Optical flow evaluation results |
Statistics:
Average
SD
R2.5
R5.0
R10.0
A90
A95
A99
Error type: endpoint angle interpolation normalized interpolation |
R10.0 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] | 8.9 | 1.35 11 | 4.72 12 | 0.01 4 | 2.31 18 | 6.03 11 | 0.07 5 | 1.11 1 | 3.99 5 | 0.12 4 | 3.07 2 | 6.79 3 | 0.40 13 | 13.6 6 | 22.1 3 | 1.00 15 | 4.53 5 | 18.6 5 | 0.11 7 | 4.33 8 | 24.1 8 | 0.07 12 | 8.94 25 | 22.6 23 | 0.19 7 |
MDP-Flow2 [68] | 12.4 | 1.29 4 | 4.46 7 | 0.01 4 | 2.18 10 | 6.04 12 | 0.06 2 | 1.11 1 | 4.13 9 | 0.14 10 | 3.08 3 | 6.86 5 | 0.36 3 | 13.5 2 | 22.1 3 | 1.01 25 | 4.96 32 | 20.4 32 | 0.18 35 | 4.31 6 | 24.0 7 | 0.07 12 | 8.94 25 | 22.6 23 | 0.20 25 |
NN-field [71] | 22.8 | 1.45 30 | 5.41 45 | 0.01 4 | 1.87 4 | 5.01 4 | 0.06 2 | 1.51 67 | 3.94 3 | 0.16 21 | 3.78 57 | 8.92 78 | 0.41 22 | 13.6 6 | 22.2 8 | 1.00 15 | 5.05 48 | 20.7 44 | 0.20 43 | 4.27 3 | 23.7 3 | 0.07 12 | 8.85 11 | 22.4 9 | 0.19 7 |
NNF-Local [87] | 23.0 | 1.41 18 | 5.16 29 | 0.01 4 | 1.85 2 | 5.00 3 | 0.07 5 | 1.12 3 | 4.05 6 | 0.14 10 | 3.68 44 | 8.61 67 | 0.42 27 | 13.6 6 | 22.3 9 | 1.00 15 | 5.21 67 | 21.4 67 | 0.24 59 | 4.35 9 | 24.1 8 | 0.11 68 | 8.85 11 | 22.4 9 | 0.19 7 |
DeepFlow [86] | 25.2 | 1.35 11 | 4.67 11 | 0.00 1 | 3.01 59 | 8.09 56 | 0.21 57 | 1.26 14 | 5.00 38 | 0.12 4 | 3.97 80 | 8.00 42 | 0.44 42 | 13.7 12 | 22.4 15 | 1.03 47 | 4.46 3 | 18.3 3 | 0.08 2 | 4.40 15 | 24.5 19 | 0.06 3 | 8.74 7 | 22.1 7 | 0.21 58 |
SepConv-v1 [127] | 25.6 | 0.93 1 | 3.75 1 | 0.02 63 | 2.74 42 | 6.52 23 | 0.46 109 | 1.13 4 | 2.38 1 | 0.42 102 | 3.64 43 | 7.10 10 | 0.94 116 | 13.6 6 | 22.1 3 | 0.82 2 | 4.01 1 | 16.5 1 | 0.05 1 | 4.02 1 | 22.3 1 | 0.12 81 | 8.05 1 | 20.3 1 | 0.12 1 |
SuperFlow [81] | 26.3 | 1.30 5 | 4.40 5 | 0.01 4 | 3.28 70 | 8.24 63 | 0.30 76 | 1.46 59 | 4.37 12 | 0.23 63 | 3.62 37 | 7.44 19 | 0.46 52 | 13.7 12 | 22.4 15 | 0.99 10 | 4.56 6 | 18.7 6 | 0.08 2 | 4.44 23 | 24.7 26 | 0.09 45 | 8.83 9 | 22.4 9 | 0.17 3 |
DF-Auto [115] | 27.0 | 1.26 2 | 3.92 2 | 0.00 1 | 3.13 64 | 7.98 52 | 0.29 71 | 1.14 5 | 4.08 8 | 0.16 21 | 3.57 31 | 7.48 22 | 0.46 52 | 13.5 2 | 22.0 2 | 1.01 25 | 4.66 7 | 19.2 8 | 0.15 19 | 4.49 33 | 24.8 30 | 0.09 45 | 9.09 43 | 23.0 45 | 0.21 58 |
DeepFlow2 [108] | 27.2 | 1.40 17 | 4.87 16 | 0.01 4 | 3.01 59 | 8.13 58 | 0.20 55 | 1.25 13 | 5.01 41 | 0.11 2 | 3.83 65 | 8.21 52 | 0.43 35 | 13.7 12 | 22.4 15 | 1.03 47 | 4.49 4 | 18.5 4 | 0.08 2 | 4.49 33 | 24.8 30 | 0.06 3 | 8.87 14 | 22.5 15 | 0.21 58 |
NNF-EAC [103] | 28.8 | 1.48 35 | 4.99 21 | 0.03 114 | 2.49 31 | 6.76 31 | 0.08 11 | 1.56 73 | 4.39 13 | 0.24 69 | 3.28 8 | 7.08 8 | 0.39 8 | 13.7 12 | 22.3 9 | 1.01 25 | 4.66 7 | 19.1 7 | 0.12 11 | 4.41 18 | 24.5 19 | 0.11 68 | 9.02 34 | 22.8 35 | 0.20 25 |
IROF++ [58] | 29.0 | 1.54 59 | 5.77 70 | 0.01 4 | 2.35 21 | 6.39 16 | 0.10 22 | 1.50 64 | 5.04 42 | 0.25 71 | 3.10 4 | 6.75 2 | 0.39 8 | 13.7 12 | 22.4 15 | 1.16 83 | 4.66 7 | 19.2 8 | 0.11 7 | 4.55 51 | 25.1 46 | 0.10 57 | 8.84 10 | 22.4 9 | 0.19 7 |
PH-Flow [101] | 30.5 | 1.54 59 | 5.76 69 | 0.01 4 | 1.93 5 | 5.29 5 | 0.08 11 | 1.17 6 | 4.41 14 | 0.17 27 | 3.05 1 | 6.71 1 | 0.39 8 | 13.6 6 | 22.3 9 | 1.01 25 | 5.35 83 | 21.9 83 | 0.34 99 | 4.36 12 | 24.3 12 | 0.10 57 | 8.92 23 | 22.6 23 | 0.22 90 |
Local-TV-L1 [65] | 31.4 | 1.34 9 | 4.39 4 | 0.02 63 | 4.19 97 | 9.42 85 | 0.37 92 | 1.34 28 | 4.34 11 | 0.15 16 | 3.61 36 | 7.78 36 | 0.44 42 | 13.8 26 | 22.5 24 | 1.04 50 | 4.71 12 | 19.5 14 | 0.16 27 | 4.40 15 | 24.5 19 | 0.07 12 | 8.70 6 | 22.0 4 | 0.20 25 |
nLayers [57] | 32.9 | 1.51 45 | 5.46 48 | 0.01 4 | 2.17 9 | 5.91 9 | 0.08 11 | 1.24 12 | 3.88 2 | 0.18 34 | 3.47 27 | 7.71 34 | 0.40 13 | 13.9 44 | 22.7 55 | 1.18 91 | 5.25 69 | 21.6 73 | 0.31 84 | 4.35 9 | 23.8 4 | 0.09 45 | 8.99 30 | 22.7 30 | 0.19 7 |
Brox et al. [5] | 33.0 | 1.43 25 | 4.82 14 | 0.01 4 | 2.95 50 | 7.80 47 | 0.18 46 | 1.40 45 | 5.17 54 | 0.16 21 | 3.73 49 | 7.70 33 | 0.45 46 | 13.7 12 | 22.4 15 | 1.00 15 | 5.04 45 | 20.6 41 | 0.27 70 | 4.52 38 | 25.1 46 | 0.09 45 | 8.87 14 | 22.4 9 | 0.19 7 |
Layers++ [37] | 33.1 | 1.44 28 | 5.17 30 | 0.01 4 | 1.83 1 | 4.87 1 | 0.05 1 | 1.32 26 | 4.86 28 | 0.22 59 | 3.34 15 | 7.36 15 | 0.43 35 | 13.8 26 | 22.6 40 | 1.13 76 | 5.37 87 | 22.0 85 | 0.24 59 | 4.39 14 | 24.3 12 | 0.05 1 | 9.00 31 | 22.7 30 | 0.22 90 |
ALD-Flow [66] | 34.8 | 1.53 56 | 5.62 58 | 0.01 4 | 2.86 47 | 7.94 49 | 0.16 35 | 1.29 17 | 5.15 51 | 0.12 4 | 3.34 15 | 7.68 31 | 0.38 6 | 14.0 75 | 22.8 73 | 1.15 81 | 4.71 12 | 19.2 8 | 0.11 7 | 4.41 18 | 24.4 16 | 0.06 3 | 9.30 72 | 23.5 73 | 0.20 25 |
CBF [12] | 35.6 | 1.28 3 | 4.40 5 | 0.01 4 | 3.24 69 | 8.20 61 | 0.25 65 | 1.59 79 | 4.63 17 | 0.18 34 | 3.60 35 | 7.61 28 | 0.49 63 | 13.8 26 | 22.5 24 | 0.99 10 | 4.89 23 | 20.2 26 | 0.18 35 | 4.56 53 | 25.3 60 | 0.07 12 | 9.21 58 | 23.3 60 | 0.18 4 |
LME [70] | 35.8 | 1.39 15 | 5.10 25 | 0.01 4 | 2.49 31 | 6.95 36 | 0.10 22 | 1.35 32 | 5.64 86 | 0.15 16 | 3.29 10 | 7.55 25 | 0.39 8 | 14.1 97 | 23.0 100 | 1.26 128 | 5.13 57 | 21.1 59 | 0.19 40 | 4.37 13 | 24.2 11 | 0.06 3 | 8.89 20 | 22.5 15 | 0.19 7 |
WLIF-Flow [93] | 35.9 | 1.44 28 | 5.19 32 | 0.01 4 | 2.52 34 | 6.81 33 | 0.16 35 | 1.38 38 | 4.67 18 | 0.20 47 | 3.21 5 | 6.98 6 | 0.42 27 | 13.7 12 | 22.4 15 | 1.07 61 | 5.37 87 | 22.1 90 | 0.28 75 | 4.43 21 | 24.4 16 | 0.08 32 | 9.09 43 | 23.0 45 | 0.21 58 |
Aniso. Huber-L1 [22] | 36.0 | 1.43 25 | 5.00 22 | 0.01 4 | 4.12 93 | 9.46 88 | 0.36 87 | 1.60 81 | 4.77 20 | 0.17 27 | 3.69 45 | 7.99 40 | 0.43 35 | 13.7 12 | 22.3 9 | 1.00 15 | 4.90 25 | 20.1 22 | 0.12 11 | 4.62 64 | 25.2 51 | 0.07 12 | 8.95 28 | 22.6 23 | 0.20 25 |
ComponentFusion [96] | 38.0 | 1.54 59 | 6.05 87 | 0.01 4 | 2.33 20 | 6.57 25 | 0.07 5 | 1.31 25 | 4.82 24 | 0.16 21 | 3.35 19 | 7.66 30 | 0.37 5 | 13.9 44 | 22.6 40 | 1.13 76 | 4.93 27 | 20.3 29 | 0.20 43 | 4.61 62 | 25.7 77 | 0.13 87 | 9.06 39 | 22.9 40 | 0.20 25 |
CLG-TV [48] | 38.2 | 1.35 11 | 4.56 8 | 0.02 63 | 3.84 82 | 9.35 80 | 0.29 71 | 1.38 38 | 5.11 46 | 0.18 34 | 3.71 47 | 7.93 38 | 0.50 72 | 13.8 26 | 22.4 15 | 1.00 15 | 4.74 15 | 19.5 14 | 0.13 14 | 4.59 60 | 25.2 51 | 0.07 12 | 9.08 40 | 22.9 40 | 0.20 25 |
TV-L1-MCT [64] | 38.5 | 1.70 103 | 6.35 101 | 0.02 63 | 2.90 48 | 7.98 52 | 0.17 40 | 1.39 42 | 5.19 58 | 0.20 47 | 3.32 14 | 7.10 10 | 0.45 46 | 13.9 44 | 22.6 40 | 1.17 88 | 4.67 10 | 19.3 12 | 0.15 19 | 4.44 23 | 24.4 16 | 0.08 32 | 8.67 4 | 22.0 4 | 0.19 7 |
COFM [59] | 40.1 | 1.49 40 | 5.57 53 | 0.01 4 | 2.37 22 | 6.48 21 | 0.11 29 | 1.30 20 | 4.84 26 | 0.23 63 | 3.29 10 | 7.33 14 | 0.40 13 | 13.8 26 | 22.5 24 | 1.02 39 | 5.52 97 | 22.7 101 | 0.39 111 | 4.04 2 | 22.5 2 | 0.11 68 | 9.33 75 | 23.6 77 | 0.20 25 |
CombBMOF [113] | 40.8 | 1.59 77 | 5.46 48 | 0.02 63 | 2.30 17 | 6.40 17 | 0.10 22 | 1.39 42 | 4.83 25 | 0.21 55 | 3.90 74 | 8.60 65 | 0.46 52 | 13.8 26 | 22.5 24 | 1.01 25 | 4.92 26 | 20.1 22 | 0.11 7 | 5.35 117 | 25.9 83 | 0.10 57 | 8.89 20 | 22.4 9 | 0.19 7 |
Sparse-NonSparse [56] | 42.2 | 1.54 59 | 5.67 63 | 0.02 63 | 2.38 25 | 6.47 20 | 0.12 30 | 1.42 51 | 5.13 48 | 0.17 27 | 3.42 24 | 7.36 15 | 0.42 27 | 13.8 26 | 22.5 24 | 1.18 91 | 5.29 75 | 21.7 78 | 0.23 54 | 4.43 21 | 24.5 19 | 0.10 57 | 9.11 47 | 23.0 45 | 0.20 25 |
ProbFlowFields [128] | 42.4 | 1.46 32 | 5.63 59 | 0.02 63 | 2.23 14 | 6.35 15 | 0.09 17 | 1.19 7 | 4.55 15 | 0.20 47 | 3.51 29 | 8.02 44 | 0.45 46 | 13.9 44 | 22.7 55 | 1.24 122 | 5.28 74 | 21.6 73 | 0.39 111 | 4.32 7 | 24.1 8 | 0.11 68 | 8.68 5 | 22.0 4 | 0.21 58 |
FlowFields [110] | 42.5 | 1.52 52 | 5.99 81 | 0.02 63 | 2.31 18 | 6.55 24 | 0.09 17 | 1.30 20 | 4.97 35 | 0.20 47 | 3.76 53 | 9.03 82 | 0.40 13 | 13.9 44 | 22.7 55 | 1.16 83 | 5.16 59 | 21.3 64 | 0.28 75 | 4.40 15 | 24.5 19 | 0.07 12 | 8.88 17 | 22.5 15 | 0.21 58 |
SIOF [67] | 42.8 | 1.53 56 | 5.38 43 | 0.01 4 | 4.17 94 | 10.1 104 | 0.31 78 | 1.40 45 | 5.38 71 | 0.14 10 | 3.62 37 | 7.99 40 | 0.61 94 | 13.5 2 | 22.1 3 | 0.98 7 | 4.87 21 | 20.1 22 | 0.13 14 | 4.49 33 | 24.9 37 | 0.08 32 | 9.34 77 | 23.6 77 | 0.20 25 |
IROF-TV [53] | 42.9 | 1.51 45 | 5.64 60 | 0.02 63 | 2.60 36 | 6.84 34 | 0.16 35 | 1.34 28 | 5.38 71 | 0.18 34 | 3.29 10 | 7.49 23 | 0.42 27 | 14.0 75 | 22.8 73 | 1.21 102 | 5.05 48 | 20.8 48 | 0.20 43 | 4.52 38 | 25.2 51 | 0.07 12 | 8.82 8 | 22.3 8 | 0.21 58 |
TF+OM [100] | 42.9 | 1.41 18 | 5.19 32 | 0.01 4 | 2.38 25 | 6.62 26 | 0.12 30 | 1.42 51 | 5.51 80 | 0.15 16 | 3.87 69 | 8.66 72 | 0.49 63 | 13.9 44 | 22.6 40 | 1.05 55 | 4.96 32 | 20.3 29 | 0.16 27 | 4.54 48 | 25.3 60 | 0.10 57 | 9.12 49 | 23.0 45 | 0.21 58 |
HAST [109] | 43.6 | 1.48 35 | 5.42 46 | 0.01 4 | 2.14 8 | 5.79 8 | 0.06 2 | 1.52 69 | 5.24 60 | 0.23 63 | 3.27 7 | 7.14 12 | 0.33 2 | 14.0 75 | 22.8 73 | 0.99 10 | 5.52 97 | 22.7 101 | 0.31 84 | 4.35 9 | 24.3 12 | 0.07 12 | 9.68 100 | 24.4 100 | 0.21 58 |
PGM-C [120] | 43.8 | 1.50 42 | 5.75 67 | 0.01 4 | 2.37 22 | 6.68 28 | 0.10 22 | 1.48 62 | 5.32 64 | 0.15 16 | 3.77 55 | 9.13 88 | 0.50 72 | 13.9 44 | 22.6 40 | 1.21 102 | 4.97 37 | 20.4 32 | 0.23 54 | 4.54 48 | 25.1 46 | 0.08 32 | 8.94 25 | 22.6 23 | 0.20 25 |
BlockOverlap [61] | 44.2 | 1.34 9 | 4.33 3 | 0.02 63 | 4.04 90 | 9.04 77 | 0.48 111 | 1.36 34 | 4.31 10 | 0.32 91 | 3.43 26 | 6.81 4 | 0.71 102 | 14.0 75 | 22.8 73 | 1.04 50 | 4.89 23 | 20.0 21 | 0.22 51 | 4.44 23 | 24.7 26 | 0.11 68 | 8.64 3 | 21.8 3 | 0.20 25 |
CPM-Flow [116] | 44.3 | 1.50 42 | 5.73 66 | 0.01 4 | 2.37 22 | 6.69 29 | 0.10 22 | 1.38 38 | 5.06 44 | 0.12 4 | 4.02 83 | 9.68 99 | 0.51 77 | 13.9 44 | 22.6 40 | 1.21 102 | 4.83 19 | 19.9 18 | 0.15 19 | 4.63 66 | 25.6 71 | 0.08 32 | 8.88 17 | 22.5 15 | 0.22 90 |
FMOF [94] | 44.4 | 1.67 100 | 5.98 79 | 0.03 114 | 2.22 12 | 6.04 12 | 0.08 11 | 1.56 73 | 5.22 59 | 0.28 84 | 3.81 63 | 8.34 56 | 0.49 63 | 13.8 26 | 22.5 24 | 1.01 25 | 5.01 42 | 20.5 37 | 0.15 19 | 4.30 5 | 23.9 6 | 0.07 12 | 9.21 58 | 23.3 60 | 0.20 25 |
Ramp [62] | 45.5 | 1.57 74 | 5.75 67 | 0.01 4 | 2.38 25 | 6.51 22 | 0.19 48 | 1.41 48 | 5.11 46 | 0.17 27 | 3.26 6 | 7.09 9 | 0.41 22 | 13.9 44 | 22.6 40 | 1.15 81 | 5.51 96 | 22.5 96 | 0.32 93 | 4.48 29 | 24.7 26 | 0.07 12 | 9.33 75 | 23.6 77 | 0.20 25 |
F-TV-L1 [15] | 46.4 | 1.54 59 | 5.24 37 | 0.02 63 | 4.11 92 | 9.73 98 | 0.32 81 | 1.50 64 | 5.44 75 | 0.23 63 | 3.76 53 | 8.03 45 | 0.47 58 | 13.5 2 | 22.1 3 | 0.94 3 | 4.71 12 | 19.4 13 | 0.17 30 | 4.61 62 | 25.3 60 | 0.13 87 | 8.92 23 | 22.6 23 | 0.19 7 |
2DHMM-SAS [92] | 46.5 | 1.65 95 | 6.25 98 | 0.02 63 | 3.41 72 | 8.58 69 | 0.22 59 | 1.36 34 | 4.84 26 | 0.19 42 | 3.28 8 | 7.02 7 | 0.43 35 | 13.8 26 | 22.5 24 | 1.19 95 | 4.94 29 | 20.1 22 | 0.09 5 | 4.52 38 | 24.8 30 | 0.12 81 | 9.25 67 | 23.4 67 | 0.20 25 |
FlowFields+ [130] | 46.8 | 1.52 52 | 5.97 78 | 0.02 63 | 2.26 15 | 6.42 19 | 0.09 17 | 1.29 17 | 5.05 43 | 0.19 42 | 3.69 45 | 8.96 79 | 0.44 42 | 14.0 75 | 22.8 73 | 1.21 102 | 5.26 70 | 21.6 73 | 0.33 96 | 4.41 18 | 24.5 19 | 0.08 32 | 8.86 13 | 22.5 15 | 0.20 25 |
Second-order prior [8] | 47.5 | 1.39 15 | 4.88 17 | 0.02 63 | 3.85 84 | 9.45 87 | 0.27 68 | 1.83 91 | 5.99 97 | 0.26 75 | 3.83 65 | 8.56 63 | 0.49 63 | 13.6 6 | 22.3 9 | 1.01 25 | 4.82 18 | 19.9 18 | 0.18 35 | 4.67 72 | 25.6 71 | 0.06 3 | 9.03 36 | 22.8 35 | 0.20 25 |
MDP-Flow [26] | 47.8 | 1.35 11 | 4.97 20 | 0.02 63 | 2.22 12 | 6.25 14 | 0.09 17 | 1.20 8 | 4.07 7 | 0.14 10 | 3.89 72 | 8.34 56 | 0.48 60 | 13.8 26 | 22.5 24 | 1.24 122 | 5.74 111 | 23.6 115 | 0.50 128 | 4.63 66 | 25.6 71 | 0.11 68 | 8.97 29 | 22.7 30 | 0.19 7 |
Classic++ [32] | 47.9 | 1.46 32 | 5.27 38 | 0.01 4 | 3.38 71 | 8.81 74 | 0.25 65 | 1.47 61 | 5.14 50 | 0.17 27 | 3.93 79 | 8.15 49 | 0.46 52 | 13.8 26 | 22.6 40 | 1.00 15 | 5.16 59 | 21.2 63 | 0.25 65 | 4.59 60 | 25.2 51 | 0.08 32 | 9.17 55 | 23.2 56 | 0.20 25 |
Classic+NL [31] | 48.1 | 1.62 85 | 5.98 79 | 0.02 63 | 2.48 28 | 6.65 27 | 0.17 40 | 1.41 48 | 5.13 48 | 0.19 42 | 3.37 21 | 7.27 13 | 0.44 42 | 13.9 44 | 22.6 40 | 1.12 75 | 5.34 81 | 21.8 80 | 0.22 51 | 4.48 29 | 24.8 30 | 0.09 45 | 9.27 70 | 23.4 67 | 0.19 7 |
LDOF [28] | 49.2 | 1.45 30 | 4.81 13 | 0.02 63 | 3.10 62 | 7.33 40 | 0.56 123 | 1.58 77 | 5.39 74 | 0.22 59 | 3.92 78 | 8.47 60 | 0.63 96 | 13.8 26 | 22.5 24 | 1.02 39 | 4.68 11 | 19.2 8 | 0.13 14 | 4.48 29 | 25.0 42 | 0.10 57 | 9.01 32 | 22.8 35 | 0.22 90 |
OAR-Flow [125] | 50.0 | 1.56 70 | 5.61 57 | 0.01 4 | 2.99 58 | 8.09 56 | 0.22 59 | 1.29 17 | 4.89 30 | 0.10 1 | 3.29 10 | 7.60 27 | 0.38 6 | 14.0 75 | 22.8 73 | 1.23 116 | 5.11 55 | 21.0 52 | 0.29 78 | 4.78 87 | 26.1 86 | 0.09 45 | 9.19 56 | 23.2 56 | 0.20 25 |
AggregFlow [97] | 50.5 | 1.89 115 | 7.48 118 | 0.01 4 | 2.95 50 | 8.13 58 | 0.16 35 | 1.22 10 | 4.86 28 | 0.14 10 | 4.18 88 | 9.69 100 | 0.45 46 | 13.9 44 | 22.6 40 | 1.04 50 | 4.94 29 | 20.2 26 | 0.15 19 | 4.51 36 | 25.0 42 | 0.12 81 | 9.24 66 | 23.3 60 | 0.21 58 |
ComplOF-FED-GPU [35] | 51.0 | 1.56 70 | 5.92 75 | 0.02 63 | 2.71 40 | 7.63 43 | 0.17 40 | 1.95 96 | 5.09 45 | 0.39 101 | 3.63 39 | 8.64 69 | 0.40 13 | 13.7 12 | 22.5 24 | 1.13 76 | 4.98 38 | 20.5 37 | 0.17 30 | 4.70 76 | 25.7 77 | 0.07 12 | 9.31 73 | 23.4 67 | 0.19 7 |
CRTflow [80] | 51.0 | 1.50 42 | 5.48 50 | 0.02 63 | 3.87 85 | 9.40 84 | 0.36 87 | 1.62 83 | 6.22 101 | 0.24 69 | 3.55 30 | 7.76 35 | 0.43 35 | 13.9 44 | 22.7 55 | 1.21 102 | 4.77 17 | 19.6 16 | 0.14 17 | 4.53 43 | 25.2 51 | 0.07 12 | 9.04 37 | 22.9 40 | 0.20 25 |
LSM [39] | 51.1 | 1.64 91 | 6.31 100 | 0.01 4 | 2.48 28 | 6.79 32 | 0.12 30 | 1.51 67 | 5.55 82 | 0.17 27 | 3.59 34 | 7.98 39 | 0.41 22 | 13.9 44 | 22.6 40 | 1.19 95 | 5.36 85 | 21.9 83 | 0.25 65 | 4.47 27 | 24.7 26 | 0.10 57 | 9.23 63 | 23.3 60 | 0.20 25 |
Kuang [131] | 52.5 | 1.52 52 | 5.99 81 | 0.01 4 | 2.57 35 | 7.16 39 | 0.12 30 | 1.55 72 | 5.50 78 | 0.21 55 | 3.84 67 | 9.18 91 | 0.41 22 | 13.9 44 | 22.7 55 | 1.17 88 | 5.04 45 | 20.7 44 | 0.21 46 | 4.79 88 | 26.4 96 | 0.14 91 | 8.87 14 | 22.5 15 | 0.19 7 |
DPOF [18] | 53.3 | 1.64 91 | 6.52 109 | 0.04 124 | 1.98 6 | 5.37 6 | 0.07 5 | 1.83 91 | 4.75 19 | 0.34 94 | 3.75 52 | 8.80 77 | 0.48 60 | 13.7 12 | 22.3 9 | 1.01 25 | 5.16 59 | 21.0 52 | 0.14 17 | 4.67 72 | 25.3 60 | 0.06 3 | 9.31 73 | 23.5 73 | 0.22 90 |
p-harmonic [29] | 53.5 | 1.42 21 | 4.96 19 | 0.01 4 | 4.00 88 | 9.50 89 | 0.39 96 | 1.38 38 | 5.68 88 | 0.19 42 | 4.20 90 | 8.58 64 | 0.49 63 | 13.9 44 | 22.6 40 | 1.01 25 | 4.93 27 | 20.3 29 | 0.21 46 | 4.81 90 | 26.1 86 | 0.11 68 | 9.12 49 | 23.1 52 | 0.20 25 |
TC-Flow [46] | 54.0 | 1.51 45 | 5.70 65 | 0.01 4 | 2.97 53 | 8.31 64 | 0.21 57 | 1.46 59 | 5.35 66 | 0.11 2 | 3.63 39 | 8.10 47 | 0.58 91 | 14.0 75 | 22.8 73 | 1.21 102 | 5.10 52 | 21.0 52 | 0.29 78 | 4.53 43 | 25.0 42 | 0.07 12 | 9.19 56 | 23.3 60 | 0.21 58 |
S2F-IF [123] | 54.3 | 1.55 67 | 6.13 91 | 0.01 4 | 2.26 15 | 6.41 18 | 0.09 17 | 1.30 20 | 5.17 54 | 0.17 27 | 3.72 48 | 9.01 81 | 0.45 46 | 14.0 75 | 22.9 92 | 1.24 122 | 5.27 71 | 21.6 73 | 0.32 93 | 4.54 48 | 25.2 51 | 0.09 45 | 8.88 17 | 22.5 15 | 0.23 114 |
OFLAF [77] | 55.1 | 1.48 35 | 5.49 52 | 0.01 4 | 2.00 7 | 5.50 7 | 0.07 5 | 1.30 20 | 4.97 35 | 0.15 16 | 3.34 15 | 7.40 18 | 0.40 13 | 14.0 75 | 22.8 73 | 1.21 102 | 5.56 102 | 22.8 105 | 0.41 114 | 4.83 92 | 26.3 93 | 0.17 106 | 9.73 103 | 24.5 105 | 0.20 25 |
SVFilterOh [111] | 55.5 | 1.51 45 | 5.48 50 | 0.02 63 | 2.19 11 | 5.94 10 | 0.10 22 | 1.50 64 | 5.00 38 | 0.25 71 | 3.78 57 | 8.01 43 | 0.40 13 | 14.3 111 | 23.2 111 | 1.22 114 | 5.35 83 | 22.0 85 | 0.23 54 | 4.27 3 | 23.8 4 | 0.06 3 | 9.55 92 | 24.1 94 | 0.22 90 |
FC-2Layers-FF [74] | 56.3 | 1.56 70 | 5.94 77 | 0.02 63 | 1.86 3 | 4.90 2 | 0.08 11 | 1.39 42 | 5.29 63 | 0.20 47 | 3.34 15 | 7.46 21 | 0.40 13 | 13.9 44 | 22.8 73 | 1.20 98 | 5.56 102 | 22.9 106 | 0.37 108 | 4.53 43 | 24.9 37 | 0.11 68 | 9.35 79 | 23.6 77 | 0.22 90 |
Occlusion-TV-L1 [63] | 57.0 | 1.43 25 | 5.20 35 | 0.01 4 | 4.18 96 | 10.3 107 | 0.37 92 | 1.34 28 | 5.35 66 | 0.27 81 | 4.19 89 | 9.14 89 | 0.56 87 | 13.7 12 | 22.4 15 | 0.97 4 | 4.99 40 | 20.6 41 | 0.33 96 | 5.12 107 | 25.4 66 | 0.27 118 | 9.01 32 | 22.7 30 | 0.19 7 |
S2D-Matching [84] | 58.5 | 1.65 95 | 6.07 89 | 0.02 63 | 3.21 66 | 8.58 69 | 0.23 63 | 1.34 28 | 5.00 38 | 0.23 63 | 3.35 19 | 7.36 15 | 0.42 27 | 13.9 44 | 22.7 55 | 1.07 61 | 5.53 100 | 22.6 98 | 0.35 102 | 4.57 56 | 24.8 30 | 0.07 12 | 9.22 62 | 23.3 60 | 0.22 90 |
RFlow [90] | 59.4 | 1.41 18 | 5.18 31 | 0.02 63 | 3.96 87 | 9.67 95 | 0.35 85 | 1.41 48 | 5.37 68 | 0.26 75 | 3.91 77 | 8.76 73 | 0.52 78 | 13.7 12 | 22.5 24 | 1.01 25 | 4.96 32 | 20.5 37 | 0.21 46 | 4.58 57 | 25.5 68 | 0.09 45 | 9.37 83 | 23.7 85 | 0.23 114 |
TC/T-Flow [76] | 59.8 | 1.69 102 | 6.24 96 | 0.02 63 | 2.98 55 | 8.17 60 | 0.19 48 | 1.27 15 | 4.78 21 | 0.13 8 | 3.58 33 | 8.26 53 | 0.36 3 | 14.1 97 | 23.0 100 | 1.23 116 | 5.03 44 | 20.6 41 | 0.12 11 | 4.85 94 | 26.3 93 | 0.16 103 | 9.40 85 | 23.8 88 | 0.19 7 |
MLDP_OF [89] | 60.8 | 1.55 67 | 6.03 85 | 0.02 63 | 3.12 63 | 8.40 67 | 0.20 55 | 1.22 10 | 4.97 35 | 0.13 8 | 3.77 55 | 7.62 29 | 0.73 104 | 13.9 44 | 22.7 55 | 1.01 25 | 5.60 106 | 23.0 110 | 0.31 84 | 4.63 66 | 25.3 60 | 0.14 91 | 9.21 58 | 23.3 60 | 0.21 58 |
HBM-GC [105] | 61.0 | 1.54 59 | 5.66 61 | 0.01 4 | 2.98 55 | 8.21 62 | 0.17 40 | 1.20 8 | 3.96 4 | 0.18 34 | 3.63 39 | 7.79 37 | 0.43 35 | 14.4 115 | 23.5 115 | 1.29 130 | 5.96 121 | 24.4 123 | 0.48 124 | 4.46 26 | 24.6 25 | 0.05 1 | 9.35 79 | 23.6 77 | 0.22 90 |
EpicFlow [102] | 61.1 | 1.51 45 | 5.82 73 | 0.01 4 | 2.90 48 | 8.08 54 | 0.18 46 | 1.43 54 | 5.25 61 | 0.16 21 | 3.88 71 | 9.42 96 | 0.54 83 | 13.9 44 | 22.7 55 | 1.21 102 | 5.13 57 | 21.1 59 | 0.31 84 | 4.71 78 | 25.8 80 | 0.14 91 | 9.13 51 | 23.1 52 | 0.21 58 |
AGIF+OF [85] | 61.2 | 1.66 99 | 6.04 86 | 0.02 63 | 2.48 28 | 6.73 30 | 0.19 48 | 1.44 55 | 4.96 34 | 0.26 75 | 3.41 23 | 7.54 24 | 0.45 46 | 14.1 97 | 23.1 108 | 1.21 102 | 5.47 94 | 22.3 92 | 0.27 70 | 4.52 38 | 24.3 12 | 0.07 12 | 9.41 87 | 23.8 88 | 0.21 58 |
OFH [38] | 62.4 | 1.56 70 | 5.79 71 | 0.01 4 | 3.55 76 | 8.78 73 | 0.30 76 | 1.62 83 | 6.44 105 | 0.16 21 | 3.57 31 | 8.52 62 | 0.39 8 | 13.8 26 | 22.6 40 | 1.16 83 | 5.18 63 | 21.3 64 | 0.31 84 | 4.94 98 | 26.6 97 | 0.15 98 | 9.36 81 | 23.6 77 | 0.19 7 |
Ad-TV-NDC [36] | 62.5 | 1.63 88 | 4.88 17 | 0.03 114 | 5.06 118 | 10.3 107 | 0.36 87 | 1.45 57 | 5.55 82 | 0.20 47 | 4.53 100 | 9.15 90 | 0.57 89 | 14.1 97 | 22.9 92 | 0.99 10 | 4.74 15 | 19.6 16 | 0.16 27 | 4.80 89 | 25.8 80 | 0.06 3 | 9.02 34 | 22.8 35 | 0.19 7 |
RNLOD-Flow [121] | 62.8 | 1.52 52 | 5.81 72 | 0.01 4 | 3.05 61 | 8.32 66 | 0.19 48 | 1.58 77 | 5.84 96 | 0.32 91 | 3.47 27 | 7.69 32 | 0.43 35 | 13.9 44 | 22.7 55 | 1.19 95 | 5.36 85 | 22.0 85 | 0.26 69 | 4.51 36 | 24.8 30 | 0.14 91 | 9.66 98 | 24.4 100 | 0.21 58 |
PMF [73] | 63.5 | 1.59 77 | 6.16 94 | 0.01 4 | 2.73 41 | 7.62 42 | 0.07 5 | 1.65 87 | 6.90 112 | 0.28 84 | 3.74 50 | 8.38 58 | 0.40 13 | 14.1 97 | 23.0 100 | 1.02 39 | 5.10 52 | 20.9 51 | 0.19 40 | 4.56 53 | 25.4 66 | 0.09 45 | 9.84 109 | 24.9 115 | 0.22 90 |
Sparse Occlusion [54] | 63.9 | 1.51 45 | 5.58 54 | 0.02 63 | 3.51 75 | 9.43 86 | 0.19 48 | 1.37 37 | 4.95 33 | 0.18 34 | 3.80 60 | 8.33 54 | 0.49 63 | 13.9 44 | 22.7 55 | 1.20 98 | 5.58 104 | 22.9 106 | 0.37 108 | 4.73 81 | 25.8 80 | 0.07 12 | 9.38 84 | 23.7 85 | 0.20 25 |
TCOF [69] | 64.5 | 1.54 59 | 5.59 55 | 0.01 4 | 4.46 105 | 10.4 110 | 0.43 105 | 1.28 16 | 5.15 51 | 0.14 10 | 3.63 39 | 8.04 46 | 0.42 27 | 13.9 44 | 22.7 55 | 0.98 7 | 5.41 89 | 22.3 92 | 0.24 59 | 5.00 102 | 26.7 98 | 0.09 45 | 9.76 104 | 24.6 107 | 0.24 119 |
Modified CLG [34] | 65.1 | 1.31 6 | 4.60 9 | 0.01 4 | 4.56 110 | 9.63 93 | 0.50 116 | 1.63 85 | 6.45 107 | 0.33 93 | 4.14 87 | 9.05 83 | 0.62 95 | 13.9 44 | 22.6 40 | 1.02 39 | 5.08 51 | 20.8 48 | 0.31 84 | 4.65 70 | 25.9 83 | 0.09 45 | 9.08 40 | 22.9 40 | 0.22 90 |
AdaConv-v1 [126] | 65.9 | 2.34 123 | 9.17 123 | 0.04 124 | 4.08 91 | 8.31 64 | 0.75 127 | 2.48 114 | 6.07 100 | 0.62 114 | 7.79 126 | 14.5 128 | 2.10 130 | 12.8 1 | 20.9 1 | 0.69 1 | 4.24 2 | 17.6 2 | 0.09 5 | 4.52 38 | 25.1 46 | 0.22 116 | 8.39 2 | 21.2 2 | 0.12 1 |
HBpMotionGpu [43] | 66.0 | 1.64 91 | 5.67 63 | 0.02 63 | 5.07 119 | 11.0 122 | 0.48 111 | 1.33 27 | 4.89 30 | 0.22 59 | 4.40 94 | 9.95 105 | 0.52 78 | 13.8 26 | 22.6 40 | 1.17 88 | 5.30 77 | 21.4 67 | 0.29 78 | 4.48 29 | 24.9 37 | 0.06 3 | 9.23 63 | 23.2 56 | 0.21 58 |
CostFilter [40] | 66.6 | 1.77 108 | 7.36 116 | 0.01 4 | 2.66 38 | 7.51 41 | 0.08 11 | 1.82 90 | 7.88 123 | 0.29 88 | 4.00 81 | 9.50 98 | 0.31 1 | 14.2 109 | 23.1 108 | 1.07 61 | 4.98 38 | 20.4 32 | 0.17 30 | 4.62 64 | 25.6 71 | 0.08 32 | 9.65 96 | 24.4 100 | 0.21 58 |
Adaptive [20] | 66.8 | 1.48 35 | 5.33 41 | 0.02 63 | 4.48 106 | 10.6 117 | 0.43 105 | 1.53 71 | 5.50 78 | 0.18 34 | 3.84 67 | 8.33 54 | 0.54 83 | 13.9 44 | 22.7 55 | 1.00 15 | 5.20 66 | 21.4 67 | 0.28 75 | 4.89 95 | 26.1 86 | 0.07 12 | 9.41 87 | 23.8 88 | 0.21 58 |
FlowNetS+ft+v [112] | 67.4 | 1.48 35 | 5.14 28 | 0.02 63 | 4.36 102 | 9.68 97 | 0.78 128 | 1.57 75 | 5.37 68 | 0.26 75 | 3.90 74 | 8.47 60 | 0.81 112 | 13.9 44 | 22.7 55 | 1.23 116 | 4.83 19 | 19.9 18 | 0.24 59 | 4.70 76 | 26.1 86 | 0.12 81 | 9.09 43 | 23.0 45 | 0.21 58 |
Bartels [41] | 68.5 | 1.59 77 | 6.24 96 | 0.03 114 | 3.20 65 | 8.92 76 | 0.31 78 | 1.40 45 | 5.17 54 | 0.25 71 | 4.09 85 | 9.05 83 | 0.86 114 | 14.1 97 | 22.9 92 | 0.97 4 | 5.43 92 | 22.2 91 | 0.25 65 | 4.47 27 | 24.8 30 | 0.10 57 | 9.15 53 | 23.1 52 | 0.20 25 |
TriFlow [95] | 70.0 | 1.65 95 | 6.58 110 | 0.01 4 | 3.78 81 | 9.53 91 | 0.29 71 | 1.45 57 | 5.99 97 | 0.18 34 | 4.07 84 | 9.25 95 | 0.49 63 | 14.0 75 | 22.9 92 | 1.18 91 | 5.34 81 | 21.4 67 | 0.15 19 | 4.64 69 | 25.2 51 | 0.08 32 | 9.40 85 | 23.6 77 | 0.21 58 |
Efficient-NL [60] | 70.5 | 1.53 56 | 5.60 56 | 0.01 4 | 2.98 55 | 7.94 49 | 0.16 35 | 2.05 98 | 5.45 77 | 0.56 110 | 3.82 64 | 8.12 48 | 0.42 27 | 13.8 26 | 22.5 24 | 1.18 91 | 5.69 109 | 23.2 112 | 0.32 93 | 4.75 84 | 26.1 86 | 0.11 68 | 9.94 117 | 24.8 112 | 0.22 90 |
Nguyen [33] | 70.8 | 1.55 67 | 5.02 23 | 0.00 1 | 5.69 124 | 10.8 119 | 0.48 111 | 1.63 85 | 6.70 110 | 0.28 84 | 5.35 112 | 10.5 111 | 0.77 109 | 13.8 26 | 22.5 24 | 1.01 25 | 4.99 40 | 20.7 44 | 0.18 35 | 5.44 119 | 28.2 113 | 0.21 112 | 9.08 40 | 22.9 40 | 0.20 25 |
EPPM w/o HM [88] | 71.2 | 1.68 101 | 7.02 112 | 0.02 63 | 2.84 45 | 8.08 54 | 0.10 22 | 2.13 106 | 7.82 122 | 0.36 96 | 3.87 69 | 9.12 86 | 0.49 63 | 13.9 44 | 22.7 55 | 1.04 50 | 5.27 71 | 21.6 73 | 0.17 30 | 4.56 53 | 25.2 51 | 0.15 98 | 9.34 77 | 23.6 77 | 0.22 90 |
2D-CLG [1] | 71.4 | 1.42 21 | 5.09 24 | 0.01 4 | 4.91 115 | 9.82 102 | 0.48 111 | 2.21 108 | 5.62 84 | 0.57 111 | 5.05 107 | 9.90 104 | 0.80 111 | 13.8 26 | 22.5 24 | 1.09 69 | 5.07 50 | 21.0 52 | 0.42 116 | 4.89 95 | 26.7 98 | 0.13 87 | 9.11 47 | 22.6 23 | 0.20 25 |
Filter Flow [19] | 72.0 | 1.54 59 | 5.28 39 | 0.01 4 | 4.52 109 | 9.97 103 | 0.35 85 | 1.61 82 | 5.53 81 | 0.20 47 | 4.55 101 | 8.61 67 | 0.46 52 | 14.3 111 | 23.2 111 | 1.08 66 | 5.10 52 | 21.0 52 | 0.21 46 | 4.76 86 | 26.1 86 | 0.11 68 | 9.65 96 | 24.3 99 | 0.20 25 |
IAOF [50] | 72.1 | 1.79 109 | 5.85 74 | 0.02 63 | 6.44 129 | 12.4 130 | 0.55 121 | 1.84 94 | 5.77 93 | 0.36 96 | 4.78 104 | 9.12 86 | 0.75 108 | 13.7 12 | 22.4 15 | 1.01 25 | 5.01 42 | 20.7 44 | 0.15 19 | 4.73 81 | 25.7 77 | 0.09 45 | 9.29 71 | 23.4 67 | 0.20 25 |
Steered-L1 [118] | 72.5 | 1.33 8 | 4.83 15 | 0.02 63 | 2.85 46 | 7.86 48 | 0.32 81 | 2.08 100 | 5.18 57 | 0.64 116 | 4.36 93 | 8.60 65 | 1.06 118 | 14.1 97 | 23.0 100 | 0.97 4 | 5.12 56 | 21.1 59 | 0.34 99 | 4.69 75 | 26.1 86 | 0.12 81 | 9.50 89 | 24.1 94 | 0.22 90 |
GraphCuts [14] | 72.6 | 1.90 116 | 6.51 108 | 0.02 63 | 2.96 52 | 7.63 43 | 0.22 59 | 3.79 127 | 5.16 53 | 0.64 116 | 4.49 99 | 9.24 94 | 0.55 86 | 13.9 44 | 22.7 55 | 0.99 10 | 5.04 45 | 20.8 48 | 0.21 46 | 4.53 43 | 25.2 51 | 0.15 98 | 9.88 113 | 24.9 115 | 0.21 58 |
FESL [72] | 72.6 | 1.63 88 | 5.93 76 | 0.02 63 | 2.50 33 | 6.86 35 | 0.14 34 | 1.49 63 | 5.44 75 | 0.26 75 | 3.80 60 | 8.16 50 | 0.50 72 | 14.1 97 | 22.9 92 | 1.21 102 | 5.60 106 | 22.9 106 | 0.40 113 | 4.58 57 | 25.0 42 | 0.08 32 | 9.50 89 | 24.0 93 | 0.22 90 |
Black & Anandan [4] | 73.0 | 1.63 88 | 5.12 26 | 0.01 4 | 5.17 121 | 10.5 114 | 0.41 100 | 2.30 110 | 6.36 103 | 0.47 106 | 5.20 110 | 9.84 103 | 0.49 63 | 14.0 75 | 22.9 92 | 1.02 39 | 4.88 22 | 20.2 26 | 0.18 35 | 5.10 106 | 27.0 103 | 0.08 32 | 9.23 63 | 23.1 52 | 0.21 58 |
Classic+CPF [83] | 73.0 | 1.65 95 | 6.18 95 | 0.02 63 | 2.63 37 | 7.07 38 | 0.17 40 | 1.44 55 | 5.38 71 | 0.23 63 | 3.40 22 | 7.45 20 | 0.42 27 | 14.3 111 | 23.3 113 | 1.21 102 | 5.68 108 | 23.2 112 | 0.31 84 | 4.71 78 | 25.3 60 | 0.10 57 | 9.77 105 | 24.6 107 | 0.22 90 |
TV-L1-improved [17] | 73.2 | 1.42 21 | 5.19 32 | 0.01 4 | 4.41 103 | 10.4 110 | 0.40 97 | 2.10 101 | 5.27 62 | 0.50 107 | 3.89 72 | 8.46 59 | 0.52 78 | 14.0 75 | 22.8 73 | 1.00 15 | 5.33 80 | 22.0 85 | 0.25 65 | 4.96 100 | 27.5 109 | 0.23 117 | 9.26 68 | 23.4 67 | 0.21 58 |
SRR-TVOF-NL [91] | 73.7 | 1.79 109 | 6.72 111 | 0.02 63 | 3.21 66 | 8.72 72 | 0.29 71 | 1.42 51 | 5.34 65 | 0.20 47 | 4.29 91 | 9.00 80 | 0.53 81 | 13.9 44 | 22.8 73 | 1.20 98 | 5.24 68 | 21.5 72 | 0.22 51 | 4.68 74 | 25.1 46 | 0.07 12 | 9.93 116 | 25.0 117 | 0.22 90 |
Complementary OF [21] | 73.8 | 1.61 82 | 6.47 105 | 0.01 4 | 2.74 42 | 7.75 46 | 0.19 48 | 2.67 120 | 5.71 89 | 0.89 129 | 3.74 50 | 8.76 73 | 0.41 22 | 13.9 44 | 22.7 55 | 1.14 80 | 5.19 64 | 21.4 67 | 0.31 84 | 4.98 101 | 26.9 102 | 0.13 87 | 9.78 107 | 24.8 112 | 0.21 58 |
CNN-flow-warp+ref [117] | 75.9 | 1.31 6 | 4.62 10 | 0.02 63 | 3.65 80 | 9.05 78 | 0.47 110 | 2.07 99 | 6.56 109 | 0.43 103 | 5.51 114 | 9.78 102 | 1.12 121 | 13.9 44 | 22.7 55 | 1.23 116 | 4.96 32 | 20.5 37 | 0.35 102 | 4.90 97 | 27.1 105 | 0.18 109 | 9.04 37 | 22.8 35 | 0.21 58 |
Fusion [6] | 76.4 | 1.46 32 | 5.40 44 | 0.02 63 | 2.70 39 | 6.97 37 | 0.17 40 | 1.30 20 | 4.55 15 | 0.29 88 | 4.32 92 | 8.77 75 | 0.46 52 | 14.3 111 | 23.5 115 | 1.02 39 | 5.93 120 | 24.4 123 | 0.47 122 | 4.83 92 | 26.7 98 | 0.12 81 | 10.4 122 | 26.1 124 | 0.22 90 |
BriefMatch [124] | 76.7 | 1.61 82 | 6.15 93 | 0.03 114 | 2.97 53 | 7.73 45 | 0.61 125 | 2.12 104 | 4.79 22 | 0.58 112 | 4.84 106 | 9.05 83 | 1.68 127 | 13.9 44 | 22.7 55 | 1.08 66 | 5.30 77 | 21.8 80 | 0.24 59 | 4.53 43 | 24.9 37 | 0.14 91 | 9.13 51 | 23.0 45 | 0.28 127 |
FlowNet2 [122] | 78.9 | 2.65 126 | 10.2 125 | 0.02 63 | 3.21 66 | 8.40 67 | 0.22 59 | 1.75 88 | 6.40 104 | 0.27 81 | 4.43 96 | 11.4 115 | 0.67 99 | 14.1 97 | 23.0 100 | 1.11 73 | 5.17 62 | 21.0 52 | 0.19 40 | 4.65 70 | 25.5 68 | 0.08 32 | 9.10 46 | 23.0 45 | 0.24 119 |
Rannacher [23] | 79.2 | 1.49 40 | 5.66 61 | 0.01 4 | 4.49 107 | 10.6 117 | 0.40 97 | 2.19 107 | 5.77 93 | 0.52 109 | 3.79 59 | 8.65 71 | 0.53 81 | 14.0 75 | 22.8 73 | 1.02 39 | 5.29 75 | 21.8 80 | 0.27 70 | 4.95 99 | 27.4 107 | 0.21 112 | 9.26 68 | 23.4 67 | 0.22 90 |
Horn & Schunck [3] | 83.5 | 1.62 85 | 5.42 46 | 0.01 4 | 5.40 123 | 10.9 120 | 0.44 108 | 2.27 109 | 6.97 113 | 0.45 104 | 6.35 122 | 11.5 117 | 0.65 98 | 14.1 97 | 23.0 100 | 1.06 58 | 4.95 31 | 20.4 32 | 0.17 30 | 5.51 120 | 28.2 113 | 0.14 91 | 9.57 93 | 23.7 85 | 0.18 4 |
SimpleFlow [49] | 83.5 | 1.60 81 | 6.00 83 | 0.01 4 | 3.50 74 | 8.63 71 | 0.32 81 | 2.53 115 | 6.06 99 | 0.79 121 | 3.42 24 | 7.56 26 | 0.48 60 | 14.0 75 | 22.8 73 | 1.20 98 | 5.77 112 | 23.7 116 | 0.43 119 | 5.01 103 | 28.0 111 | 0.42 127 | 9.68 100 | 24.5 105 | 0.20 25 |
TriangleFlow [30] | 84.1 | 1.73 105 | 6.50 106 | 0.02 63 | 3.84 82 | 9.51 90 | 0.31 78 | 1.78 89 | 5.71 89 | 0.27 81 | 4.43 96 | 10.1 107 | 0.57 89 | 13.7 12 | 22.5 24 | 0.98 7 | 5.69 109 | 22.9 106 | 0.23 54 | 5.16 108 | 28.3 116 | 0.21 112 | 10.1 118 | 25.4 119 | 0.21 58 |
LocallyOriented [52] | 84.1 | 1.61 82 | 6.13 91 | 0.01 4 | 4.64 111 | 10.9 120 | 0.40 97 | 1.83 91 | 6.44 105 | 0.26 75 | 4.45 98 | 10.2 109 | 0.47 58 | 14.0 75 | 22.8 73 | 1.01 25 | 5.58 104 | 22.6 98 | 0.27 70 | 5.18 110 | 26.8 101 | 0.15 98 | 9.66 98 | 24.4 100 | 0.20 25 |
2bit-BM-tele [98] | 86.0 | 1.51 45 | 5.13 27 | 0.04 124 | 4.17 94 | 10.2 105 | 0.41 100 | 1.52 69 | 4.90 32 | 0.46 105 | 4.01 82 | 8.64 69 | 0.60 93 | 14.4 115 | 23.3 113 | 1.05 55 | 5.89 116 | 24.1 118 | 0.45 121 | 5.77 126 | 32.3 130 | 0.60 130 | 8.90 22 | 22.5 15 | 0.21 58 |
Shiralkar [42] | 86.4 | 1.85 112 | 7.19 113 | 0.01 4 | 4.31 101 | 9.75 99 | 0.37 92 | 2.10 101 | 7.58 117 | 0.36 96 | 5.54 115 | 11.4 115 | 0.63 96 | 13.8 26 | 22.5 24 | 1.07 61 | 5.46 93 | 22.4 94 | 0.34 99 | 5.32 115 | 27.8 110 | 0.20 111 | 9.36 81 | 23.5 73 | 0.20 25 |
Aniso-Texture [82] | 87.0 | 1.42 21 | 5.22 36 | 0.02 63 | 4.23 98 | 10.4 110 | 0.48 111 | 2.33 111 | 5.37 68 | 0.25 71 | 4.09 85 | 9.18 91 | 0.87 115 | 14.1 97 | 23.1 108 | 1.24 122 | 6.03 124 | 24.8 128 | 0.57 130 | 4.55 51 | 24.9 37 | 0.08 32 | 9.58 95 | 24.1 94 | 0.22 90 |
TI-DOFE [24] | 88.4 | 1.76 107 | 6.02 84 | 0.01 4 | 6.21 128 | 11.7 128 | 0.51 118 | 1.97 97 | 7.23 115 | 0.28 84 | 6.30 121 | 11.3 114 | 0.85 113 | 14.0 75 | 22.8 73 | 1.02 39 | 4.96 32 | 20.4 32 | 0.15 19 | 5.21 111 | 27.1 105 | 0.15 98 | 9.86 111 | 23.8 88 | 0.26 125 |
Correlation Flow [75] | 89.0 | 1.71 104 | 6.50 106 | 0.02 63 | 4.00 88 | 10.2 105 | 0.36 87 | 1.35 32 | 4.81 23 | 0.19 42 | 3.90 74 | 8.77 75 | 0.50 72 | 14.0 75 | 22.9 92 | 1.05 55 | 6.25 130 | 24.8 128 | 0.43 119 | 5.22 112 | 28.1 112 | 0.19 110 | 9.80 108 | 24.7 110 | 0.23 114 |
SPSA-learn [13] | 89.5 | 1.59 77 | 5.32 40 | 0.01 4 | 4.23 98 | 9.36 82 | 0.42 103 | 2.54 117 | 6.27 102 | 0.80 122 | 5.24 111 | 9.44 97 | 0.73 104 | 14.0 75 | 22.8 73 | 1.03 47 | 5.27 71 | 21.7 78 | 0.31 84 | 5.93 128 | 33.0 131 | 0.86 131 | 10.4 122 | 26.2 126 | 0.20 25 |
ROF-ND [107] | 90.2 | 1.73 105 | 5.36 42 | 0.01 4 | 3.43 73 | 9.15 79 | 0.27 68 | 1.59 79 | 5.63 85 | 0.21 55 | 5.50 113 | 12.2 123 | 0.77 109 | 14.0 75 | 22.8 73 | 1.21 102 | 5.96 121 | 24.2 121 | 0.42 116 | 5.51 120 | 28.7 119 | 0.11 68 | 9.90 115 | 24.7 110 | 0.22 90 |
IAOF2 [51] | 90.7 | 1.81 111 | 6.46 104 | 0.02 63 | 4.65 112 | 11.3 126 | 0.36 87 | 1.57 75 | 5.79 95 | 0.21 55 | 4.61 103 | 10.0 106 | 0.56 87 | 14.4 115 | 23.5 115 | 1.16 83 | 5.48 95 | 22.6 98 | 0.29 78 | 4.75 84 | 25.6 71 | 0.11 68 | 9.57 93 | 24.1 94 | 0.21 58 |
StereoFlow [44] | 92.0 | 4.05 131 | 12.8 131 | 0.02 63 | 5.34 122 | 12.0 129 | 0.29 71 | 1.36 34 | 5.65 87 | 0.22 59 | 3.80 60 | 8.17 51 | 0.50 72 | 16.7 130 | 27.3 130 | 1.13 76 | 7.27 131 | 29.7 131 | 0.42 116 | 4.58 57 | 25.5 68 | 0.10 57 | 10.3 119 | 26.1 124 | 0.21 58 |
SegOF [10] | 95.2 | 1.57 74 | 6.05 87 | 0.02 63 | 3.63 78 | 8.91 75 | 0.24 64 | 2.71 121 | 6.79 111 | 0.74 120 | 4.81 105 | 11.7 119 | 0.73 104 | 14.0 75 | 22.8 73 | 1.22 114 | 5.52 97 | 22.7 101 | 0.48 124 | 5.17 109 | 28.7 119 | 0.33 123 | 9.21 58 | 23.2 56 | 0.23 114 |
SILK [79] | 98.6 | 1.86 113 | 7.35 115 | 0.01 4 | 5.84 125 | 11.2 123 | 0.60 124 | 3.00 123 | 7.69 118 | 0.83 125 | 5.68 117 | 10.6 112 | 0.69 101 | 14.1 97 | 23.0 100 | 1.00 15 | 5.31 79 | 21.3 64 | 0.36 105 | 5.02 104 | 27.0 103 | 0.28 119 | 9.50 89 | 23.5 73 | 0.24 119 |
ACK-Prior [27] | 98.7 | 1.64 91 | 6.39 103 | 0.02 63 | 2.81 44 | 7.97 51 | 0.19 48 | 2.53 115 | 5.74 91 | 0.63 115 | 4.56 102 | 10.1 107 | 1.09 120 | 14.7 123 | 24.0 124 | 1.27 129 | 6.04 125 | 24.5 126 | 0.29 78 | 5.04 105 | 27.4 107 | 0.10 57 | 11.0 127 | 27.7 128 | 0.22 90 |
Dynamic MRF [7] | 100.3 | 1.58 76 | 6.37 102 | 0.02 63 | 3.55 76 | 9.67 95 | 0.27 68 | 2.35 112 | 7.75 121 | 0.50 107 | 5.74 118 | 10.9 113 | 1.06 118 | 14.0 75 | 22.8 73 | 1.23 116 | 5.90 118 | 24.1 118 | 0.51 129 | 5.28 114 | 28.7 119 | 0.34 124 | 9.71 102 | 23.9 92 | 0.21 58 |
Adaptive flow [45] | 101.0 | 2.02 118 | 6.27 99 | 0.03 114 | 5.93 126 | 11.2 123 | 0.55 121 | 1.87 95 | 5.74 91 | 0.37 100 | 5.16 108 | 9.21 93 | 0.73 104 | 14.7 123 | 24.0 124 | 1.04 50 | 5.89 116 | 24.3 122 | 0.37 108 | 4.72 80 | 26.3 93 | 0.14 91 | 9.84 109 | 24.8 112 | 0.18 4 |
Learning Flow [11] | 102.6 | 1.62 85 | 6.12 90 | 0.02 63 | 4.43 104 | 10.4 110 | 0.33 84 | 2.86 122 | 7.92 124 | 0.82 124 | 5.19 109 | 9.72 101 | 0.59 92 | 14.6 120 | 23.8 122 | 1.10 71 | 5.42 90 | 22.4 94 | 0.27 70 | 5.24 113 | 28.3 116 | 0.17 106 | 10.3 119 | 25.4 119 | 0.23 114 |
StereoOF-V1MT [119] | 103.1 | 1.94 117 | 7.40 117 | 0.02 63 | 3.93 86 | 9.53 91 | 0.41 100 | 2.57 118 | 7.29 116 | 0.60 113 | 6.23 120 | 11.5 117 | 0.94 116 | 14.2 109 | 23.0 100 | 1.24 122 | 5.81 114 | 22.7 101 | 0.48 124 | 5.53 123 | 28.2 113 | 0.29 120 | 9.16 54 | 22.7 30 | 0.22 90 |
FOLKI [16] | 104.1 | 1.88 114 | 7.22 114 | 0.02 63 | 6.20 127 | 11.2 123 | 0.86 129 | 2.60 119 | 9.02 126 | 0.64 116 | 7.81 127 | 12.1 122 | 1.70 128 | 14.6 120 | 23.7 119 | 1.06 58 | 5.19 64 | 21.0 52 | 0.24 59 | 5.43 118 | 28.9 123 | 0.32 121 | 9.77 105 | 24.1 94 | 0.21 58 |
UnFlow [129] | 105.3 | 2.13 121 | 8.96 122 | 0.03 114 | 4.27 100 | 9.79 101 | 0.42 103 | 2.12 104 | 7.71 119 | 0.35 95 | 4.42 95 | 10.4 110 | 0.67 99 | 14.0 75 | 22.9 92 | 1.16 83 | 5.85 115 | 23.3 114 | 0.47 122 | 4.82 91 | 25.6 71 | 0.16 103 | 11.0 127 | 25.7 122 | 0.33 129 |
NL-TV-NCC [25] | 105.4 | 2.10 120 | 7.63 119 | 0.03 114 | 3.63 78 | 9.77 100 | 0.25 65 | 2.10 101 | 6.55 108 | 0.29 88 | 5.56 116 | 12.2 123 | 0.54 83 | 14.5 118 | 23.5 115 | 1.08 66 | 6.15 126 | 24.4 123 | 0.36 105 | 6.66 131 | 29.8 127 | 0.16 103 | 10.3 119 | 25.8 123 | 0.21 58 |
SLK [47] | 108.5 | 2.08 119 | 8.23 121 | 0.01 4 | 5.10 120 | 9.38 83 | 0.50 116 | 3.21 124 | 7.73 120 | 0.83 125 | 8.10 129 | 14.2 127 | 1.61 126 | 14.5 118 | 23.7 119 | 1.06 58 | 5.77 112 | 22.5 96 | 0.36 105 | 5.84 127 | 30.6 128 | 0.36 126 | 9.87 112 | 24.4 100 | 0.22 90 |
HCIC-L [99] | 109.1 | 3.37 130 | 10.9 128 | 0.07 130 | 4.98 116 | 10.3 107 | 0.38 95 | 2.44 113 | 8.00 125 | 0.36 96 | 7.09 123 | 13.0 125 | 0.71 102 | 14.9 126 | 24.2 126 | 1.07 61 | 5.99 123 | 24.1 118 | 0.23 54 | 4.74 83 | 26.0 85 | 0.11 68 | 12.3 131 | 30.4 131 | 0.25 123 |
FFV1MT [106] | 114.1 | 2.64 125 | 10.4 127 | 0.03 114 | 4.88 114 | 9.35 80 | 0.52 120 | 4.45 128 | 13.1 130 | 0.71 119 | 7.42 124 | 11.9 120 | 1.22 123 | 14.6 120 | 23.7 119 | 1.09 69 | 5.53 100 | 21.1 59 | 0.33 96 | 6.16 129 | 29.7 126 | 0.35 125 | 10.5 125 | 25.6 121 | 0.26 125 |
PGAM+LK [55] | 118.2 | 2.35 124 | 9.74 124 | 0.05 129 | 5.04 117 | 10.5 114 | 0.66 126 | 3.38 125 | 9.68 127 | 0.84 127 | 7.99 128 | 14.6 129 | 1.24 125 | 14.7 123 | 23.8 122 | 1.10 71 | 5.92 119 | 23.9 117 | 0.35 102 | 5.33 116 | 28.5 118 | 0.17 106 | 9.88 113 | 24.6 107 | 0.32 128 |
Pyramid LK [2] | 119.8 | 2.13 121 | 8.10 120 | 0.04 124 | 7.17 130 | 11.5 127 | 0.99 131 | 6.22 130 | 6.97 113 | 1.21 130 | 13.9 131 | 24.7 131 | 2.97 131 | 15.7 129 | 25.7 129 | 1.11 73 | 5.42 90 | 22.0 85 | 0.30 83 | 5.55 124 | 29.5 125 | 0.52 128 | 11.9 129 | 29.7 130 | 0.54 131 |
Heeger++ [104] | 121.8 | 3.11 129 | 11.9 130 | 0.03 114 | 4.77 113 | 9.65 94 | 0.51 118 | 4.48 129 | 12.0 129 | 0.80 122 | 7.42 124 | 11.9 120 | 1.22 123 | 15.0 127 | 24.5 127 | 1.23 116 | 6.24 128 | 23.0 110 | 0.60 131 | 6.37 130 | 29.4 124 | 0.32 121 | 10.4 122 | 25.1 118 | 0.25 123 |
GroupFlow [9] | 122.6 | 2.80 128 | 11.2 129 | 0.04 124 | 4.49 107 | 10.5 114 | 0.43 105 | 3.45 126 | 9.80 128 | 0.88 128 | 5.90 119 | 13.7 126 | 1.21 122 | 15.1 128 | 24.6 128 | 1.24 122 | 6.24 128 | 25.2 130 | 0.49 127 | 5.51 120 | 28.7 119 | 0.21 112 | 10.6 126 | 26.4 127 | 0.24 119 |
Periodicity [78] | 128.7 | 2.65 126 | 10.3 126 | 0.09 131 | 9.86 131 | 13.0 131 | 0.95 130 | 7.07 131 | 15.7 131 | 2.07 131 | 9.47 130 | 22.6 130 | 1.94 129 | 16.9 131 | 27.6 131 | 1.35 131 | 6.22 127 | 24.5 126 | 0.41 114 | 5.73 125 | 30.9 129 | 0.58 129 | 12.2 130 | 29.3 129 | 0.40 130 |
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. |