 12 | 0.61 5 | 0.45 4 | 2.21 17 | 0.57 5 | 1.46 18 | 2.22 4 | 2.09 73 | 0.62 26 | 0.95 42 | 0.68 42 |
ComponentFusion [96] | 15.3 | 1.24 2 | 2.24 8 | 1.15 3 | 0.66 5 | 1.61 7 | 0.79 7 | 0.78 2 | 1.37 3 | 0.88 1 | 0.43 3 | 1.27 31 | 0.37 1 | 0.77 10 | 1.18 21 | 0.64 7 | 0.64 24 | 3.07 54 | 0.71 17 | 1.66 34 | 2.85 57 | 0.98 9 | 0.55 18 | 0.78 27 | 0.56 16 |
NN-field [71] | 16.6 | 1.36 9 | 2.10 4 | 1.32 5 | 0.74 11 | 2.17 35 | 0.86 15 | 0.85 9 | 1.48 10 | 0.93 7 | 0.50 9 | 1.10 12 | 0.43 6 | 0.76 9 | 1.03 5 | 0.71 9 | 0.71 39 | 1.57 3 | 0.79 29 | 1.93 52 | 2.58 34 | 1.79 53 | 0.49 13 | 0.74 15 | 0.37 6 |
NNF-Local [87] | 16.8 | 1.29 5 | 2.06 3 | 1.24 4 | 0.66 5 | 1.96 21 | 0.77 6 | 0.84 8 | 1.53 14 | 0.91 5 | 0.61 39 | 1.09 11 | 0.56 43 | 0.77 10 | 1.04 6 | 0.75 12 | 0.71 39 | 2.07 11 | 0.92 56 | 1.55 23 | 2.27 7 | 1.37 30 | 0.49 13 | 0.75 21 | 0.44 10 |
OAR-Flow [125] | 17.9 | 1.44 11 | 2.90 39 | 1.55 15 | 0.68 7 | 1.87 16 | 0.80 8 | 0.88 10 | 1.98 42 | 1.04 17 | 0.45 5 | 0.93 5 | 0.42 5 | 0.71 6 | 1.25 27 | 0.59 4 | 0.37 2 | 2.14 15 | 0.46 2 | 1.14 2 | 1.87 2 | 1.49 43 | 0.66 45 | 0.95 42 | 0.77 60 |
RNLOD-Flow [121] | 20.1 | 1.28 4 | 2.10 4 | 1.36 8 | 0.69 8 | 2.01 25 | 0.75 4 | 0.80 3 | 1.38 4 | 0.93 7 | 0.52 13 | 0.92 4 | 0.47 15 | 0.74 8 | 1.09 10 | 0.73 10 | 0.65 26 | 2.10 13 | 0.76 24 | 2.13 74 | 3.27 77 | 2.37 92 | 0.56 19 | 0.74 15 | 0.54 15 |
nLayers [57] | 21.4 | 1.35 7 | 1.80 1 | 1.44 11 | 1.07 81 | 2.03 26 | 1.33 87 | 0.88 10 | 1.47 9 | 1.14 45 | 0.42 1 | 0.70 1 | 0.38 3 | 0.62 2 | 0.81 2 | 0.55 1 | 0.57 14 | 1.73 4 | 0.68 8 | 1.73 42 | 2.67 43 | 1.53 45 | 0.62 26 | 0.73 11 | 0.67 34 |
TC-Flow [46] | 22.0 | 1.44 11 | 2.66 31 | 1.55 15 | 0.56 1 | 1.52 3 | 0.65 2 | 0.81 4 | 1.53 14 | 0.89 3 | 0.54 16 | 0.85 2 | 0.51 31 | 0.79 12 | 1.31 35 | 0.76 14 | 0.63 20 | 2.77 38 | 0.76 24 | 1.32 6 | 2.34 12 | 1.83 57 | 0.68 48 | 1.09 60 | 0.82 69 |
WLIF-Flow [93] | 23.0 | 1.54 25 | 2.30 11 | 1.58 19 | 0.85 31 | 2.13 34 | 0.99 33 | 0.90 15 | 1.57 20 | 1.04 17 | 0.55 21 | 1.18 22 | 0.49 22 | 0.86 26 | 1.18 21 | 0.83 19 | 0.69 35 | 2.26 22 | 0.78 27 | 1.58 25 | 2.28 8 | 1.67 49 | 0.60 22 | 0.71 9 | 0.60 19 |
HAST [109] | 24.4 | 1.24 2 | 1.88 2 | 1.14 2 | 0.60 2 | 1.59 6 | 0.64 1 | 0.74 1 | 0.94 1 | 0.88 1 | 0.48 6 | 0.94 7 | 0.43 6 | 0.59 1 | 0.77 1 | 0.57 3 | 0.78 55 | 2.30 25 | 0.94 58 | 2.06 68 | 3.22 74 | 3.57 116 | 0.65 40 | 0.80 28 | 0.92 80 |
Layers++ [37] | 24.7 | 1.55 27 | 2.42 15 | 1.77 43 | 0.94 55 | 2.05 30 | 1.14 66 | 0.90 15 | 1.42 6 | 1.10 31 | 0.49 8 | 0.89 3 | 0.44 8 | 0.64 3 | 0.82 3 | 0.56 2 | 0.64 24 | 2.04 9 | 0.72 18 | 1.98 59 | 2.87 59 | 1.80 54 | 0.61 23 | 0.70 8 | 0.63 23 |
AGIF+OF [85] | 26.8 | 1.46 13 | 2.42 15 | 1.56 18 | 0.95 60 | 2.75 65 | 1.10 59 | 0.89 13 | 1.73 31 | 1.13 34 | 0.52 13 | 1.43 38 | 0.45 11 | 0.79 12 | 1.12 11 | 0.77 15 | 0.63 20 | 2.40 28 | 0.73 20 | 1.61 31 | 2.47 21 | 1.73 52 | 0.61 23 | 0.73 11 | 0.64 28 |
Classic+CPF [83] | 27.6 | 1.47 14 | 2.60 26 | 1.54 14 | 0.88 38 | 2.71 64 | 1.03 43 | 0.89 13 | 1.72 30 | 1.13 34 | 0.51 10 | 1.24 29 | 0.45 11 | 0.79 12 | 1.14 14 | 0.81 17 | 0.63 20 | 2.32 26 | 0.75 23 | 1.82 47 | 2.50 23 | 2.28 86 | 0.62 26 | 0.71 9 | 0.66 33 |
OFLAF [77] | 28.2 | 1.67 34 | 2.34 12 | 1.68 32 | 0.76 13 | 1.65 9 | 0.85 14 | 0.90 15 | 1.31 2 | 0.98 12 | 0.63 44 | 0.93 5 | 0.55 40 | 0.67 4 | 0.89 4 | 0.65 8 | 0.92 74 | 1.83 6 | 0.99 65 | 1.63 32 | 2.53 26 | 1.70 50 | 0.76 66 | 0.90 38 | 0.83 71 |
PH-Flow [101] | 28.2 | 1.49 17 | 2.62 27 | 1.60 24 | 0.81 21 | 2.42 50 | 0.99 33 | 0.93 22 | 1.67 24 | 1.13 34 | 0.54 16 | 1.16 20 | 0.48 17 | 0.82 18 | 1.14 14 | 0.85 20 | 0.68 31 | 2.53 30 | 0.80 31 | 1.96 56 | 2.42 18 | 2.27 83 | 0.61 23 | 0.76 24 | 0.63 23 |
Sparse-NonSparse [56] | 28.2 | 1.48 15 | 2.64 29 | 1.62 26 | 0.84 29 | 2.37 46 | 1.03 43 | 0.91 19 | 1.79 34 | 1.11 33 | 0.51 10 | 1.20 25 | 0.45 11 | 0.84 22 | 1.29 31 | 0.85 20 | 0.63 20 | 2.55 31 | 0.74 21 | 1.81 46 | 2.29 9 | 2.12 77 | 0.63 31 | 0.74 15 | 0.67 34 |
COFM [59] | 28.2 | 1.29 5 | 2.43 17 | 1.33 6 | 0.70 9 | 1.82 14 | 0.83 10 | 0.81 4 | 1.56 18 | 1.08 27 | 0.43 3 | 1.01 9 | 0.39 4 | 0.82 18 | 1.20 23 | 1.16 65 | 0.67 30 | 1.98 7 | 0.79 29 | 1.57 24 | 2.33 11 | 2.20 79 | 0.97 98 | 1.06 55 | 1.44 112 |
IROF++ [58] | 29.4 | 1.54 25 | 2.55 21 | 1.63 27 | 0.82 23 | 2.51 58 | 1.01 37 | 0.94 26 | 1.80 35 | 1.14 45 | 0.56 25 | 1.17 21 | 0.50 27 | 0.86 26 | 1.24 25 | 0.89 30 | 0.66 28 | 3.01 52 | 0.78 27 | 1.47 19 | 2.47 21 | 0.83 3 | 0.65 40 | 0.84 31 | 0.67 34 |
LME [70] | 29.4 | 1.74 38 | 2.59 25 | 1.51 12 | 0.77 15 | 1.49 2 | 0.87 18 | 0.95 30 | 1.49 11 | 1.04 17 | 0.70 48 | 1.47 43 | 0.64 54 | 0.99 42 | 1.32 36 | 0.96 44 | 0.66 28 | 2.39 27 | 0.76 24 | 1.67 37 | 2.50 23 | 1.36 29 | 0.66 45 | 0.89 35 | 0.63 23 |
MDP-Flow2 [68] | 29.6 | 1.91 52 | 2.84 37 | 1.84 49 | 0.75 12 | 1.58 5 | 0.84 12 | 0.95 30 | 1.39 5 | 0.96 10 | 0.70 48 | 1.06 10 | 0.64 54 | 0.96 38 | 1.22 24 | 0.90 34 | 0.76 51 | 2.07 11 | 0.81 35 | 1.58 25 | 2.70 45 | 1.20 20 | 0.68 48 | 0.89 35 | 0.62 21 |
Efficient-NL [60] | 29.8 | 1.39 10 | 2.12 6 | 1.38 9 | 0.84 29 | 2.75 65 | 0.96 28 | 0.90 15 | 1.53 14 | 1.08 27 | 0.51 10 | 1.12 15 | 0.44 8 | 0.80 15 | 1.16 18 | 0.75 12 | 0.82 59 | 2.43 29 | 0.84 41 | 1.95 54 | 2.53 26 | 2.04 70 | 0.72 56 | 0.91 39 | 0.77 60 |
NNF-EAC [103] | 33.3 | 1.89 50 | 2.83 36 | 1.82 47 | 0.77 15 | 1.70 10 | 0.86 15 | 0.98 39 | 1.52 12 | 1.01 14 | 0.71 53 | 1.15 17 | 0.65 57 | 0.97 40 | 1.25 27 | 0.95 41 | 0.81 57 | 2.12 14 | 0.86 46 | 1.72 41 | 2.45 19 | 1.40 33 | 0.70 51 | 0.95 42 | 0.63 23 |
LSM [39] | 33.3 | 1.50 20 | 2.56 22 | 1.64 29 | 0.86 32 | 2.42 50 | 1.07 51 | 0.93 22 | 1.68 26 | 1.13 34 | 0.55 21 | 1.15 17 | 0.49 22 | 0.85 24 | 1.24 25 | 0.89 30 | 0.70 36 | 2.55 31 | 0.81 35 | 2.08 70 | 2.46 20 | 2.31 89 | 0.63 31 | 0.75 21 | 0.68 42 |
FC-2Layers-FF [74] | 33.5 | 1.53 24 | 2.41 14 | 1.66 31 | 0.87 33 | 2.35 45 | 1.06 49 | 0.92 20 | 1.44 7 | 1.13 34 | 0.57 28 | 1.11 13 | 0.51 31 | 0.80 15 | 1.05 8 | 0.85 20 | 0.74 45 | 2.26 22 | 0.86 46 | 2.24 78 | 2.87 59 | 2.31 89 | 0.63 31 | 0.75 21 | 0.68 42 |
2DHMM-SAS [92] | 34.0 | 1.49 17 | 2.65 30 | 1.59 22 | 0.82 23 | 2.69 61 | 0.99 33 | 0.96 33 | 1.92 39 | 1.14 45 | 0.54 16 | 1.22 27 | 0.48 17 | 0.87 28 | 1.28 30 | 0.89 30 | 0.70 36 | 2.96 50 | 0.81 35 | 1.98 59 | 2.40 16 | 2.22 81 | 0.62 26 | 0.84 31 | 0.65 31 |
Ramp [62] | 34.2 | 1.49 17 | 2.69 32 | 1.61 25 | 0.87 33 | 2.37 46 | 1.08 53 | 0.94 26 | 1.70 28 | 1.14 45 | 0.57 28 | 1.18 22 | 0.50 27 | 0.87 28 | 1.29 31 | 0.90 34 | 0.72 41 | 2.66 33 | 0.84 41 | 1.95 54 | 2.21 3 | 2.27 83 | 0.63 31 | 0.80 28 | 0.65 31 |
ProbFlowFields [128] | 36.1 | 1.71 37 | 4.58 73 | 1.78 45 | 0.82 23 | 2.05 30 | 0.98 31 | 0.96 33 | 2.07 46 | 1.13 34 | 0.54 16 | 1.52 45 | 0.48 17 | 0.90 32 | 1.37 39 | 0.88 27 | 0.55 13 | 2.25 20 | 0.67 6 | 1.47 19 | 2.78 52 | 1.43 40 | 0.76 66 | 1.08 56 | 0.81 67 |
FMOF [94] | 36.3 | 1.51 22 | 2.53 20 | 1.58 19 | 0.95 60 | 2.85 75 | 1.11 62 | 0.93 22 | 1.58 21 | 1.17 58 | 0.52 13 | 1.26 30 | 0.46 14 | 0.81 17 | 1.14 14 | 0.88 27 | 0.77 54 | 2.25 20 | 0.81 35 | 1.97 58 | 2.76 49 | 2.45 95 | 0.64 37 | 0.74 15 | 0.67 34 |
FESL [72] | 37.3 | 1.52 23 | 2.27 10 | 1.77 43 | 0.94 55 | 2.76 67 | 1.08 53 | 0.92 20 | 1.59 22 | 1.13 34 | 0.61 39 | 1.30 33 | 0.56 43 | 0.84 22 | 1.16 18 | 0.90 34 | 0.75 49 | 2.14 15 | 0.96 61 | 2.05 67 | 3.16 70 | 2.08 71 | 0.57 20 | 0.64 6 | 0.61 20 |
Adaptive [20] | 38.9 | 1.50 20 | 2.72 33 | 1.34 7 | 0.88 38 | 2.24 38 | 1.00 36 | 1.03 44 | 2.29 54 | 1.13 34 | 0.57 28 | 1.65 53 | 0.49 22 | 2.18 113 | 2.69 109 | 2.56 111 | 0.45 4 | 2.26 22 | 0.49 3 | 1.73 42 | 2.94 64 | 1.23 21 | 0.54 17 | 0.63 5 | 0.56 16 |
Classic+NL [31] | 39.3 | 1.55 27 | 2.35 13 | 1.70 34 | 0.88 38 | 2.41 49 | 1.08 53 | 0.94 26 | 1.62 23 | 1.15 53 | 0.58 33 | 1.19 24 | 0.52 33 | 0.90 32 | 1.30 34 | 0.96 44 | 0.74 45 | 2.76 36 | 0.85 44 | 2.23 77 | 2.64 40 | 2.27 83 | 0.63 31 | 0.76 24 | 0.69 47 |
PMMST [114] | 39.5 | 2.17 75 | 3.12 43 | 2.08 68 | 0.88 38 | 2.04 29 | 1.02 41 | 1.03 44 | 1.74 32 | 1.09 30 | 0.84 73 | 1.11 13 | 0.78 73 | 0.92 34 | 1.14 14 | 0.85 20 | 0.76 51 | 2.03 8 | 0.80 31 | 1.60 29 | 2.54 28 | 1.29 24 | 0.73 60 | 0.94 40 | 0.70 50 |
SimpleFlow [49] | 40.5 | 1.56 29 | 2.63 28 | 1.74 42 | 0.94 55 | 2.69 61 | 1.18 79 | 0.98 39 | 1.96 40 | 1.20 67 | 0.56 25 | 1.23 28 | 0.50 27 | 0.92 34 | 1.36 38 | 1.02 55 | 0.85 62 | 2.74 35 | 0.89 50 | 1.85 49 | 2.38 14 | 1.71 51 | 0.62 26 | 0.73 11 | 0.64 28 |
TV-L1-MCT [64] | 40.6 | 1.48 15 | 2.45 18 | 1.52 13 | 1.00 70 | 2.95 81 | 1.16 74 | 0.94 26 | 1.67 24 | 1.18 63 | 0.56 25 | 1.28 32 | 0.49 22 | 0.96 38 | 1.37 39 | 1.11 61 | 0.74 45 | 2.87 44 | 0.90 54 | 1.60 29 | 2.62 39 | 1.05 13 | 0.71 52 | 0.85 33 | 0.80 64 |
S2D-Matching [84] | 40.9 | 1.57 31 | 2.52 19 | 1.71 35 | 0.88 38 | 2.34 44 | 1.08 53 | 0.96 33 | 1.75 33 | 1.13 34 | 0.58 33 | 1.15 17 | 0.52 33 | 0.89 31 | 1.27 29 | 0.92 37 | 0.76 51 | 2.90 45 | 0.89 50 | 2.35 91 | 2.71 46 | 2.53 100 | 0.64 37 | 0.74 15 | 0.69 47 |
AggregFlow [97] | 45.1 | 1.86 49 | 2.73 34 | 1.91 54 | 1.01 76 | 2.88 77 | 1.14 66 | 1.09 59 | 2.34 56 | 1.29 75 | 0.76 62 | 1.46 40 | 0.70 63 | 0.82 18 | 1.35 37 | 0.88 27 | 0.54 11 | 1.53 2 | 0.70 14 | 1.50 21 | 2.35 13 | 0.90 4 | 0.78 77 | 0.97 48 | 1.23 99 |
Classic++ [32] | 45.9 | 1.57 31 | 2.57 23 | 1.72 40 | 0.87 33 | 2.03 26 | 1.08 53 | 0.96 33 | 1.98 42 | 1.15 53 | 0.58 33 | 1.41 36 | 0.52 33 | 1.03 44 | 1.80 58 | 1.04 58 | 0.75 49 | 3.62 68 | 0.86 46 | 2.32 84 | 2.84 55 | 2.48 97 | 0.64 37 | 0.89 35 | 0.67 34 |
Occlusion-TV-L1 [63] | 46.3 | 1.82 47 | 3.15 44 | 1.64 29 | 0.91 47 | 2.12 33 | 1.04 46 | 1.13 67 | 2.49 58 | 1.19 66 | 0.70 48 | 1.75 54 | 0.62 48 | 1.28 72 | 1.97 75 | 1.44 81 | 0.61 18 | 2.77 38 | 0.83 40 | 1.59 27 | 2.65 41 | 1.05 13 | 0.65 40 | 1.00 51 | 0.64 28 |
OFH [38] | 47.7 | 2.14 72 | 3.60 50 | 2.44 81 | 0.76 13 | 1.81 13 | 0.87 18 | 0.88 10 | 2.28 52 | 0.90 4 | 0.55 21 | 1.20 25 | 0.52 33 | 1.20 59 | 1.83 63 | 1.37 77 | 0.84 60 | 4.30 87 | 1.05 68 | 1.37 10 | 2.85 57 | 1.49 43 | 0.76 66 | 1.35 79 | 0.96 83 |
IROF-TV [53] | 47.8 | 1.69 36 | 2.85 38 | 1.84 49 | 0.90 46 | 2.58 59 | 1.10 59 | 0.95 30 | 1.86 37 | 1.15 53 | 0.74 59 | 1.80 58 | 0.68 60 | 1.41 81 | 1.71 52 | 1.51 84 | 0.90 69 | 3.88 77 | 1.03 67 | 1.30 4 | 2.41 17 | 0.82 2 | 0.65 40 | 0.80 28 | 0.68 42 |
PMF [73] | 48.2 | 2.05 63 | 2.90 39 | 1.89 53 | 0.87 33 | 1.97 23 | 0.96 28 | 1.05 49 | 1.83 36 | 1.15 53 | 0.87 75 | 1.45 39 | 0.80 75 | 0.85 24 | 1.08 9 | 0.73 10 | 0.98 76 | 3.46 63 | 1.10 75 | 3.02 107 | 4.28 110 | 2.92 107 | 0.36 5 | 0.60 3 | 0.27 3 |
DeepFlow2 [108] | 48.4 | 2.02 57 | 4.03 61 | 2.29 76 | 0.81 21 | 2.27 40 | 0.93 25 | 1.09 59 | 2.78 65 | 1.27 73 | 0.66 46 | 1.77 56 | 0.60 46 | 0.93 36 | 1.59 44 | 0.86 25 | 0.57 14 | 2.93 46 | 0.68 8 | 1.66 34 | 2.25 6 | 1.97 65 | 0.83 84 | 1.44 86 | 1.03 89 |
MDP-Flow [26] | 48.5 | 1.90 51 | 3.71 54 | 1.94 58 | 0.92 50 | 1.90 18 | 1.15 68 | 1.01 42 | 2.00 45 | 1.14 45 | 0.70 48 | 1.76 55 | 0.64 54 | 1.06 45 | 1.59 44 | 0.96 44 | 0.68 31 | 3.58 67 | 0.84 41 | 1.68 38 | 3.01 65 | 1.19 19 | 0.75 64 | 1.33 76 | 0.68 42 |
Aniso-Texture [82] | 48.7 | 1.56 29 | 2.57 23 | 1.58 19 | 0.92 50 | 1.80 12 | 1.15 68 | 0.93 22 | 1.70 28 | 1.18 63 | 0.61 39 | 1.46 40 | 0.55 40 | 1.20 59 | 1.80 58 | 1.05 59 | 0.91 71 | 2.99 51 | 1.07 71 | 3.29 114 | 4.61 116 | 2.66 104 | 0.48 12 | 0.73 11 | 0.43 9 |
SVFilterOh [111] | 48.7 | 2.00 56 | 2.79 35 | 1.99 62 | 0.99 67 | 1.96 21 | 1.12 63 | 1.07 55 | 1.56 18 | 1.17 58 | 0.83 72 | 1.82 61 | 0.73 69 | 0.82 18 | 1.04 6 | 0.79 16 | 0.90 69 | 2.23 18 | 0.97 63 | 2.60 98 | 4.73 118 | 3.08 109 | 0.42 9 | 0.55 2 | 0.34 5 |
BriefMatch [124] | 49.3 | 1.91 52 | 3.16 45 | 1.92 55 | 0.77 15 | 1.87 16 | 0.83 10 | 0.96 33 | 1.53 14 | 0.95 9 | 0.78 64 | 1.37 35 | 0.71 65 | 0.97 40 | 1.54 42 | 1.02 55 | 1.69 111 | 4.79 90 | 1.92 109 | 2.34 89 | 3.28 78 | 3.43 115 | 0.40 6 | 0.76 24 | 0.48 11 |
Correlation Flow [75] | 49.8 | 1.96 55 | 2.90 39 | 2.04 65 | 0.82 23 | 2.03 26 | 0.88 20 | 1.04 47 | 1.99 44 | 1.02 15 | 0.75 61 | 1.51 44 | 0.68 60 | 1.12 53 | 1.58 43 | 0.98 50 | 1.04 78 | 3.01 52 | 1.14 79 | 2.01 63 | 2.61 38 | 2.35 91 | 0.71 52 | 0.98 50 | 0.69 47 |
S2F-IF [123] | 51.0 | 1.80 42 | 5.50 95 | 1.68 32 | 0.95 60 | 2.97 83 | 1.10 59 | 1.13 67 | 3.58 87 | 1.28 74 | 0.58 33 | 2.17 74 | 0.50 27 | 1.09 50 | 1.86 66 | 0.93 38 | 0.52 7 | 2.93 46 | 0.67 6 | 1.36 9 | 2.58 34 | 1.37 30 | 0.77 72 | 1.22 65 | 0.82 69 |
Kuang [131] | 52.0 | 1.67 34 | 5.53 96 | 1.59 22 | 0.91 47 | 3.32 96 | 1.01 37 | 1.06 53 | 4.03 93 | 1.14 45 | 0.62 43 | 1.94 66 | 0.56 43 | 1.19 58 | 2.09 81 | 0.95 41 | 0.62 19 | 3.87 75 | 0.80 31 | 1.33 8 | 2.55 30 | 1.40 33 | 0.74 62 | 1.34 77 | 0.75 58 |
PGM-C [120] | 52.8 | 1.81 43 | 5.14 83 | 1.71 35 | 1.00 70 | 2.94 80 | 1.17 76 | 1.14 70 | 3.46 85 | 1.31 77 | 0.60 37 | 2.30 78 | 0.52 33 | 1.06 45 | 1.82 62 | 0.93 38 | 0.53 10 | 2.95 48 | 0.68 8 | 1.42 14 | 2.51 25 | 1.41 36 | 0.77 72 | 1.29 69 | 0.84 72 |
CPM-Flow [116] | 53.0 | 1.81 43 | 5.14 83 | 1.71 35 | 1.00 70 | 2.93 78 | 1.17 76 | 1.14 70 | 3.37 83 | 1.31 77 | 0.60 37 | 2.30 78 | 0.52 33 | 1.07 47 | 1.83 63 | 0.95 41 | 0.54 11 | 2.81 41 | 0.68 8 | 1.45 17 | 2.57 33 | 1.41 36 | 0.77 72 | 1.29 69 | 0.84 72 |
FlowFields+ [130] | 53.0 | 1.78 39 | 5.67 99 | 1.63 27 | 0.99 67 | 3.15 93 | 1.15 68 | 1.15 75 | 3.92 91 | 1.32 81 | 0.57 28 | 2.13 73 | 0.48 17 | 1.13 54 | 1.89 69 | 0.97 48 | 0.52 7 | 3.12 55 | 0.68 8 | 1.37 10 | 2.66 42 | 1.31 26 | 0.76 66 | 1.27 68 | 0.79 62 |
TV-L1-improved [17] | 54.7 | 1.58 33 | 3.18 46 | 1.55 15 | 0.78 18 | 1.98 24 | 0.90 21 | 0.99 41 | 2.28 52 | 1.07 24 | 0.55 21 | 1.52 45 | 0.48 17 | 1.32 77 | 2.10 83 | 1.01 52 | 1.82 113 | 6.46 108 | 2.25 114 | 2.59 96 | 3.51 89 | 2.52 99 | 0.65 40 | 1.17 64 | 0.62 21 |
EpicFlow [102] | 55.4 | 1.81 43 | 5.16 86 | 1.71 35 | 1.00 70 | 2.98 85 | 1.17 76 | 1.14 70 | 3.63 89 | 1.31 77 | 0.61 39 | 2.31 80 | 0.52 33 | 1.07 47 | 1.86 66 | 0.97 48 | 0.59 16 | 2.95 48 | 0.70 14 | 1.42 14 | 2.60 36 | 1.41 36 | 0.78 77 | 1.31 73 | 0.84 72 |
FlowFields [110] | 56.8 | 1.81 43 | 5.57 97 | 1.71 35 | 0.99 67 | 3.12 92 | 1.15 68 | 1.16 78 | 3.92 91 | 1.32 81 | 0.63 44 | 2.19 75 | 0.55 40 | 1.13 54 | 1.91 72 | 1.00 51 | 0.52 7 | 3.18 57 | 0.68 8 | 1.37 10 | 2.77 51 | 1.38 32 | 0.77 72 | 1.31 73 | 0.80 64 |
CostFilter [40] | 58.0 | 2.22 77 | 3.43 48 | 2.15 71 | 0.96 64 | 2.10 32 | 1.07 51 | 1.10 62 | 2.13 48 | 1.17 58 | 1.10 90 | 1.58 50 | 1.06 91 | 0.88 30 | 1.13 12 | 0.85 20 | 1.07 81 | 3.96 80 | 1.24 82 | 3.04 108 | 4.87 122 | 3.15 111 | 0.15 1 | 0.34 1 | 0.15 1 |
DeepFlow [86] | 59.6 | 2.36 85 | 4.50 72 | 3.07 98 | 0.88 38 | 2.30 42 | 1.01 37 | 1.15 75 | 3.29 80 | 1.38 86 | 0.82 71 | 1.80 58 | 0.77 72 | 0.95 37 | 1.66 48 | 0.87 26 | 0.59 16 | 3.80 71 | 0.70 14 | 1.59 27 | 2.31 10 | 2.02 69 | 0.94 97 | 1.68 100 | 1.28 101 |
Steered-L1 [118] | 61.5 | 2.04 60 | 3.66 52 | 2.29 76 | 0.71 10 | 1.31 1 | 0.81 9 | 0.97 38 | 1.89 38 | 0.99 13 | 0.72 57 | 1.42 37 | 0.66 59 | 1.26 67 | 1.81 60 | 1.39 79 | 1.02 77 | 4.27 86 | 0.96 61 | 3.23 112 | 3.77 96 | 5.23 124 | 0.87 88 | 1.40 80 | 1.13 95 |
TF+OM [100] | 62.4 | 2.02 57 | 2.92 42 | 1.85 51 | 0.93 54 | 1.53 4 | 1.15 68 | 1.05 49 | 1.69 27 | 1.37 85 | 1.12 91 | 1.35 34 | 1.15 93 | 1.07 47 | 1.45 41 | 1.52 85 | 1.06 79 | 2.82 42 | 1.22 81 | 2.45 92 | 3.60 92 | 2.19 78 | 0.74 62 | 1.25 67 | 0.87 76 |
Sparse Occlusion [54] | 62.4 | 2.04 60 | 3.32 47 | 1.92 55 | 1.08 83 | 2.38 48 | 1.27 85 | 1.11 65 | 2.16 49 | 1.18 63 | 0.74 59 | 1.56 49 | 0.65 57 | 1.21 61 | 1.76 56 | 0.81 17 | 0.91 71 | 2.84 43 | 0.98 64 | 3.84 121 | 4.85 121 | 2.41 93 | 0.71 52 | 1.08 56 | 0.63 23 |
RFlow [90] | 63.0 | 2.23 79 | 4.33 68 | 2.52 85 | 0.96 64 | 1.84 15 | 1.05 48 | 1.08 57 | 2.71 61 | 1.05 21 | 0.76 62 | 1.58 50 | 0.71 65 | 1.27 70 | 1.89 69 | 1.31 73 | 0.80 56 | 3.40 62 | 0.94 58 | 2.00 62 | 2.76 49 | 1.98 67 | 0.93 96 | 1.42 82 | 1.12 94 |
CombBMOF [113] | 63.9 | 2.19 76 | 4.69 77 | 1.87 52 | 1.01 76 | 2.82 73 | 1.08 53 | 1.03 44 | 2.21 50 | 1.07 24 | 0.86 74 | 1.89 64 | 0.78 73 | 1.27 70 | 1.71 52 | 1.03 57 | 1.39 96 | 3.89 78 | 1.61 96 | 2.80 104 | 3.78 97 | 2.30 87 | 0.49 13 | 0.88 34 | 0.50 14 |
Second-order prior [8] | 64.0 | 1.95 54 | 4.68 76 | 2.03 63 | 0.79 19 | 2.65 60 | 0.84 12 | 1.10 62 | 3.80 90 | 1.14 45 | 0.54 16 | 1.52 45 | 0.47 15 | 1.42 84 | 2.45 103 | 0.94 40 | 0.88 65 | 6.63 109 | 0.89 50 | 2.87 106 | 3.46 86 | 2.67 105 | 0.77 72 | 1.60 96 | 0.80 64 |
EPPM w/o HM [88] | 64.7 | 2.15 74 | 5.62 98 | 2.03 63 | 0.88 38 | 2.76 67 | 0.91 23 | 1.06 53 | 3.03 73 | 1.08 27 | 0.89 78 | 2.04 71 | 0.83 80 | 1.22 62 | 1.66 48 | 1.12 62 | 1.19 85 | 5.06 96 | 1.37 87 | 2.25 79 | 3.56 90 | 3.98 120 | 0.51 16 | 1.00 51 | 0.48 11 |
LDOF [28] | 66.4 | 2.07 65 | 4.67 75 | 2.39 79 | 0.92 50 | 3.01 87 | 1.03 43 | 1.22 87 | 3.52 86 | 1.22 69 | 0.73 58 | 3.52 99 | 0.62 48 | 1.18 56 | 1.94 73 | 1.25 69 | 0.68 31 | 4.07 85 | 0.74 21 | 1.70 40 | 2.87 59 | 1.29 24 | 0.89 90 | 1.82 108 | 1.08 91 |
MLDP_OF [89] | 66.6 | 2.44 88 | 5.18 87 | 2.59 87 | 0.95 60 | 2.46 54 | 1.04 46 | 1.15 75 | 2.81 66 | 1.13 34 | 0.78 64 | 1.60 52 | 0.71 65 | 1.22 62 | 1.68 51 | 1.16 65 | 1.06 79 | 3.17 56 | 1.28 84 | 2.45 92 | 3.17 72 | 3.24 113 | 0.68 48 | 0.97 48 | 0.70 50 |
Aniso. Huber-L1 [22] | 68.1 | 1.79 40 | 3.70 53 | 1.82 47 | 1.37 93 | 3.01 87 | 1.79 94 | 1.19 83 | 3.00 71 | 1.68 92 | 0.79 68 | 2.49 83 | 0.70 63 | 1.26 67 | 1.96 74 | 0.96 44 | 0.81 57 | 3.19 58 | 0.95 60 | 2.54 94 | 3.34 80 | 1.99 68 | 0.71 52 | 1.03 54 | 0.73 53 |
Rannacher [23] | 68.5 | 2.22 77 | 4.10 63 | 2.36 78 | 1.02 79 | 2.44 52 | 1.20 82 | 1.23 88 | 3.15 74 | 1.29 75 | 0.70 48 | 1.88 63 | 0.62 48 | 1.37 80 | 2.27 90 | 1.16 65 | 1.14 83 | 5.21 98 | 1.11 77 | 2.21 75 | 2.88 63 | 1.97 65 | 0.66 45 | 0.95 42 | 0.67 34 |
F-TV-L1 [15] | 69.7 | 3.66 105 | 6.22 101 | 4.52 111 | 1.19 87 | 2.78 70 | 1.40 88 | 1.21 85 | 3.24 78 | 1.31 77 | 1.15 92 | 2.69 87 | 1.07 92 | 1.70 100 | 2.30 92 | 1.86 94 | 0.72 41 | 3.33 60 | 0.90 54 | 1.76 45 | 2.74 47 | 1.44 41 | 0.46 10 | 0.61 4 | 0.48 11 |
FlowNetS+ft+v [112] | 69.8 | 1.85 48 | 4.25 65 | 2.07 66 | 0.92 50 | 2.76 67 | 1.06 49 | 1.16 78 | 3.42 84 | 1.42 87 | 0.68 47 | 2.62 85 | 0.60 46 | 1.53 91 | 2.44 102 | 1.36 76 | 0.72 41 | 3.63 69 | 0.80 31 | 2.30 81 | 3.46 86 | 1.96 63 | 0.82 82 | 1.58 95 | 0.97 85 |
Complementary OF [21] | 69.9 | 2.60 95 | 5.23 89 | 2.96 96 | 0.83 27 | 1.75 11 | 0.93 25 | 1.12 66 | 2.23 51 | 1.20 67 | 1.05 86 | 1.52 45 | 1.02 90 | 1.24 64 | 1.79 57 | 1.56 87 | 1.21 86 | 4.83 91 | 1.25 83 | 1.73 42 | 2.60 36 | 1.87 59 | 1.07 108 | 1.69 102 | 1.55 114 |
ComplOF-FED-GPU [35] | 70.1 | 2.48 91 | 5.29 91 | 2.73 90 | 0.79 19 | 2.20 36 | 0.86 15 | 1.07 55 | 2.73 62 | 1.04 17 | 0.88 77 | 1.46 40 | 0.83 80 | 1.30 75 | 2.00 78 | 1.31 73 | 1.28 91 | 5.12 97 | 1.32 85 | 2.33 86 | 3.03 67 | 2.54 101 | 0.85 86 | 1.43 84 | 0.99 87 |
TriangleFlow [30] | 71.4 | 2.07 65 | 3.84 57 | 2.12 69 | 0.94 55 | 2.79 71 | 0.98 31 | 1.10 62 | 2.81 66 | 1.06 23 | 0.57 28 | 1.81 60 | 0.49 22 | 1.81 104 | 2.82 111 | 1.91 96 | 1.43 98 | 4.91 93 | 1.66 98 | 2.08 70 | 3.66 95 | 1.93 61 | 0.84 85 | 1.62 97 | 1.15 96 |
TCOF [69] | 71.7 | 2.27 81 | 4.40 71 | 2.56 86 | 1.07 81 | 2.95 81 | 1.19 80 | 1.24 89 | 3.15 74 | 1.50 89 | 1.30 95 | 1.99 68 | 1.40 97 | 1.49 90 | 2.38 96 | 1.06 60 | 0.70 36 | 2.06 10 | 0.89 50 | 2.63 99 | 3.64 94 | 1.25 22 | 0.76 66 | 1.30 72 | 0.67 34 |
DF-Auto [115] | 72.0 | 2.06 64 | 4.06 62 | 1.72 40 | 1.83 99 | 4.07 102 | 2.46 100 | 1.43 96 | 4.16 95 | 2.15 100 | 1.06 87 | 2.83 90 | 0.97 87 | 1.18 56 | 1.81 60 | 1.42 80 | 0.51 6 | 1.82 5 | 0.72 18 | 2.21 75 | 3.87 99 | 1.02 11 | 0.98 99 | 1.82 108 | 1.05 90 |
SuperFlow [81] | 72.7 | 1.79 40 | 3.74 55 | 1.80 46 | 1.37 93 | 3.00 86 | 1.90 95 | 1.05 49 | 2.84 68 | 1.75 95 | 1.38 99 | 3.64 100 | 1.48 101 | 1.25 66 | 1.71 52 | 1.94 97 | 0.68 31 | 3.53 65 | 0.85 44 | 2.30 81 | 3.28 78 | 1.41 36 | 0.79 80 | 1.64 99 | 1.01 88 |
Brox et al. [5] | 73.0 | 2.11 69 | 5.20 88 | 2.91 94 | 1.01 76 | 2.81 72 | 1.19 80 | 1.13 67 | 3.02 72 | 1.17 58 | 0.71 53 | 2.49 83 | 0.63 53 | 1.45 87 | 2.09 81 | 2.27 106 | 0.84 60 | 4.00 81 | 1.05 68 | 1.69 39 | 3.01 65 | 0.95 8 | 0.92 94 | 1.81 107 | 1.08 91 |
HBM-GC [105] | 73.3 | 3.83 107 | 4.28 67 | 3.79 102 | 1.43 95 | 2.46 54 | 1.65 92 | 1.68 100 | 2.38 57 | 1.73 93 | 1.61 101 | 2.23 76 | 1.47 100 | 1.09 50 | 1.29 31 | 1.14 63 | 1.24 90 | 2.69 34 | 1.41 90 | 3.76 120 | 4.84 120 | 3.05 108 | 0.21 2 | 0.64 6 | 0.22 2 |
ACK-Prior [27] | 73.5 | 2.81 97 | 3.93 60 | 2.98 97 | 0.91 47 | 1.91 19 | 0.97 30 | 1.09 59 | 1.97 41 | 1.13 34 | 0.97 84 | 1.85 62 | 0.85 83 | 1.24 64 | 1.67 50 | 1.30 72 | 1.56 105 | 3.87 75 | 1.53 93 | 3.14 109 | 3.16 70 | 4.47 122 | 1.08 110 | 1.41 81 | 1.27 100 |
CNN-flow-warp+ref [117] | 74.0 | 2.10 68 | 5.00 80 | 2.49 84 | 1.21 89 | 2.97 83 | 1.59 90 | 1.29 92 | 4.29 96 | 1.77 96 | 0.78 64 | 3.05 95 | 0.68 60 | 1.35 79 | 2.01 79 | 1.81 91 | 0.73 44 | 4.03 82 | 0.87 49 | 1.42 14 | 2.54 28 | 1.06 15 | 0.89 90 | 1.70 103 | 1.32 105 |
LocallyOriented [52] | 74.1 | 2.04 60 | 3.52 49 | 1.97 60 | 1.12 84 | 3.79 101 | 1.25 83 | 1.21 85 | 3.59 88 | 1.33 83 | 0.87 75 | 1.77 56 | 0.82 77 | 1.47 88 | 2.21 87 | 1.45 83 | 0.88 65 | 2.76 36 | 1.06 70 | 2.03 66 | 3.17 72 | 1.96 63 | 0.81 81 | 1.47 89 | 0.88 77 |
Bartels [41] | 75.0 | 2.42 87 | 3.60 50 | 3.08 99 | 1.15 86 | 1.94 20 | 1.41 89 | 1.17 80 | 2.07 46 | 1.34 84 | 1.33 97 | 1.96 67 | 1.33 96 | 1.28 72 | 1.87 68 | 1.81 91 | 1.22 89 | 3.82 73 | 1.86 108 | 2.65 101 | 3.84 98 | 3.33 114 | 0.57 20 | 0.96 46 | 0.57 18 |
Local-TV-L1 [65] | 75.7 | 2.87 98 | 5.15 85 | 3.73 101 | 1.72 97 | 3.02 89 | 2.23 98 | 1.52 98 | 4.41 98 | 1.84 97 | 1.19 93 | 2.74 88 | 1.20 94 | 1.01 43 | 1.60 46 | 1.01 52 | 0.65 26 | 3.50 64 | 0.81 35 | 1.52 22 | 2.38 14 | 1.80 54 | 1.00 103 | 1.90 110 | 1.43 111 |
CBF [12] | 76.2 | 2.11 69 | 4.34 70 | 2.45 82 | 1.74 98 | 2.70 63 | 2.64 102 | 1.08 57 | 2.57 60 | 1.25 72 | 0.71 53 | 2.03 70 | 0.62 48 | 1.44 85 | 2.08 80 | 1.22 68 | 0.89 67 | 3.26 59 | 1.09 73 | 3.31 115 | 3.98 103 | 2.65 103 | 0.82 82 | 1.31 73 | 0.88 77 |
CRTflow [80] | 76.2 | 2.07 65 | 4.91 79 | 1.98 61 | 1.00 70 | 2.45 53 | 1.12 63 | 1.05 49 | 3.25 79 | 1.05 21 | 0.79 68 | 1.91 65 | 0.73 69 | 1.29 74 | 1.97 75 | 1.28 71 | 2.16 115 | 6.87 110 | 2.90 118 | 2.02 64 | 3.44 84 | 1.95 62 | 1.03 106 | 1.78 105 | 1.30 104 |
Dynamic MRF [7] | 77.0 | 2.47 90 | 5.33 92 | 2.82 93 | 0.83 27 | 2.26 39 | 0.90 21 | 1.02 43 | 3.21 76 | 1.02 15 | 0.79 68 | 2.10 72 | 0.74 71 | 1.68 99 | 2.40 99 | 2.10 101 | 1.48 102 | 7.55 113 | 1.71 99 | 1.96 56 | 2.87 59 | 2.82 106 | 0.98 99 | 1.63 98 | 1.39 110 |
DPOF [18] | 77.0 | 2.25 80 | 5.04 82 | 1.92 55 | 1.05 80 | 3.43 97 | 1.15 68 | 1.14 70 | 2.94 69 | 1.24 70 | 0.92 80 | 3.04 94 | 0.82 77 | 1.09 50 | 1.83 63 | 0.89 30 | 1.08 82 | 3.79 70 | 1.10 75 | 2.34 89 | 2.80 54 | 5.03 123 | 1.01 104 | 1.47 89 | 1.17 97 |
SRR-TVOF-NL [91] | 77.1 | 2.36 85 | 4.25 65 | 2.14 70 | 0.96 64 | 2.84 74 | 1.02 41 | 1.14 70 | 3.33 81 | 1.24 70 | 0.71 53 | 2.28 77 | 0.62 48 | 1.41 81 | 1.90 71 | 1.27 70 | 0.89 67 | 3.55 66 | 1.02 66 | 3.32 116 | 4.34 112 | 2.47 96 | 1.10 113 | 1.46 87 | 1.36 108 |
CLG-TV [48] | 77.8 | 2.12 71 | 3.91 59 | 2.20 72 | 1.56 96 | 2.93 78 | 2.16 96 | 1.36 94 | 3.36 82 | 1.84 97 | 1.09 89 | 3.37 98 | 0.98 88 | 1.44 85 | 2.22 88 | 1.37 77 | 0.85 62 | 4.38 88 | 1.12 78 | 2.30 81 | 3.06 68 | 1.54 47 | 0.72 56 | 1.11 62 | 0.74 56 |
SIOF [67] | 78.3 | 2.51 93 | 3.80 56 | 2.39 79 | 1.00 70 | 2.31 43 | 1.12 63 | 1.41 95 | 2.99 70 | 1.58 91 | 1.34 98 | 2.32 81 | 1.41 98 | 1.48 89 | 2.11 84 | 1.59 88 | 1.14 83 | 3.81 72 | 1.37 87 | 1.94 53 | 2.69 44 | 1.35 28 | 1.15 116 | 1.48 91 | 1.36 108 |
NL-TV-NCC [25] | 80.8 | 2.32 83 | 3.85 58 | 2.27 73 | 1.13 85 | 3.05 91 | 1.16 74 | 1.18 82 | 2.31 55 | 1.17 58 | 0.93 81 | 2.02 69 | 0.80 75 | 1.54 92 | 2.41 100 | 1.01 52 | 1.44 100 | 4.66 89 | 1.52 92 | 2.32 84 | 4.08 108 | 2.30 87 | 0.90 93 | 1.43 84 | 0.85 75 |
ROF-ND [107] | 81.4 | 2.46 89 | 4.33 68 | 2.48 83 | 1.26 90 | 2.23 37 | 1.25 83 | 1.17 80 | 2.50 59 | 1.15 53 | 1.08 88 | 2.75 89 | 0.92 84 | 1.41 81 | 2.11 84 | 1.31 73 | 1.40 97 | 3.85 74 | 1.38 89 | 3.26 113 | 4.06 106 | 3.09 110 | 0.86 87 | 1.29 69 | 0.81 67 |
p-harmonic [29] | 83.3 | 2.52 94 | 7.07 105 | 2.78 91 | 1.19 87 | 2.87 76 | 1.29 86 | 1.46 97 | 4.64 100 | 1.46 88 | 0.91 79 | 3.66 101 | 0.82 77 | 1.66 97 | 2.25 89 | 1.82 93 | 0.86 64 | 5.52 101 | 1.14 79 | 2.29 80 | 3.36 81 | 1.80 54 | 0.75 64 | 1.14 63 | 0.73 53 |
TriFlow [95] | 83.9 | 2.34 84 | 4.21 64 | 2.07 66 | 1.34 92 | 2.46 54 | 1.75 93 | 1.26 90 | 2.75 63 | 1.74 94 | 1.31 96 | 2.46 82 | 1.22 95 | 1.30 75 | 1.72 55 | 1.88 95 | 0.95 75 | 2.79 40 | 1.08 72 | 4.93 126 | 4.06 106 | 16.4 128 | 0.89 90 | 1.49 92 | 0.98 86 |
Learning Flow [11] | 84.2 | 2.14 72 | 4.65 74 | 2.28 74 | 1.32 91 | 3.15 93 | 1.63 91 | 1.27 91 | 3.23 77 | 1.52 90 | 0.94 82 | 3.23 96 | 0.83 80 | 1.86 106 | 2.85 112 | 2.31 107 | 1.21 86 | 4.99 95 | 1.36 86 | 2.33 86 | 3.44 84 | 2.08 71 | 0.73 60 | 1.23 66 | 0.72 52 |
StereoFlow [44] | 85.2 | 11.7 129 | 23.3 129 | 12.9 127 | 10.4 128 | 17.4 129 | 9.59 124 | 10.3 129 | 21.5 129 | 5.63 121 | 14.7 129 | 21.8 127 | 12.3 127 | 2.66 117 | 3.01 114 | 2.67 112 | 0.23 1 | 1.30 1 | 0.31 1 | 0.88 1 | 2.22 4 | 0.54 1 | 0.72 56 | 0.96 46 | 0.79 62 |
Fusion [6] | 85.4 | 2.02 57 | 6.55 103 | 2.28 74 | 0.87 33 | 2.50 57 | 1.01 37 | 1.04 47 | 2.76 64 | 1.14 45 | 0.78 64 | 2.85 91 | 0.72 68 | 1.87 107 | 2.39 98 | 2.25 104 | 1.62 107 | 5.52 101 | 2.03 113 | 3.47 118 | 4.40 114 | 2.25 82 | 1.98 125 | 2.15 117 | 2.60 123 |
SegOF [10] | 90.2 | 2.88 99 | 5.24 90 | 1.95 59 | 3.25 112 | 5.52 112 | 4.24 114 | 2.17 106 | 5.82 104 | 3.40 111 | 1.77 102 | 4.89 106 | 1.51 102 | 1.94 108 | 2.32 93 | 2.83 114 | 1.36 93 | 6.92 111 | 1.59 94 | 1.32 6 | 3.10 69 | 0.93 7 | 0.87 88 | 1.42 82 | 0.95 82 |
Shiralkar [42] | 90.5 | 2.29 82 | 9.09 114 | 2.70 89 | 0.89 45 | 3.49 98 | 0.95 27 | 1.20 84 | 5.92 105 | 1.07 24 | 0.96 83 | 3.95 102 | 0.92 84 | 1.60 94 | 2.47 105 | 1.60 89 | 1.66 109 | 7.74 114 | 1.83 107 | 2.63 99 | 3.41 82 | 3.86 118 | 0.99 102 | 2.01 114 | 1.28 101 |
StereoOF-V1MT [119] | 92.5 | 2.48 91 | 8.60 110 | 2.79 92 | 0.94 55 | 5.07 110 | 0.92 24 | 1.33 93 | 7.67 109 | 1.10 31 | 0.99 85 | 5.20 107 | 0.93 86 | 2.20 114 | 3.24 115 | 2.31 107 | 1.69 111 | 9.96 119 | 1.75 101 | 1.99 61 | 3.46 86 | 2.42 94 | 1.07 108 | 1.99 113 | 1.19 98 |
Filter Flow [19] | 92.6 | 3.22 100 | 5.46 94 | 2.91 94 | 1.91 101 | 4.47 106 | 2.45 99 | 1.99 102 | 5.00 101 | 2.64 103 | 2.64 114 | 7.42 114 | 2.52 110 | 2.02 110 | 2.47 105 | 2.90 115 | 1.54 103 | 5.42 99 | 1.80 105 | 4.36 124 | 5.78 126 | 2.11 75 | 0.31 4 | 0.74 15 | 0.32 4 |
BlockOverlap [61] | 92.8 | 4.29 111 | 5.43 93 | 4.08 106 | 2.37 106 | 3.20 95 | 3.03 107 | 2.32 109 | 4.36 97 | 2.64 103 | 2.44 110 | 2.87 92 | 2.58 111 | 1.34 78 | 1.60 46 | 2.16 103 | 1.68 110 | 3.89 78 | 1.75 101 | 4.00 122 | 4.80 119 | 4.06 121 | 0.30 3 | 1.02 53 | 0.73 53 |
Ad-TV-NDC [36] | 93.0 | 4.41 114 | 6.97 104 | 7.36 122 | 3.30 113 | 4.58 107 | 4.69 116 | 2.60 110 | 6.97 107 | 3.33 110 | 2.16 107 | 4.56 105 | 2.33 108 | 1.26 67 | 1.99 77 | 1.14 63 | 0.91 71 | 3.37 61 | 1.09 73 | 1.88 51 | 2.74 47 | 1.62 48 | 1.09 112 | 2.40 121 | 1.93 118 |
Modified CLG [34] | 93.2 | 3.61 104 | 7.80 108 | 3.85 103 | 2.69 109 | 4.23 105 | 3.87 112 | 2.73 112 | 9.13 112 | 3.49 112 | 2.34 108 | 5.73 110 | 2.33 108 | 1.57 93 | 2.45 103 | 2.02 100 | 0.74 45 | 5.44 100 | 0.93 57 | 1.64 33 | 2.84 55 | 1.12 17 | 1.08 110 | 2.09 116 | 1.33 106 |
IAOF2 [51] | 98.4 | 2.79 96 | 4.89 78 | 2.69 88 | 1.86 100 | 3.78 100 | 2.57 101 | 1.57 99 | 4.12 94 | 2.00 99 | 4.95 120 | 6.55 112 | 6.90 123 | 1.75 102 | 2.49 107 | 1.54 86 | 1.60 106 | 4.88 92 | 1.78 104 | 3.14 109 | 3.92 100 | 1.91 60 | 0.98 99 | 1.55 94 | 1.10 93 |
HBpMotionGpu [43] | 98.8 | 3.50 101 | 5.01 81 | 3.35 100 | 3.02 111 | 4.08 103 | 4.11 113 | 2.06 103 | 5.55 103 | 2.91 106 | 1.90 104 | 3.25 97 | 1.85 103 | 1.71 101 | 2.29 91 | 2.31 107 | 1.43 98 | 4.05 83 | 1.81 106 | 3.49 119 | 4.03 104 | 2.51 98 | 0.76 66 | 1.50 93 | 0.89 79 |
GroupFlow [9] | 99.1 | 4.01 109 | 8.96 112 | 5.33 114 | 4.08 117 | 10.0 121 | 5.03 117 | 2.71 111 | 11.2 114 | 3.56 114 | 1.47 100 | 4.31 103 | 1.41 98 | 2.46 115 | 3.47 118 | 1.68 90 | 2.65 122 | 8.76 116 | 3.71 122 | 1.30 4 | 2.56 32 | 0.92 6 | 1.03 106 | 1.90 110 | 1.34 107 |
2D-CLG [1] | 100.8 | 4.35 113 | 11.2 117 | 3.92 104 | 4.00 116 | 5.65 113 | 6.06 121 | 4.79 121 | 14.1 119 | 5.16 119 | 6.50 124 | 14.0 123 | 6.55 122 | 1.76 103 | 2.41 100 | 2.94 116 | 1.21 86 | 6.32 107 | 1.62 97 | 1.40 13 | 2.55 30 | 0.90 4 | 1.27 117 | 2.20 118 | 1.69 115 |
SPSA-learn [13] | 101.6 | 3.57 103 | 9.65 115 | 4.33 110 | 2.13 104 | 4.20 104 | 2.79 105 | 2.06 103 | 6.85 106 | 2.87 105 | 1.88 103 | 5.24 108 | 1.95 104 | 1.82 105 | 2.38 96 | 2.35 110 | 1.54 103 | 6.21 106 | 1.94 110 | 2.02 64 | 3.22 74 | 1.47 42 | 1.41 119 | 2.22 119 | 2.28 120 |
IAOF [50] | 102.0 | 3.55 102 | 6.41 102 | 4.27 108 | 2.52 108 | 3.73 99 | 3.48 110 | 2.09 105 | 7.46 108 | 2.47 102 | 2.56 111 | 5.54 109 | 3.29 116 | 1.62 95 | 2.36 95 | 1.44 81 | 1.46 101 | 5.99 104 | 1.44 91 | 2.79 103 | 3.42 83 | 2.11 75 | 1.12 115 | 1.94 112 | 1.49 113 |
Black & Anandan [4] | 102.3 | 3.90 108 | 8.79 111 | 5.34 115 | 2.10 103 | 4.91 109 | 2.68 103 | 2.24 107 | 7.98 111 | 2.91 106 | 1.98 105 | 6.06 111 | 2.01 105 | 1.97 109 | 2.68 108 | 2.11 102 | 1.38 94 | 6.99 112 | 1.59 94 | 2.55 95 | 3.97 102 | 1.10 16 | 1.11 114 | 2.04 115 | 1.28 101 |
Heeger++ [104] | 103.7 | 7.15 123 | 17.6 126 | 4.62 112 | 4.20 120 | 14.6 128 | 3.84 111 | 7.59 128 | 16.7 123 | 6.25 125 | 4.14 119 | 9.75 118 | 4.00 117 | 3.45 124 | 3.91 125 | 3.63 122 | 5.33 127 | 17.9 127 | 6.70 127 | 2.06 68 | 4.70 117 | 1.40 33 | 0.40 6 | 1.08 56 | 0.39 7 |
GraphCuts [14] | 104.3 | 3.73 106 | 6.14 100 | 4.13 107 | 1.95 102 | 5.36 111 | 2.22 97 | 1.84 101 | 5.39 102 | 3.06 108 | 1.23 94 | 4.38 104 | 0.99 89 | 1.67 98 | 2.32 93 | 1.95 98 | 2.18 116 | 4.06 84 | 1.96 111 | 3.32 116 | 4.15 109 | 3.73 117 | 1.67 122 | 1.68 100 | 2.14 119 |
FlowNet2 [122] | 105.4 | 4.70 116 | 7.33 107 | 4.31 109 | 4.14 119 | 7.37 114 | 5.19 118 | 3.19 115 | 7.71 110 | 3.60 115 | 2.14 106 | 2.64 86 | 2.05 106 | 2.09 112 | 2.89 113 | 1.98 99 | 1.31 92 | 4.96 94 | 1.71 99 | 3.19 111 | 5.29 125 | 2.55 102 | 0.92 94 | 1.46 87 | 0.92 80 |
FFV1MT [106] | 107.0 | 6.53 122 | 14.4 122 | 5.37 116 | 3.40 114 | 12.8 127 | 3.44 109 | 7.00 125 | 18.3 126 | 6.41 126 | 4.10 118 | 15.6 124 | 4.04 118 | 3.98 127 | 4.51 127 | 5.29 128 | 5.96 128 | 18.3 128 | 7.50 128 | 2.59 96 | 4.32 111 | 2.20 79 | 0.40 6 | 1.08 56 | 0.39 7 |
Nguyen [33] | 107.4 | 4.50 115 | 8.30 109 | 4.64 113 | 6.04 123 | 4.82 108 | 11.0 125 | 3.37 116 | 12.8 116 | 4.28 117 | 6.23 123 | 9.23 117 | 7.98 124 | 2.07 111 | 2.73 110 | 3.19 117 | 1.38 94 | 6.12 105 | 1.76 103 | 2.12 73 | 3.24 76 | 1.34 27 | 1.34 118 | 2.23 120 | 1.90 117 |
UnFlow [129] | 107.8 | 8.73 125 | 14.6 123 | 5.86 117 | 5.18 122 | 8.43 118 | 5.98 120 | 6.12 124 | 18.7 127 | 5.81 122 | 2.89 116 | 8.11 115 | 2.63 112 | 3.59 125 | 3.82 124 | 4.60 127 | 2.49 120 | 9.81 118 | 3.55 120 | 2.73 102 | 3.95 101 | 1.14 18 | 0.72 56 | 1.34 77 | 0.75 58 |
2bit-BM-tele [98] | 108.0 | 5.50 120 | 7.24 106 | 7.00 120 | 2.15 105 | 3.03 90 | 2.69 104 | 2.27 108 | 4.45 99 | 2.35 101 | 2.63 113 | 2.90 93 | 2.77 114 | 1.63 96 | 2.17 86 | 2.25 104 | 2.29 118 | 5.52 101 | 2.82 116 | 4.33 123 | 5.11 124 | 5.62 125 | 1.01 104 | 1.79 106 | 1.72 116 |
SILK [79] | 108.4 | 4.92 119 | 10.7 116 | 7.59 123 | 3.66 115 | 8.19 116 | 4.66 115 | 3.10 114 | 12.4 115 | 3.75 116 | 2.78 115 | 7.03 113 | 2.83 115 | 2.79 119 | 3.42 117 | 3.35 118 | 2.18 116 | 9.05 117 | 2.40 115 | 1.66 34 | 2.78 52 | 1.85 58 | 1.44 120 | 2.57 122 | 2.47 122 |
Horn & Schunck [3] | 110.7 | 4.32 112 | 13.5 120 | 5.90 118 | 2.42 107 | 7.53 115 | 2.88 106 | 2.91 113 | 13.4 118 | 3.32 109 | 2.58 112 | 9.94 119 | 2.71 113 | 2.62 116 | 3.37 116 | 2.79 113 | 1.64 108 | 10.2 121 | 1.98 112 | 2.82 105 | 4.37 113 | 1.28 23 | 1.68 123 | 3.07 124 | 2.30 121 |
Periodicity [78] | 111.4 | 4.83 118 | 9.05 113 | 3.96 105 | 2.71 110 | 9.89 119 | 3.14 108 | 6.02 123 | 13.1 117 | 6.02 124 | 2.38 109 | 8.20 116 | 2.32 107 | 5.70 128 | 9.33 129 | 4.34 126 | 4.45 125 | 24.5 129 | 4.11 124 | 1.87 50 | 4.04 105 | 1.02 11 | 1.75 124 | 4.56 128 | 2.87 125 |
Adaptive flow [45] | 114.6 | 9.90 127 | 13.2 118 | 12.5 126 | 6.60 124 | 8.42 117 | 8.27 123 | 4.70 120 | 14.5 121 | 5.85 123 | 5.26 122 | 12.1 121 | 5.52 121 | 3.21 123 | 3.56 121 | 3.59 120 | 4.42 124 | 11.0 122 | 4.71 125 | 9.00 128 | 7.86 128 | 15.7 127 | 0.46 10 | 1.74 104 | 0.74 56 |
SLK [47] | 115.3 | 4.20 110 | 17.7 127 | 6.35 119 | 7.25 125 | 11.8 125 | 11.0 125 | 4.27 119 | 18.2 125 | 5.29 120 | 10.8 127 | 13.4 122 | 16.5 129 | 3.11 122 | 3.74 122 | 4.22 125 | 2.43 119 | 11.4 125 | 3.32 119 | 1.83 48 | 3.60 92 | 1.53 45 | 2.53 127 | 3.59 125 | 4.73 126 |
TI-DOFE [24] | 115.6 | 8.79 126 | 16.1 124 | 13.8 128 | 8.00 126 | 10.2 122 | 11.4 127 | 7.45 126 | 19.0 128 | 7.28 127 | 9.61 126 | 16.2 125 | 11.2 126 | 2.78 118 | 3.54 120 | 3.59 120 | 1.94 114 | 10.1 120 | 2.84 117 | 2.10 72 | 3.58 91 | 0.98 9 | 2.55 128 | 3.92 127 | 4.92 127 |
HCIC-L [99] | 118.7 | 11.0 128 | 13.3 119 | 7.06 121 | 13.2 129 | 12.4 126 | 16.8 129 | 7.58 127 | 10.5 113 | 10.5 129 | 13.4 128 | 17.8 126 | 13.8 128 | 3.73 126 | 4.09 126 | 3.40 119 | 4.45 125 | 7.90 115 | 5.38 126 | 17.2 129 | 13.0 129 | 17.3 129 | 0.78 77 | 1.10 61 | 0.96 83 |
FOLKI [16] | 119.7 | 4.72 117 | 16.7 125 | 8.10 124 | 4.62 121 | 11.0 124 | 8.17 122 | 3.43 117 | 16.8 124 | 3.51 113 | 3.53 117 | 10.3 120 | 4.31 119 | 2.89 120 | 3.80 123 | 3.81 123 | 2.60 121 | 13.1 126 | 4.02 123 | 2.33 86 | 4.53 115 | 3.23 112 | 2.39 126 | 3.89 126 | 7.04 128 |
PGAM+LK [55] | 122.4 | 6.47 121 | 18.3 128 | 8.42 125 | 4.08 117 | 10.7 123 | 5.38 119 | 3.66 118 | 14.4 120 | 4.35 118 | 5.05 121 | 29.1 129 | 5.23 120 | 2.92 121 | 3.48 119 | 3.99 124 | 3.48 123 | 11.1 124 | 3.57 121 | 5.90 127 | 5.96 127 | 5.68 126 | 1.57 121 | 2.70 123 | 2.60 123 |
Pyramid LK [2] | 125.8 | 8.02 124 | 14.1 121 | 14.8 129 | 8.54 127 | 9.96 120 | 16.3 128 | 5.59 122 | 14.6 122 | 8.07 128 | 7.36 125 | 22.2 128 | 9.72 125 | 5.85 129 | 7.94 128 | 7.95 129 | 7.37 129 | 11.0 122 | 8.48 129 | 4.42 125 | 4.94 123 | 3.92 119 | 5.06 129 | 8.31 129 | 17.6 129 |
AdaConv-v1 [126] | 130.0 | 40.6 130 | 40.4 130 | 41.9 130 | 76.0 130 | 77.9 130 | 75.2 130 | 71.5 130 | 69.6 130 | 73.8 130 | 60.3 130 | 73.5 130 | 60.3 130 | 80.9 130 | 81.7 130 | 81.8 130 | 78.7 130 | 70.9 130 | 77.5 130 | 58.2 130 | 47.8 130 | 72.2 130 | 82.4 130 | 83.1 130 | 82.9 130 |
SepConv-v1 [127] | 130.0 | 40.6 130 | 40.4 130 | 41.9 130 | 76.0 130 | 77.9 130 | 75.2 130 | 71.5 130 | 69.6 130 | 73.8 130 | 60.3 130 | 73.5 130 | 60.3 130 | 80.9 130 | 81.7 130 | 81.8 130 | 78.7 130 | 70.9 130 | 77.5 130 | 58.2 130 | 47.8 130 | 72.2 130 | 82.4 130 | 83.1 130 | 82.9 130 |
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. |