Optical flow evaluation results |
Statistics:
Average
SD
R2.5
R5.0
R10.0
A90
A95
A99
Error type: endpoint angle interpolation normalized interpolation |
R2.5 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 | |
NNF-Local [87] | 12.2 | 13.4 3 | 36.1 7 | 1.56 2 | 24.2 2 | 35.7 6 | 2.60 5 | 18.4 21 | 30.4 6 | 1.43 3 | 59.2 19 | 68.4 50 | 41.6 17 | 79.1 11 | 87.3 5 | 43.1 21 | 36.4 14 | 66.6 16 | 25.0 22 | 31.7 11 | 63.5 9 | 4.66 19 | 38.9 9 | 78.4 8 | 3.01 6 |
PH-Flow [101] | 13.9 | 13.7 20 | 37.1 30 | 1.77 24 | 24.3 5 | 35.5 5 | 2.58 3 | 18.5 23 | 30.6 9 | 1.54 16 | 58.8 1 | 66.8 4 | 41.6 17 | 79.0 5 | 87.2 3 | 42.9 11 | 36.3 9 | 67.1 50 | 24.6 7 | 31.6 2 | 63.7 18 | 4.64 11 | 39.0 17 | 78.6 16 | 3.10 27 |
NN-field [71] | 15.0 | 13.5 8 | 36.9 25 | 1.67 10 | 24.2 2 | 35.4 4 | 2.54 1 | 18.7 36 | 30.6 9 | 1.52 13 | 59.3 29 | 68.5 55 | 41.7 25 | 79.1 11 | 87.3 5 | 43.2 37 | 36.4 14 | 66.2 4 | 25.0 22 | 31.6 2 | 63.6 12 | 4.64 11 | 38.9 9 | 78.1 5 | 3.05 12 |
MDP-Flow2 [68] | 17.2 | 13.3 2 | 35.1 3 | 1.62 5 | 24.6 13 | 36.5 13 | 2.63 10 | 18.5 23 | 30.5 7 | 1.42 2 | 59.0 9 | 67.8 26 | 41.4 4 | 79.1 11 | 87.3 5 | 43.4 62 | 36.5 20 | 66.4 9 | 25.0 22 | 32.0 49 | 63.9 24 | 4.64 11 | 39.3 42 | 78.7 21 | 3.08 19 |
PMMST [114] | 17.8 | 13.4 3 | 35.0 2 | 1.70 14 | 25.1 32 | 37.1 25 | 2.73 19 | 18.5 23 | 30.5 7 | 1.39 1 | 58.9 4 | 67.4 13 | 41.5 11 | 79.2 30 | 87.4 12 | 43.4 62 | 36.3 9 | 66.2 4 | 24.9 16 | 31.8 20 | 63.8 20 | 4.67 21 | 39.2 36 | 78.7 21 | 3.09 23 |
COFM [59] | 18.2 | 13.6 15 | 36.0 5 | 1.89 48 | 24.6 13 | 36.4 11 | 2.71 17 | 18.5 23 | 30.3 4 | 1.59 25 | 58.8 1 | 66.8 4 | 41.1 2 | 79.0 5 | 87.4 12 | 42.6 6 | 35.8 3 | 67.2 60 | 24.1 2 | 31.2 1 | 61.6 1 | 4.89 86 | 38.5 3 | 78.1 5 | 3.34 85 |
Layers++ [37] | 18.7 | 14.0 48 | 37.5 42 | 1.91 51 | 24.3 5 | 35.3 3 | 2.75 21 | 18.3 19 | 31.0 18 | 1.56 20 | 59.2 19 | 67.5 16 | 41.7 25 | 79.2 30 | 87.4 12 | 43.1 21 | 36.4 14 | 66.5 13 | 25.0 22 | 31.6 2 | 63.2 4 | 4.60 1 | 38.7 6 | 77.7 4 | 3.12 33 |
AGIF+OF [85] | 20.0 | 13.9 41 | 37.5 42 | 1.67 10 | 24.6 13 | 36.5 13 | 2.68 13 | 18.1 15 | 31.0 18 | 1.61 30 | 58.9 4 | 66.9 6 | 41.4 4 | 79.2 30 | 87.5 37 | 43.1 21 | 36.6 32 | 67.2 60 | 25.0 22 | 31.8 20 | 63.6 12 | 4.60 1 | 39.0 17 | 78.6 16 | 2.98 3 |
HAST [109] | 20.3 | 13.7 20 | 36.2 12 | 1.93 57 | 24.7 21 | 37.0 23 | 2.77 30 | 18.8 39 | 32.2 42 | 1.66 38 | 59.1 13 | 67.9 34 | 41.4 4 | 79.0 5 | 87.4 12 | 42.6 6 | 36.3 9 | 66.9 32 | 24.6 7 | 31.6 2 | 63.3 6 | 4.71 37 | 39.0 17 | 78.4 8 | 3.06 13 |
Sparse-NonSparse [56] | 20.3 | 13.8 29 | 37.3 35 | 1.81 28 | 24.4 8 | 36.0 8 | 2.61 6 | 18.0 13 | 31.2 23 | 1.52 13 | 59.0 9 | 67.1 9 | 42.0 45 | 79.2 30 | 87.4 12 | 43.1 21 | 36.7 46 | 66.7 20 | 25.3 62 | 31.7 11 | 63.6 12 | 4.63 6 | 38.9 9 | 78.5 14 | 3.08 19 |
nLayers [57] | 20.8 | 13.9 41 | 36.7 21 | 1.85 39 | 24.5 10 | 36.1 9 | 2.76 24 | 17.7 6 | 30.0 3 | 1.44 4 | 59.2 19 | 67.6 18 | 41.6 17 | 79.3 54 | 87.5 37 | 43.3 47 | 36.4 14 | 66.8 27 | 25.1 36 | 31.7 11 | 63.2 4 | 4.72 42 | 38.7 6 | 77.6 3 | 3.03 8 |
ProbFlowFields [128] | 23.4 | 13.5 8 | 36.6 16 | 1.82 30 | 24.4 8 | 36.4 11 | 2.68 13 | 18.5 23 | 31.2 23 | 1.49 8 | 59.2 19 | 67.2 10 | 42.1 49 | 79.3 54 | 87.5 37 | 43.6 86 | 36.5 20 | 67.0 40 | 25.2 46 | 31.6 2 | 63.5 9 | 4.64 11 | 39.0 17 | 78.4 8 | 3.06 13 |
2DHMM-SAS [92] | 24.8 | 14.1 58 | 38.9 86 | 1.82 30 | 25.5 50 | 38.0 39 | 2.77 30 | 17.2 2 | 30.9 15 | 1.56 20 | 58.9 4 | 66.5 1 | 41.7 25 | 79.1 11 | 87.4 12 | 42.9 11 | 36.5 20 | 66.6 16 | 24.9 16 | 31.7 11 | 63.9 24 | 4.68 29 | 39.2 36 | 79.0 33 | 3.07 16 |
LSM [39] | 25.0 | 13.9 41 | 38.0 59 | 1.78 26 | 24.6 13 | 36.5 13 | 2.61 6 | 18.1 15 | 32.0 37 | 1.55 18 | 59.2 19 | 67.6 18 | 42.1 49 | 79.2 30 | 87.4 12 | 43.1 21 | 36.7 46 | 66.9 32 | 25.3 62 | 31.7 11 | 63.6 12 | 4.65 18 | 38.9 9 | 78.6 16 | 3.07 16 |
OFLAF [77] | 25.2 | 13.5 8 | 36.1 7 | 1.62 5 | 24.3 5 | 35.8 7 | 2.62 9 | 18.7 36 | 31.5 30 | 1.47 6 | 59.1 13 | 67.8 26 | 41.2 3 | 79.3 54 | 87.4 12 | 43.4 62 | 36.6 32 | 67.4 74 | 25.0 22 | 31.9 35 | 64.3 35 | 4.79 67 | 38.9 9 | 78.7 21 | 3.10 27 |
FMOF [94] | 26.4 | 14.2 66 | 38.6 74 | 1.91 51 | 24.5 10 | 36.2 10 | 2.70 16 | 18.4 21 | 31.2 23 | 1.77 55 | 59.5 38 | 68.0 36 | 41.5 11 | 79.2 30 | 87.4 12 | 43.1 21 | 36.6 32 | 66.8 27 | 25.0 22 | 31.6 2 | 63.3 6 | 4.61 4 | 39.1 26 | 78.4 8 | 3.11 32 |
CombBMOF [113] | 28.2 | 13.6 15 | 36.4 14 | 1.71 16 | 24.5 10 | 36.9 22 | 2.58 3 | 18.1 15 | 31.5 30 | 1.81 62 | 59.5 38 | 68.2 41 | 41.6 17 | 79.1 11 | 87.3 5 | 43.0 16 | 36.8 58 | 66.5 13 | 25.0 22 | 33.9 125 | 65.2 81 | 4.68 29 | 39.1 26 | 78.4 8 | 2.92 1 |
ComponentFusion [96] | 28.4 | 13.4 3 | 36.1 7 | 1.72 18 | 24.6 13 | 36.8 21 | 2.57 2 | 18.9 46 | 32.9 55 | 1.69 41 | 59.1 13 | 67.8 26 | 41.4 4 | 79.2 30 | 87.4 12 | 43.6 86 | 36.5 20 | 66.3 6 | 25.1 36 | 32.0 49 | 64.8 57 | 4.76 62 | 39.1 26 | 78.7 21 | 3.10 27 |
IROF++ [58] | 29.3 | 13.8 29 | 37.8 51 | 1.72 18 | 24.6 13 | 36.6 18 | 2.61 6 | 18.6 32 | 31.3 26 | 1.64 36 | 58.8 1 | 66.7 3 | 41.8 32 | 79.0 5 | 87.3 5 | 42.7 9 | 36.5 20 | 66.6 16 | 25.0 22 | 32.0 49 | 65.0 65 | 4.74 54 | 39.5 67 | 79.2 46 | 3.30 81 |
SepConv-v1 [127] | 30.9 | 9.23 1 | 28.0 1 | 1.08 1 | 20.5 1 | 32.4 1 | 3.35 81 | 8.95 1 | 20.5 1 | 2.08 82 | 60.8 96 | 66.9 6 | 44.2 109 | 79.1 11 | 87.1 2 | 43.2 37 | 35.6 1 | 62.4 1 | 25.1 36 | 32.2 74 | 62.3 2 | 5.34 117 | 37.6 1 | 76.4 1 | 3.28 78 |
Ramp [62] | 31.3 | 14.1 58 | 38.7 78 | 1.92 55 | 24.6 13 | 36.6 18 | 2.69 15 | 17.9 10 | 31.0 18 | 1.47 6 | 58.9 4 | 67.0 8 | 41.9 37 | 79.2 30 | 87.5 37 | 43.1 21 | 37.0 74 | 67.4 74 | 25.5 75 | 31.6 2 | 63.5 9 | 4.63 6 | 39.1 26 | 78.9 29 | 3.19 48 |
S2F-IF [123] | 32.1 | 13.5 8 | 36.6 16 | 1.70 14 | 24.9 26 | 37.9 35 | 2.77 30 | 18.8 39 | 32.7 51 | 1.54 16 | 59.1 13 | 67.7 22 | 41.6 17 | 79.3 54 | 87.5 37 | 43.3 47 | 36.5 20 | 67.1 50 | 25.0 22 | 31.9 35 | 64.7 53 | 4.74 54 | 39.3 42 | 79.1 43 | 3.10 27 |
TV-L1-MCT [64] | 33.3 | 14.5 87 | 39.7 106 | 1.86 42 | 25.2 35 | 37.8 33 | 2.78 33 | 17.3 3 | 31.1 22 | 1.59 25 | 58.9 4 | 66.6 2 | 41.6 17 | 79.1 11 | 87.4 12 | 42.9 11 | 36.8 58 | 66.4 9 | 25.6 80 | 31.8 20 | 64.0 29 | 4.73 46 | 39.1 26 | 79.0 33 | 3.20 56 |
FlowFields+ [130] | 33.8 | 13.5 8 | 37.0 28 | 1.69 13 | 25.0 29 | 38.2 45 | 2.78 33 | 18.9 46 | 33.3 66 | 1.55 18 | 59.1 13 | 67.6 18 | 41.9 37 | 79.3 54 | 87.5 37 | 43.3 47 | 36.6 32 | 67.2 60 | 25.1 36 | 31.8 20 | 64.5 45 | 4.67 21 | 39.3 42 | 79.2 46 | 3.07 16 |
RNLOD-Flow [121] | 34.0 | 13.9 41 | 37.9 56 | 1.86 42 | 25.2 35 | 37.9 35 | 2.78 33 | 19.0 52 | 32.1 39 | 1.78 56 | 59.2 19 | 67.8 26 | 41.5 11 | 79.1 11 | 87.4 12 | 43.1 21 | 36.7 46 | 66.8 27 | 25.2 46 | 31.9 35 | 64.2 32 | 4.75 58 | 39.2 36 | 79.0 33 | 3.06 13 |
FC-2Layers-FF [74] | 34.8 | 14.0 48 | 38.6 74 | 1.84 35 | 24.2 2 | 35.1 2 | 2.82 43 | 17.9 10 | 31.3 26 | 1.51 11 | 59.3 29 | 67.7 22 | 42.1 49 | 79.3 54 | 87.6 70 | 43.3 47 | 36.7 46 | 67.4 74 | 25.3 62 | 31.6 2 | 63.6 12 | 4.67 21 | 39.1 26 | 78.7 21 | 3.19 48 |
Classic+NL [31] | 35.0 | 14.2 66 | 38.8 80 | 1.98 60 | 24.6 13 | 36.5 13 | 2.65 11 | 17.7 6 | 30.9 15 | 1.51 11 | 59.2 19 | 67.5 16 | 42.2 60 | 79.2 30 | 87.5 37 | 43.3 47 | 37.0 74 | 67.1 50 | 25.5 75 | 31.7 11 | 63.6 12 | 4.67 21 | 39.2 36 | 79.0 33 | 3.18 45 |
Classic+CPF [83] | 35.8 | 14.1 58 | 38.3 67 | 1.74 21 | 24.9 26 | 37.1 25 | 2.73 19 | 17.6 5 | 31.4 28 | 1.60 28 | 59.0 9 | 67.3 11 | 41.4 4 | 79.3 54 | 87.6 70 | 43.3 47 | 36.9 65 | 67.9 99 | 25.2 46 | 31.9 35 | 64.3 35 | 4.64 11 | 39.3 42 | 79.2 46 | 3.04 9 |
FlowFields [110] | 36.2 | 13.6 15 | 37.1 30 | 1.74 21 | 25.0 29 | 38.1 40 | 2.75 21 | 18.8 39 | 33.2 64 | 1.53 15 | 59.4 33 | 68.0 36 | 42.3 69 | 79.3 54 | 87.5 37 | 43.2 37 | 36.5 20 | 67.0 40 | 25.0 22 | 31.8 20 | 64.7 53 | 4.69 32 | 39.4 54 | 79.3 54 | 3.13 34 |
NNF-EAC [103] | 38.6 | 14.2 66 | 37.3 35 | 2.09 71 | 25.3 40 | 37.6 31 | 2.76 24 | 18.9 46 | 30.6 9 | 1.61 30 | 59.8 60 | 68.5 55 | 43.3 102 | 79.1 11 | 87.3 5 | 43.1 21 | 36.5 20 | 66.5 13 | 25.0 22 | 32.1 61 | 64.3 35 | 4.73 46 | 39.4 54 | 79.0 33 | 3.14 36 |
LME [70] | 39.0 | 13.5 8 | 36.1 7 | 1.62 5 | 25.3 40 | 37.8 33 | 3.44 86 | 19.0 52 | 32.8 53 | 1.63 34 | 59.0 9 | 67.8 26 | 41.5 11 | 79.7 115 | 87.9 110 | 44.4 116 | 36.5 20 | 67.0 40 | 24.9 16 | 32.0 49 | 64.2 32 | 4.66 19 | 39.0 17 | 78.6 16 | 3.09 23 |
S2D-Matching [84] | 39.7 | 14.2 66 | 38.9 86 | 1.96 58 | 25.3 40 | 37.9 35 | 2.76 24 | 17.5 4 | 31.0 18 | 1.60 28 | 59.3 29 | 67.4 13 | 42.8 88 | 79.2 30 | 87.5 37 | 43.2 37 | 36.9 65 | 67.3 69 | 25.4 72 | 31.8 20 | 63.8 20 | 4.64 11 | 39.1 26 | 78.6 16 | 3.21 60 |
WLIF-Flow [93] | 39.7 | 13.8 29 | 37.4 39 | 1.73 20 | 24.9 26 | 37.1 25 | 2.81 40 | 18.5 23 | 30.9 15 | 1.49 8 | 59.4 33 | 67.8 26 | 42.5 81 | 79.2 30 | 87.4 12 | 43.8 107 | 37.2 87 | 67.5 81 | 25.9 96 | 31.8 20 | 63.9 24 | 4.64 11 | 39.4 54 | 78.9 29 | 3.14 36 |
FESL [72] | 42.3 | 14.4 83 | 39.1 92 | 1.83 33 | 25.0 29 | 37.4 29 | 2.76 24 | 18.2 18 | 31.6 32 | 1.70 42 | 59.7 51 | 68.5 55 | 41.7 25 | 79.3 54 | 87.6 70 | 43.3 47 | 36.9 65 | 67.9 99 | 25.2 46 | 31.8 20 | 63.8 20 | 4.61 4 | 39.3 42 | 78.8 26 | 3.04 9 |
PGM-C [120] | 46.1 | 13.8 29 | 37.7 50 | 1.85 39 | 25.1 32 | 38.1 40 | 2.90 50 | 19.1 62 | 33.6 70 | 1.59 25 | 59.3 29 | 68.2 41 | 41.9 37 | 79.3 54 | 87.5 37 | 43.5 69 | 36.6 32 | 67.2 60 | 25.2 46 | 31.9 35 | 64.8 57 | 4.67 21 | 39.5 67 | 79.4 62 | 3.22 62 |
PMF [73] | 46.8 | 13.7 20 | 37.1 30 | 1.66 9 | 25.5 50 | 39.3 67 | 2.71 17 | 19.0 52 | 34.9 98 | 1.74 52 | 59.4 33 | 68.4 50 | 41.8 32 | 79.4 84 | 87.6 70 | 43.3 47 | 37.3 91 | 66.9 32 | 26.2 104 | 31.9 35 | 64.3 35 | 4.73 46 | 39.3 42 | 78.8 26 | 2.93 2 |
MDP-Flow [26] | 46.9 | 13.4 3 | 36.1 7 | 1.67 10 | 24.8 24 | 37.2 28 | 2.79 37 | 18.8 39 | 32.0 37 | 1.70 42 | 59.8 60 | 68.9 84 | 42.1 49 | 79.3 54 | 87.6 70 | 43.5 69 | 36.7 46 | 67.7 91 | 25.2 46 | 32.5 92 | 65.5 91 | 4.77 65 | 39.1 26 | 79.0 33 | 3.09 23 |
SuperFlow [81] | 47.6 | 13.8 29 | 36.2 12 | 2.27 87 | 26.3 76 | 38.7 56 | 4.39 98 | 19.1 62 | 33.1 61 | 1.99 77 | 59.6 44 | 67.7 22 | 42.2 60 | 79.4 84 | 87.5 37 | 43.7 100 | 36.1 7 | 65.9 3 | 24.8 13 | 31.7 11 | 64.5 45 | 4.80 73 | 38.9 9 | 78.9 29 | 3.19 48 |
Efficient-NL [60] | 48.5 | 14.3 78 | 38.7 78 | 1.77 24 | 25.2 35 | 37.6 31 | 2.76 24 | 19.0 52 | 31.8 33 | 2.08 82 | 59.8 60 | 68.7 74 | 41.4 4 | 79.1 11 | 87.4 12 | 43.0 16 | 36.9 65 | 68.4 112 | 24.6 7 | 32.1 61 | 64.7 53 | 4.69 32 | 40.1 98 | 79.8 86 | 3.14 36 |
SVFilterOh [111] | 48.9 | 14.1 58 | 37.3 35 | 1.96 58 | 24.7 21 | 36.6 18 | 2.87 48 | 18.3 19 | 30.8 13 | 1.63 34 | 59.9 69 | 68.5 55 | 43.1 101 | 79.5 109 | 87.7 96 | 44.5 117 | 36.6 32 | 66.7 20 | 25.3 62 | 31.6 2 | 62.8 3 | 5.05 100 | 38.6 4 | 78.2 7 | 3.37 92 |
TC-Flow [46] | 49.2 | 13.7 20 | 36.9 25 | 1.91 51 | 25.3 40 | 38.5 51 | 3.05 62 | 19.3 80 | 34.1 87 | 1.73 48 | 59.2 19 | 67.8 26 | 42.2 60 | 79.3 54 | 87.5 37 | 43.5 69 | 37.1 80 | 68.0 102 | 25.6 80 | 31.9 35 | 64.3 35 | 4.71 37 | 39.0 17 | 79.0 33 | 3.13 34 |
AggregFlow [97] | 50.9 | 14.5 87 | 38.3 67 | 2.20 81 | 25.7 64 | 38.5 51 | 3.23 75 | 18.6 32 | 30.8 13 | 1.44 4 | 59.7 51 | 68.4 50 | 41.7 25 | 79.4 84 | 87.6 70 | 43.8 107 | 37.5 96 | 66.9 32 | 26.4 109 | 31.8 20 | 64.2 32 | 4.70 36 | 38.9 9 | 78.4 8 | 3.08 19 |
Second-order prior [8] | 51.0 | 14.0 48 | 37.1 30 | 2.11 72 | 26.2 73 | 39.3 67 | 2.93 52 | 19.4 85 | 35.1 101 | 2.16 92 | 59.4 33 | 67.8 26 | 41.8 32 | 79.1 11 | 87.3 5 | 43.1 21 | 36.5 20 | 66.7 20 | 25.0 22 | 32.3 81 | 65.4 87 | 4.74 54 | 39.5 67 | 79.6 77 | 3.19 48 |
EPPM w/o HM [88] | 51.3 | 13.4 3 | 36.6 16 | 1.61 3 | 25.5 50 | 39.3 67 | 2.76 24 | 19.4 85 | 35.7 107 | 1.99 77 | 59.6 44 | 69.3 93 | 41.9 37 | 79.2 30 | 87.4 12 | 43.1 21 | 37.0 74 | 67.5 81 | 25.3 62 | 32.8 102 | 65.0 65 | 4.85 82 | 39.4 54 | 79.0 33 | 3.04 9 |
IROF-TV [53] | 51.7 | 14.0 48 | 38.1 61 | 1.99 61 | 24.7 21 | 36.5 13 | 2.65 11 | 19.1 62 | 34.2 88 | 1.78 56 | 59.1 13 | 67.4 13 | 42.4 76 | 79.4 84 | 87.7 96 | 43.6 86 | 36.0 4 | 66.4 9 | 24.4 5 | 32.1 61 | 64.6 49 | 4.75 58 | 39.8 87 | 79.9 90 | 3.35 88 |
TF+OM [100] | 52.0 | 13.7 20 | 36.5 15 | 2.17 74 | 25.2 35 | 37.4 29 | 3.76 88 | 17.9 10 | 32.7 51 | 1.76 54 | 59.8 60 | 68.5 55 | 42.3 69 | 79.3 54 | 87.5 37 | 43.7 100 | 36.9 65 | 66.7 20 | 25.7 89 | 31.8 20 | 64.3 35 | 4.79 67 | 39.3 42 | 79.3 54 | 3.47 104 |
DeepFlow2 [108] | 52.1 | 13.9 41 | 36.6 16 | 2.07 69 | 25.6 58 | 38.4 48 | 3.08 64 | 19.1 62 | 33.6 70 | 1.70 42 | 59.6 44 | 68.5 55 | 41.9 37 | 79.4 84 | 87.5 37 | 43.7 100 | 36.7 46 | 66.3 6 | 25.4 72 | 31.9 35 | 64.7 53 | 4.67 21 | 39.4 54 | 79.4 62 | 3.26 75 |
CPM-Flow [116] | 52.7 | 13.8 29 | 37.8 51 | 1.87 46 | 25.1 32 | 38.2 45 | 2.93 52 | 19.0 52 | 33.4 68 | 1.61 30 | 59.6 44 | 68.7 74 | 42.1 49 | 79.3 54 | 87.5 37 | 43.5 69 | 36.8 58 | 66.9 32 | 25.5 75 | 32.0 49 | 65.2 81 | 4.68 29 | 39.5 67 | 79.5 70 | 3.25 71 |
TriFlow [95] | 53.6 | 14.2 66 | 39.0 90 | 2.20 81 | 26.6 81 | 39.3 67 | 4.59 102 | 19.0 52 | 33.7 73 | 1.71 47 | 59.9 69 | 68.7 74 | 41.4 4 | 79.2 30 | 87.5 37 | 43.5 69 | 36.7 46 | 67.1 50 | 25.2 46 | 31.8 20 | 63.9 24 | 4.69 32 | 39.1 26 | 79.0 33 | 3.23 67 |
SimpleFlow [49] | 53.9 | 14.1 58 | 38.9 86 | 1.92 55 | 25.5 50 | 37.9 35 | 2.85 46 | 19.0 52 | 32.3 44 | 2.26 97 | 59.2 19 | 67.3 11 | 42.4 76 | 79.2 30 | 87.5 37 | 43.2 37 | 36.7 46 | 67.6 87 | 25.1 36 | 32.0 49 | 66.1 102 | 5.29 113 | 39.3 42 | 79.2 46 | 3.15 39 |
EpicFlow [102] | 53.9 | 13.8 29 | 37.6 45 | 1.87 46 | 25.5 50 | 38.9 59 | 2.96 54 | 18.9 46 | 33.7 73 | 1.64 36 | 59.5 38 | 68.5 55 | 42.3 69 | 79.4 84 | 87.6 70 | 43.5 69 | 36.5 20 | 67.5 81 | 24.9 16 | 32.0 49 | 65.1 72 | 4.74 54 | 39.4 54 | 79.4 62 | 3.22 62 |
SRR-TVOF-NL [91] | 54.1 | 14.2 66 | 37.6 45 | 2.07 69 | 26.1 70 | 39.8 80 | 3.30 79 | 19.4 85 | 33.9 80 | 1.82 63 | 59.8 60 | 68.6 68 | 41.0 1 | 79.1 11 | 87.5 37 | 42.9 11 | 36.0 4 | 66.9 32 | 24.1 2 | 32.9 105 | 64.8 57 | 4.81 76 | 39.6 74 | 79.4 62 | 3.22 62 |
Kuang [131] | 54.1 | 13.8 29 | 38.3 67 | 1.78 26 | 25.6 58 | 39.2 65 | 2.84 45 | 19.1 62 | 33.4 68 | 1.66 38 | 59.5 38 | 68.5 55 | 42.2 60 | 79.3 54 | 87.6 70 | 43.2 37 | 36.5 20 | 67.1 50 | 24.8 13 | 32.2 74 | 65.9 98 | 4.80 73 | 39.6 74 | 79.6 77 | 3.19 48 |
DeepFlow [86] | 54.3 | 13.7 20 | 35.7 4 | 2.03 66 | 25.6 58 | 38.2 45 | 3.30 79 | 19.2 73 | 33.9 80 | 1.74 52 | 59.7 51 | 68.0 36 | 42.2 60 | 79.4 84 | 87.5 37 | 43.7 100 | 37.3 91 | 66.4 9 | 26.2 104 | 31.8 20 | 64.8 57 | 4.63 6 | 39.3 42 | 79.3 54 | 3.26 75 |
CostFilter [40] | 55.1 | 13.6 15 | 37.4 39 | 1.63 8 | 25.5 50 | 39.7 78 | 2.75 21 | 19.0 52 | 36.0 110 | 1.79 58 | 59.4 33 | 68.8 80 | 42.0 45 | 79.4 84 | 87.6 70 | 43.7 100 | 38.6 113 | 67.1 50 | 28.1 122 | 31.9 35 | 64.6 49 | 4.81 76 | 39.0 17 | 78.5 14 | 3.00 4 |
OFH [38] | 55.4 | 14.1 58 | 38.2 66 | 2.03 66 | 25.6 58 | 38.4 48 | 3.01 58 | 19.4 85 | 35.1 101 | 1.79 58 | 59.5 38 | 68.8 80 | 42.3 69 | 79.1 11 | 87.4 12 | 43.1 21 | 36.7 46 | 67.6 87 | 25.2 46 | 32.1 61 | 65.1 72 | 4.79 67 | 39.2 36 | 79.2 46 | 3.15 39 |
Complementary OF [21] | 55.7 | 13.7 20 | 37.8 51 | 1.71 16 | 25.2 35 | 38.6 54 | 2.81 40 | 19.8 106 | 33.7 73 | 2.38 102 | 59.9 69 | 69.2 92 | 42.8 88 | 79.2 30 | 87.5 37 | 43.1 21 | 36.6 32 | 67.4 74 | 25.2 46 | 32.3 81 | 65.4 87 | 4.79 67 | 38.8 8 | 78.9 29 | 3.29 79 |
RFlow [90] | 55.9 | 13.8 29 | 37.8 51 | 2.02 64 | 26.0 67 | 39.1 64 | 2.85 46 | 19.0 52 | 33.1 61 | 1.86 66 | 59.7 51 | 68.4 50 | 42.2 60 | 79.2 30 | 87.6 70 | 43.4 62 | 36.1 7 | 66.8 27 | 24.5 6 | 32.2 74 | 65.1 72 | 4.82 81 | 39.7 81 | 79.8 86 | 3.34 85 |
DPOF [18] | 56.1 | 14.2 66 | 39.1 92 | 2.19 80 | 24.8 24 | 37.0 23 | 2.80 38 | 19.3 80 | 31.9 34 | 2.01 79 | 60.2 84 | 69.5 101 | 42.3 69 | 79.1 11 | 87.4 12 | 43.1 21 | 36.7 46 | 67.1 50 | 24.6 7 | 32.4 87 | 65.3 85 | 4.81 76 | 39.5 67 | 79.5 70 | 3.18 45 |
Aniso. Huber-L1 [22] | 56.3 | 14.3 78 | 38.5 72 | 2.17 74 | 26.6 81 | 39.5 76 | 3.21 74 | 19.2 73 | 32.5 49 | 1.83 65 | 59.7 51 | 68.7 74 | 41.9 37 | 79.2 30 | 87.4 12 | 43.2 37 | 36.3 9 | 67.1 50 | 24.6 7 | 32.2 74 | 64.9 63 | 4.71 37 | 39.7 81 | 79.6 77 | 3.24 70 |
OAR-Flow [125] | 56.9 | 14.0 48 | 36.9 25 | 2.05 68 | 25.3 40 | 38.1 40 | 3.11 67 | 19.1 62 | 34.0 86 | 1.70 42 | 59.2 19 | 68.6 68 | 41.9 37 | 79.4 84 | 87.6 70 | 43.5 69 | 36.9 65 | 67.8 94 | 25.3 62 | 32.0 49 | 65.1 72 | 4.75 58 | 39.3 42 | 79.3 54 | 3.18 45 |
TC/T-Flow [76] | 57.1 | 14.3 78 | 38.8 80 | 1.84 35 | 25.3 40 | 38.6 54 | 2.81 40 | 18.9 46 | 32.4 48 | 1.58 23 | 59.9 69 | 69.5 101 | 42.1 49 | 79.3 54 | 87.5 37 | 43.5 69 | 37.1 80 | 68.0 102 | 25.2 46 | 32.1 61 | 65.2 81 | 4.81 76 | 39.2 36 | 79.4 62 | 3.00 4 |
Brox et al. [5] | 57.4 | 14.0 48 | 37.4 39 | 1.90 49 | 26.4 78 | 40.1 87 | 3.08 64 | 19.3 80 | 35.0 100 | 1.97 74 | 59.7 51 | 68.2 41 | 41.7 25 | 79.4 84 | 87.6 70 | 43.6 86 | 36.6 32 | 66.9 32 | 25.1 36 | 31.9 35 | 64.8 57 | 4.73 46 | 39.4 54 | 79.5 70 | 3.15 39 |
Sparse Occlusion [54] | 58.3 | 14.2 66 | 38.6 74 | 1.99 61 | 25.8 66 | 39.2 65 | 2.78 33 | 19.3 80 | 32.3 44 | 1.80 61 | 59.8 60 | 68.8 80 | 41.7 25 | 79.3 54 | 87.5 37 | 43.2 37 | 37.1 80 | 68.4 112 | 25.3 62 | 32.1 61 | 64.4 44 | 4.60 1 | 39.7 81 | 79.6 77 | 3.15 39 |
ComplOF-FED-GPU [35] | 58.8 | 14.0 48 | 38.0 59 | 1.91 51 | 25.3 40 | 38.5 51 | 2.90 50 | 20.2 110 | 34.6 94 | 2.16 92 | 59.5 38 | 68.5 55 | 42.5 81 | 79.2 30 | 87.4 12 | 43.2 37 | 36.6 32 | 67.4 74 | 25.0 22 | 32.2 74 | 65.4 87 | 4.75 58 | 39.7 81 | 79.8 86 | 3.19 48 |
Aniso-Texture [82] | 59.0 | 13.6 15 | 36.6 16 | 1.82 30 | 26.2 73 | 39.3 67 | 3.20 73 | 19.6 95 | 33.0 58 | 1.96 73 | 59.7 51 | 68.5 55 | 42.6 84 | 79.4 84 | 87.6 70 | 43.6 86 | 37.0 74 | 68.4 112 | 25.7 89 | 31.9 35 | 63.8 20 | 4.63 6 | 39.4 54 | 79.3 54 | 3.16 43 |
GraphCuts [14] | 59.1 | 15.1 103 | 39.3 96 | 2.68 99 | 26.4 78 | 39.4 74 | 4.50 100 | 19.2 73 | 30.7 12 | 2.69 108 | 60.7 94 | 68.6 68 | 42.8 88 | 79.0 5 | 87.4 12 | 42.5 5 | 35.6 1 | 66.7 20 | 23.7 1 | 32.0 49 | 65.0 65 | 5.04 99 | 39.0 17 | 79.2 46 | 3.48 105 |
Fusion [6] | 59.9 | 13.8 29 | 38.4 71 | 1.84 35 | 25.3 40 | 38.1 40 | 2.88 49 | 19.1 62 | 32.2 42 | 1.90 69 | 60.9 98 | 69.8 105 | 41.8 32 | 79.1 11 | 87.9 110 | 42.1 2 | 36.0 4 | 67.8 94 | 24.1 2 | 32.7 100 | 66.3 106 | 4.88 85 | 39.5 67 | 80.4 109 | 3.26 75 |
DF-Auto [115] | 60.4 | 14.2 66 | 36.7 21 | 2.25 85 | 26.5 80 | 39.0 62 | 4.23 94 | 18.8 39 | 31.4 28 | 1.58 23 | 60.1 80 | 69.3 93 | 41.6 17 | 79.3 54 | 87.5 37 | 43.6 86 | 36.6 32 | 67.0 40 | 25.1 36 | 32.3 81 | 65.1 72 | 4.81 76 | 39.9 88 | 80.1 98 | 3.22 62 |
Classic++ [32] | 60.6 | 14.0 48 | 38.1 61 | 2.17 74 | 25.7 64 | 38.8 57 | 2.96 54 | 19.3 80 | 33.9 80 | 1.93 70 | 59.7 51 | 67.9 34 | 42.8 88 | 79.2 30 | 87.5 37 | 43.3 47 | 37.4 95 | 67.0 40 | 26.6 111 | 31.8 20 | 64.3 35 | 4.78 66 | 39.4 54 | 79.5 70 | 3.36 89 |
Steered-L1 [118] | 60.8 | 13.7 20 | 37.5 42 | 1.84 35 | 25.5 50 | 38.9 59 | 3.17 72 | 19.7 102 | 33.1 61 | 2.40 103 | 60.2 84 | 68.5 55 | 42.8 88 | 79.4 84 | 87.7 96 | 43.5 69 | 36.6 32 | 67.0 40 | 25.6 80 | 31.8 20 | 64.6 49 | 4.96 93 | 38.6 4 | 79.0 33 | 3.36 89 |
ALD-Flow [66] | 61.9 | 14.1 58 | 37.9 56 | 2.17 74 | 25.4 48 | 38.4 48 | 3.14 69 | 19.1 62 | 33.9 80 | 1.73 48 | 59.6 44 | 69.0 88 | 42.6 84 | 79.4 84 | 87.6 70 | 43.6 86 | 37.0 74 | 67.5 81 | 25.6 80 | 31.7 11 | 64.0 29 | 4.69 32 | 39.4 54 | 79.5 70 | 3.20 56 |
p-harmonic [29] | 62.1 | 13.5 8 | 36.7 21 | 1.85 39 | 26.7 88 | 39.9 84 | 3.25 77 | 19.4 85 | 35.2 103 | 2.10 85 | 60.1 80 | 68.7 74 | 42.2 60 | 79.3 54 | 87.5 37 | 43.3 47 | 36.7 46 | 66.7 20 | 25.3 62 | 32.6 96 | 65.8 95 | 4.76 62 | 39.4 54 | 79.5 70 | 3.17 44 |
Shiralkar [42] | 65.1 | 14.2 66 | 39.0 90 | 2.02 64 | 26.8 89 | 40.3 91 | 2.98 56 | 18.5 23 | 38.0 122 | 2.48 106 | 60.1 80 | 67.7 22 | 41.8 32 | 78.8 2 | 87.2 3 | 42.3 4 | 37.7 102 | 67.2 60 | 26.2 104 | 33.2 111 | 67.1 110 | 4.94 90 | 39.4 54 | 79.3 54 | 3.10 27 |
HBM-GC [105] | 68.0 | 14.7 93 | 39.4 101 | 2.41 94 | 25.4 48 | 38.1 40 | 3.07 63 | 18.0 13 | 29.8 2 | 1.56 20 | 59.8 60 | 68.2 41 | 42.8 88 | 80.1 121 | 88.0 116 | 45.9 124 | 37.5 96 | 68.2 108 | 26.1 102 | 31.9 35 | 63.3 6 | 4.99 95 | 39.3 42 | 79.1 43 | 3.30 81 |
FlowNet2 [122] | 68.1 | 15.9 115 | 41.4 114 | 2.76 102 | 27.1 93 | 40.2 89 | 4.29 96 | 19.6 95 | 34.3 90 | 1.88 67 | 60.0 74 | 70.2 109 | 42.0 45 | 79.4 84 | 87.7 96 | 43.3 47 | 36.4 14 | 66.3 6 | 24.9 16 | 32.1 61 | 64.5 45 | 4.71 37 | 39.6 74 | 79.2 46 | 3.08 19 |
SIOF [67] | 69.2 | 14.7 93 | 39.5 103 | 2.23 84 | 27.1 93 | 40.3 91 | 4.25 95 | 19.1 62 | 32.9 55 | 1.82 63 | 59.8 60 | 68.6 68 | 42.1 49 | 79.1 11 | 87.4 12 | 43.0 16 | 37.1 80 | 67.1 50 | 25.5 75 | 32.4 87 | 64.9 63 | 4.79 67 | 40.1 98 | 79.9 90 | 3.40 96 |
CLG-TV [48] | 69.5 | 14.3 78 | 38.8 80 | 2.17 74 | 26.6 81 | 39.8 80 | 3.24 76 | 19.5 92 | 33.9 80 | 2.11 87 | 60.0 74 | 69.0 88 | 42.4 76 | 79.3 54 | 87.6 70 | 43.5 69 | 36.6 32 | 66.9 32 | 25.1 36 | 32.1 61 | 65.1 72 | 4.71 37 | 39.9 88 | 80.0 94 | 3.20 56 |
MLDP_OF [89] | 70.1 | 13.9 41 | 38.1 61 | 1.81 28 | 25.6 58 | 38.9 59 | 2.80 38 | 18.8 39 | 32.3 44 | 1.61 30 | 59.6 44 | 68.3 47 | 42.3 69 | 79.3 54 | 87.6 70 | 43.9 110 | 39.6 125 | 68.7 117 | 28.5 124 | 33.0 109 | 65.3 85 | 5.09 103 | 39.6 74 | 79.2 46 | 3.51 107 |
Local-TV-L1 [65] | 71.9 | 14.9 98 | 37.3 35 | 3.21 114 | 27.3 98 | 39.5 76 | 4.67 103 | 18.9 46 | 32.3 44 | 1.70 42 | 61.3 109 | 68.6 68 | 47.1 123 | 79.3 54 | 87.6 70 | 43.6 86 | 39.0 117 | 66.7 20 | 28.9 126 | 31.7 11 | 64.3 35 | 4.79 67 | 39.3 42 | 79.1 43 | 3.41 99 |
IAOF [50] | 72.6 | 15.5 111 | 39.2 95 | 2.93 111 | 29.4 113 | 43.0 115 | 5.18 112 | 17.8 8 | 33.0 58 | 2.04 80 | 60.8 96 | 68.9 84 | 42.2 60 | 79.2 30 | 87.4 12 | 43.3 47 | 36.8 58 | 67.2 60 | 25.1 36 | 32.7 100 | 65.6 94 | 4.67 21 | 40.0 92 | 80.0 94 | 3.20 56 |
F-TV-L1 [15] | 72.7 | 15.0 99 | 39.3 96 | 2.88 109 | 27.2 96 | 40.2 89 | 3.69 87 | 19.2 73 | 34.5 93 | 2.19 94 | 59.7 51 | 68.4 50 | 42.8 88 | 78.9 3 | 87.4 12 | 42.7 9 | 37.3 91 | 67.0 40 | 25.6 80 | 32.1 61 | 64.5 45 | 4.89 86 | 40.1 98 | 80.0 94 | 3.42 100 |
TCOF [69] | 73.6 | 14.4 83 | 39.3 96 | 1.83 33 | 27.3 98 | 40.9 101 | 3.35 81 | 18.7 36 | 32.1 39 | 1.50 10 | 60.2 84 | 70.2 109 | 42.1 49 | 79.3 54 | 87.6 70 | 43.2 37 | 36.9 65 | 68.5 115 | 24.8 13 | 33.3 112 | 65.8 95 | 4.72 42 | 41.2 120 | 81.4 122 | 3.46 102 |
BriefMatch [124] | 74.6 | 14.0 48 | 37.0 28 | 2.17 74 | 25.6 58 | 38.8 57 | 3.98 92 | 19.7 102 | 33.0 58 | 2.69 108 | 61.1 105 | 69.0 88 | 46.4 120 | 79.3 54 | 87.6 70 | 43.8 107 | 40.5 128 | 67.9 99 | 30.6 128 | 31.8 20 | 64.0 29 | 4.94 90 | 39.0 17 | 78.8 26 | 3.34 85 |
CNN-flow-warp+ref [117] | 76.2 | 13.8 29 | 36.0 5 | 2.35 91 | 26.6 81 | 39.8 80 | 3.83 89 | 20.0 108 | 35.5 106 | 2.34 99 | 60.9 98 | 68.9 84 | 43.0 99 | 79.4 84 | 87.6 70 | 43.7 100 | 36.8 58 | 67.0 40 | 25.6 80 | 32.1 61 | 66.2 104 | 4.94 90 | 39.4 54 | 79.5 70 | 3.19 48 |
Adaptive [20] | 76.2 | 14.5 87 | 39.6 105 | 2.31 89 | 27.1 93 | 40.4 94 | 3.35 81 | 18.6 32 | 33.7 73 | 1.98 75 | 59.6 44 | 68.2 41 | 42.4 76 | 79.4 84 | 87.6 70 | 43.4 62 | 37.1 80 | 67.5 81 | 25.7 89 | 32.4 87 | 64.8 57 | 4.73 46 | 40.0 92 | 80.1 98 | 3.38 94 |
FlowNetS+ft+v [112] | 77.0 | 14.7 93 | 38.1 61 | 2.80 105 | 27.5 100 | 40.6 98 | 4.81 106 | 19.6 95 | 34.9 98 | 2.07 81 | 60.1 80 | 69.5 101 | 42.2 60 | 79.4 84 | 87.7 96 | 43.4 62 | 36.6 32 | 67.1 50 | 25.2 46 | 32.0 49 | 65.4 87 | 4.73 46 | 39.6 74 | 79.7 84 | 3.21 60 |
SPSA-learn [13] | 77.6 | 14.8 97 | 37.8 51 | 2.72 101 | 27.6 101 | 40.1 87 | 4.71 104 | 20.5 112 | 33.7 73 | 2.97 115 | 60.4 88 | 67.6 18 | 41.5 11 | 79.3 54 | 87.5 37 | 43.5 69 | 36.8 58 | 67.2 60 | 25.2 46 | 33.4 114 | 70.8 131 | 6.21 130 | 39.7 81 | 79.6 77 | 3.19 48 |
AdaConv-v1 [126] | 78.1 | 16.5 119 | 42.3 117 | 4.36 121 | 30.4 119 | 43.8 118 | 9.06 127 | 20.6 115 | 36.3 114 | 4.45 127 | 64.5 125 | 71.3 123 | 45.3 115 | 78.4 1 | 86.7 1 | 42.2 3 | 36.3 9 | 64.9 2 | 25.4 72 | 32.5 92 | 63.7 18 | 5.53 123 | 38.0 2 | 77.4 2 | 3.53 109 |
LDOF [28] | 78.8 | 15.0 99 | 38.8 80 | 2.92 110 | 28.0 106 | 41.1 103 | 5.03 109 | 19.7 102 | 34.8 97 | 2.15 90 | 60.0 74 | 68.9 84 | 42.6 84 | 79.4 84 | 87.6 70 | 43.5 69 | 36.9 65 | 66.8 27 | 25.5 75 | 31.9 35 | 65.1 72 | 4.73 46 | 39.5 67 | 79.6 77 | 3.23 67 |
HBpMotionGpu [43] | 79.2 | 15.8 114 | 40.2 110 | 3.66 119 | 29.5 114 | 42.8 114 | 6.27 117 | 18.5 23 | 31.9 34 | 1.73 48 | 61.3 109 | 69.9 106 | 43.9 107 | 79.1 11 | 87.6 70 | 43.0 16 | 37.6 101 | 67.6 87 | 25.9 96 | 32.0 49 | 64.3 35 | 4.67 21 | 40.0 92 | 79.9 90 | 3.75 118 |
ROF-ND [107] | 79.2 | 15.1 103 | 37.9 56 | 1.86 42 | 26.3 76 | 40.5 97 | 3.12 68 | 19.6 95 | 32.8 53 | 1.68 40 | 60.9 98 | 71.1 122 | 41.9 37 | 79.3 54 | 87.5 37 | 43.5 69 | 37.0 74 | 68.2 108 | 24.9 16 | 34.3 128 | 68.3 119 | 5.28 112 | 40.5 114 | 80.5 112 | 3.25 71 |
CRTflow [80] | 79.4 | 14.4 83 | 38.9 86 | 2.38 92 | 26.0 67 | 39.0 62 | 3.14 69 | 20.2 110 | 36.2 113 | 2.37 101 | 60.5 91 | 69.5 101 | 44.1 108 | 79.3 54 | 87.5 37 | 43.4 62 | 37.1 80 | 67.3 69 | 25.7 89 | 32.0 49 | 64.6 49 | 4.85 82 | 39.6 74 | 79.6 77 | 3.45 101 |
Occlusion-TV-L1 [63] | 79.8 | 14.3 78 | 39.1 92 | 2.21 83 | 26.6 81 | 40.0 85 | 3.14 69 | 19.2 73 | 34.2 88 | 2.15 90 | 60.0 74 | 68.5 55 | 42.8 88 | 79.3 54 | 87.5 37 | 43.6 86 | 37.5 96 | 67.0 40 | 26.2 104 | 32.9 105 | 65.1 72 | 5.16 107 | 40.0 92 | 79.8 86 | 3.30 81 |
Modified CLG [34] | 79.9 | 14.1 58 | 37.6 45 | 2.33 90 | 28.5 110 | 41.4 107 | 5.68 113 | 19.6 95 | 35.8 109 | 2.31 98 | 60.2 84 | 68.6 68 | 42.1 49 | 79.4 84 | 87.5 37 | 43.5 69 | 36.7 46 | 67.2 60 | 25.2 46 | 32.3 81 | 66.0 100 | 4.76 62 | 40.2 102 | 80.4 109 | 3.40 96 |
CBF [12] | 81.1 | 13.7 20 | 37.2 34 | 2.15 73 | 26.0 67 | 39.4 74 | 3.28 78 | 19.1 62 | 32.1 39 | 1.79 58 | 61.0 103 | 70.0 108 | 45.8 117 | 79.6 112 | 87.8 109 | 44.9 120 | 36.8 58 | 67.4 74 | 25.2 46 | 32.2 74 | 65.5 91 | 5.22 109 | 40.0 92 | 80.2 105 | 3.99 123 |
TriangleFlow [30] | 82.9 | 14.7 93 | 40.0 109 | 2.29 88 | 26.6 81 | 40.8 99 | 3.03 60 | 19.4 85 | 33.3 66 | 2.10 85 | 60.4 88 | 69.9 106 | 42.8 88 | 79.0 5 | 87.4 12 | 42.6 6 | 37.7 102 | 68.3 111 | 25.3 62 | 33.1 110 | 67.8 114 | 5.24 111 | 40.4 110 | 80.6 114 | 3.32 84 |
2D-CLG [1] | 83.0 | 14.5 87 | 37.6 45 | 2.76 102 | 29.8 116 | 42.4 111 | 6.69 121 | 19.7 102 | 35.2 103 | 2.74 111 | 60.7 94 | 68.7 74 | 41.5 11 | 79.4 84 | 87.7 96 | 43.5 69 | 36.6 32 | 67.0 40 | 25.1 36 | 32.5 92 | 66.7 108 | 4.90 88 | 40.2 102 | 80.1 98 | 3.25 71 |
Nguyen [33] | 83.4 | 15.6 112 | 38.5 72 | 3.62 118 | 30.1 118 | 43.2 116 | 6.04 115 | 19.6 95 | 36.3 114 | 2.25 96 | 61.1 105 | 69.4 98 | 42.0 45 | 79.2 30 | 87.5 37 | 43.1 21 | 36.4 14 | 67.2 60 | 24.7 12 | 34.3 128 | 67.4 113 | 5.00 96 | 40.2 102 | 80.3 106 | 3.29 79 |
BlockOverlap [61] | 83.5 | 15.1 103 | 37.6 45 | 3.31 116 | 27.7 103 | 39.3 67 | 5.73 114 | 18.6 32 | 30.3 4 | 2.09 84 | 60.9 98 | 68.2 41 | 47.1 123 | 80.2 122 | 87.9 110 | 46.5 125 | 39.0 117 | 67.3 69 | 28.4 123 | 31.9 35 | 63.9 24 | 5.09 103 | 39.7 81 | 79.3 54 | 3.55 110 |
SegOF [10] | 84.0 | 14.2 66 | 36.8 24 | 2.54 96 | 27.0 92 | 40.0 85 | 4.18 93 | 21.1 118 | 36.1 112 | 3.15 120 | 60.5 91 | 70.7 118 | 41.6 17 | 79.4 84 | 87.6 70 | 43.6 86 | 36.9 65 | 68.2 108 | 25.2 46 | 32.5 92 | 68.0 118 | 5.31 116 | 39.6 74 | 79.4 62 | 3.22 62 |
ACK-Prior [27] | 84.3 | 13.8 29 | 38.1 61 | 1.74 21 | 25.5 50 | 39.3 67 | 2.82 43 | 19.6 95 | 33.8 79 | 2.45 105 | 60.5 91 | 70.3 111 | 42.3 69 | 80.2 122 | 88.0 116 | 45.8 123 | 38.2 107 | 67.8 94 | 26.9 115 | 32.6 96 | 66.2 104 | 5.35 118 | 38.9 9 | 79.7 84 | 3.60 114 |
IAOF2 [51] | 84.9 | 15.6 112 | 41.3 113 | 2.58 98 | 27.6 101 | 41.4 107 | 4.29 96 | 17.8 8 | 33.6 70 | 1.94 71 | 61.2 108 | 70.8 119 | 42.8 88 | 79.4 84 | 87.7 96 | 43.3 47 | 37.2 87 | 67.5 81 | 25.6 80 | 32.3 81 | 65.0 65 | 4.63 6 | 40.6 115 | 80.4 109 | 3.40 96 |
StereoOF-V1MT [119] | 86.0 | 14.6 91 | 39.9 107 | 2.00 63 | 27.2 96 | 41.9 109 | 3.04 61 | 20.9 117 | 37.8 120 | 2.85 114 | 61.3 109 | 68.3 47 | 43.8 105 | 79.2 30 | 87.5 37 | 42.9 11 | 38.2 107 | 67.8 94 | 26.3 108 | 33.8 121 | 68.5 120 | 5.36 119 | 40.0 92 | 79.4 62 | 3.09 23 |
Dynamic MRF [7] | 86.5 | 13.9 41 | 38.6 74 | 1.90 49 | 26.1 70 | 40.4 94 | 3.08 64 | 20.0 108 | 37.7 119 | 2.73 110 | 61.3 109 | 69.3 93 | 44.6 110 | 79.1 11 | 87.6 70 | 43.0 16 | 37.7 102 | 68.0 102 | 25.9 96 | 32.6 96 | 67.2 111 | 5.08 102 | 40.4 110 | 80.5 112 | 3.49 106 |
TV-L1-improved [17] | 86.7 | 14.2 66 | 38.8 80 | 2.25 85 | 26.9 90 | 40.3 91 | 3.40 85 | 19.5 92 | 33.9 80 | 2.44 104 | 59.9 69 | 69.0 88 | 42.7 87 | 79.4 84 | 87.7 96 | 43.5 69 | 37.2 87 | 67.6 87 | 25.8 93 | 32.1 61 | 66.1 102 | 5.05 100 | 39.9 88 | 80.0 94 | 3.46 102 |
Correlation Flow [75] | 86.7 | 14.0 48 | 38.3 67 | 1.61 3 | 26.2 73 | 39.8 80 | 2.98 56 | 19.1 62 | 31.9 34 | 1.73 48 | 60.4 88 | 69.4 98 | 43.6 104 | 80.2 122 | 87.9 110 | 47.8 128 | 38.0 106 | 68.7 117 | 26.0 100 | 33.4 114 | 67.2 111 | 5.29 113 | 40.1 98 | 80.3 106 | 3.39 95 |
Black & Anandan [4] | 89.5 | 15.3 107 | 38.8 80 | 2.96 112 | 28.4 108 | 40.9 101 | 4.78 105 | 20.5 112 | 35.2 103 | 2.74 111 | 60.9 98 | 69.3 93 | 42.1 49 | 79.4 84 | 87.7 96 | 43.6 86 | 37.1 80 | 66.6 16 | 25.6 80 | 32.9 105 | 65.9 98 | 4.72 42 | 40.3 105 | 80.3 106 | 3.25 71 |
Rannacher [23] | 90.0 | 14.4 83 | 39.3 96 | 2.38 92 | 26.9 90 | 40.4 94 | 3.36 84 | 19.5 92 | 34.6 94 | 2.58 107 | 59.8 60 | 68.8 80 | 42.8 88 | 79.4 84 | 87.7 96 | 43.6 86 | 37.2 87 | 67.8 94 | 25.8 93 | 32.2 74 | 66.0 100 | 5.02 97 | 39.9 88 | 79.9 90 | 3.56 111 |
LocallyOriented [52] | 91.5 | 15.0 99 | 40.3 111 | 2.53 95 | 27.7 103 | 41.3 105 | 3.86 90 | 19.4 85 | 34.4 91 | 1.95 72 | 61.1 105 | 70.6 115 | 43.3 102 | 79.2 30 | 87.5 37 | 43.3 47 | 39.1 121 | 68.1 106 | 27.6 120 | 32.9 105 | 65.8 95 | 4.72 42 | 40.6 115 | 80.6 114 | 3.37 92 |
UnFlow [129] | 93.0 | 16.0 116 | 42.8 119 | 2.87 108 | 30.6 121 | 45.2 126 | 4.52 101 | 21.3 121 | 39.4 125 | 2.81 113 | 60.0 74 | 68.3 47 | 42.1 49 | 79.2 30 | 87.4 12 | 43.5 69 | 37.5 96 | 68.0 102 | 25.2 46 | 33.8 121 | 65.1 72 | 4.98 94 | 43.2 130 | 81.8 125 | 3.67 116 |
Filter Flow [19] | 96.8 | 15.0 99 | 39.4 101 | 2.78 104 | 28.4 108 | 40.8 99 | 6.31 118 | 18.5 23 | 32.9 55 | 2.14 88 | 61.7 113 | 69.3 93 | 45.3 115 | 79.7 115 | 88.0 116 | 44.5 117 | 37.3 91 | 67.7 91 | 26.1 102 | 32.1 61 | 65.2 81 | 4.93 89 | 40.3 105 | 80.7 117 | 3.97 122 |
StereoFlow [44] | 98.5 | 22.8 130 | 51.1 131 | 4.80 122 | 36.2 130 | 51.1 131 | 6.57 120 | 19.2 73 | 34.6 94 | 1.89 68 | 60.0 74 | 68.5 55 | 42.4 76 | 80.3 125 | 89.1 130 | 43.9 110 | 39.0 117 | 74.1 131 | 25.3 62 | 32.1 61 | 65.0 65 | 4.73 46 | 40.3 105 | 80.9 118 | 3.36 89 |
Ad-TV-NDC [36] | 101.9 | 17.2 121 | 39.9 107 | 5.26 124 | 29.6 115 | 42.1 110 | 6.18 116 | 19.2 73 | 33.7 73 | 1.98 75 | 62.4 115 | 70.3 111 | 45.2 114 | 79.6 112 | 87.9 110 | 43.9 110 | 38.3 110 | 67.3 69 | 27.2 119 | 32.3 81 | 65.5 91 | 4.80 73 | 40.3 105 | 80.1 98 | 3.58 113 |
Bartels [41] | 103.1 | 14.6 91 | 39.3 96 | 2.80 105 | 26.1 70 | 39.7 78 | 4.45 99 | 19.0 52 | 33.2 64 | 2.14 88 | 62.1 114 | 70.9 120 | 48.9 125 | 80.7 129 | 88.1 119 | 49.2 130 | 43.7 130 | 69.0 123 | 34.8 130 | 32.4 87 | 65.0 65 | 5.76 127 | 40.4 110 | 80.1 98 | 4.26 125 |
TI-DOFE [24] | 104.5 | 17.9 124 | 43.0 120 | 5.41 125 | 32.3 126 | 46.2 128 | 7.98 125 | 20.5 112 | 38.1 123 | 2.97 115 | 63.1 122 | 70.6 115 | 43.8 105 | 79.1 11 | 87.6 70 | 43.1 21 | 37.7 102 | 67.4 74 | 25.8 93 | 33.4 114 | 67.8 114 | 5.09 103 | 41.6 124 | 81.5 124 | 3.68 117 |
Horn & Schunck [3] | 105.5 | 15.3 107 | 40.4 112 | 2.69 100 | 29.0 111 | 42.7 112 | 5.10 110 | 21.1 118 | 37.9 121 | 3.33 121 | 62.5 117 | 70.3 111 | 43.0 99 | 79.3 54 | 87.7 96 | 43.6 86 | 37.5 96 | 67.3 69 | 25.9 96 | 33.9 125 | 68.5 120 | 5.03 98 | 41.2 120 | 81.2 121 | 3.57 112 |
GroupFlow [9] | 107.6 | 16.8 120 | 43.4 121 | 3.43 117 | 29.1 112 | 43.9 119 | 5.11 111 | 22.2 124 | 39.3 124 | 3.53 122 | 61.0 103 | 70.6 115 | 42.5 81 | 79.7 115 | 88.1 119 | 44.0 113 | 39.0 117 | 69.4 127 | 26.8 114 | 32.8 102 | 66.8 109 | 4.87 84 | 40.4 110 | 80.1 98 | 3.01 6 |
2bit-BM-tele [98] | 109.3 | 15.3 107 | 39.5 103 | 3.22 115 | 27.8 105 | 41.2 104 | 4.90 107 | 18.8 39 | 32.5 49 | 2.34 99 | 62.4 115 | 71.0 121 | 49.0 126 | 80.6 127 | 88.2 124 | 47.9 129 | 42.8 129 | 69.3 126 | 32.9 129 | 33.4 114 | 70.0 129 | 6.77 131 | 40.3 105 | 79.4 62 | 4.33 128 |
SLK [47] | 109.9 | 17.4 122 | 43.9 123 | 4.90 123 | 30.5 120 | 44.0 120 | 7.18 123 | 22.5 125 | 39.8 126 | 4.15 126 | 64.5 125 | 70.5 114 | 46.7 121 | 78.9 3 | 87.7 96 | 41.6 1 | 38.5 112 | 68.8 119 | 26.0 100 | 33.8 121 | 70.1 130 | 5.50 122 | 41.6 124 | 81.4 122 | 3.91 120 |
SILK [79] | 110.9 | 16.3 117 | 42.0 116 | 4.01 120 | 29.9 117 | 43.5 117 | 6.44 119 | 21.6 122 | 37.4 118 | 3.55 123 | 62.6 118 | 69.4 98 | 47.0 122 | 79.3 54 | 87.7 96 | 43.6 86 | 39.9 126 | 68.1 106 | 29.2 127 | 32.8 102 | 67.8 114 | 5.14 106 | 40.6 115 | 80.6 114 | 3.52 108 |
NL-TV-NCC [25] | 111.1 | 15.1 103 | 41.6 115 | 1.86 42 | 26.6 81 | 41.3 105 | 3.02 59 | 20.8 116 | 35.7 107 | 2.24 95 | 63.2 123 | 73.9 127 | 45.9 118 | 81.3 131 | 88.7 129 | 49.9 131 | 38.6 113 | 69.8 129 | 25.6 80 | 37.6 131 | 69.5 126 | 5.62 125 | 42.4 129 | 82.1 128 | 4.00 124 |
HCIC-L [99] | 112.2 | 23.2 131 | 49.0 130 | 11.0 131 | 32.1 125 | 44.4 121 | 9.93 128 | 23.2 127 | 36.4 116 | 3.02 117 | 64.4 124 | 72.1 124 | 44.9 111 | 80.6 127 | 88.5 127 | 46.6 126 | 39.1 121 | 68.9 121 | 27.1 117 | 32.4 87 | 65.0 65 | 5.53 123 | 39.1 26 | 79.3 54 | 3.65 115 |
Heeger++ [104] | 113.4 | 17.5 123 | 47.2 129 | 2.80 105 | 31.1 123 | 44.9 124 | 4.93 108 | 26.6 129 | 47.7 130 | 4.79 128 | 62.6 118 | 68.0 36 | 45.1 112 | 79.8 119 | 88.4 126 | 44.1 114 | 39.1 121 | 68.9 121 | 26.5 110 | 34.8 130 | 67.9 117 | 5.23 110 | 41.5 123 | 80.1 98 | 3.23 67 |
FFV1MT [106] | 115.5 | 16.4 118 | 44.7 125 | 3.13 113 | 31.9 124 | 44.7 122 | 7.15 122 | 25.4 128 | 45.6 129 | 5.04 129 | 62.6 118 | 68.0 36 | 45.1 112 | 79.6 112 | 87.9 110 | 44.1 114 | 38.9 116 | 67.7 91 | 27.1 117 | 34.0 127 | 68.5 120 | 5.29 113 | 41.8 127 | 81.0 119 | 4.48 130 |
Learning Flow [11] | 115.8 | 15.3 107 | 42.7 118 | 2.55 97 | 28.0 106 | 42.7 112 | 3.95 91 | 21.1 118 | 37.0 117 | 3.03 119 | 63.0 121 | 73.3 126 | 46.2 119 | 80.0 120 | 88.2 124 | 45.1 121 | 38.2 107 | 68.6 116 | 26.7 112 | 33.8 121 | 68.5 120 | 5.21 108 | 41.9 128 | 82.3 129 | 3.95 121 |
Adaptive flow [45] | 118.8 | 19.6 126 | 44.1 124 | 6.76 126 | 32.8 127 | 45.7 127 | 10.2 129 | 19.8 106 | 34.4 91 | 3.02 117 | 64.7 127 | 72.1 124 | 49.4 127 | 80.3 125 | 88.6 128 | 45.6 122 | 38.3 110 | 69.2 124 | 26.7 112 | 32.6 96 | 66.5 107 | 5.45 121 | 41.0 118 | 81.1 120 | 3.75 118 |
Pyramid LK [2] | 120.5 | 21.2 129 | 43.7 122 | 10.7 130 | 33.1 129 | 45.1 125 | 11.9 130 | 27.3 130 | 36.0 110 | 6.46 130 | 70.7 131 | 78.5 131 | 57.7 131 | 79.5 109 | 88.1 119 | 43.3 47 | 38.6 113 | 68.8 119 | 27.0 116 | 33.5 119 | 68.8 124 | 6.00 128 | 41.0 118 | 81.8 125 | 4.31 127 |
FOLKI [16] | 123.8 | 20.9 127 | 46.0 126 | 9.48 129 | 32.8 127 | 47.4 129 | 8.75 126 | 21.6 122 | 40.7 127 | 4.10 125 | 67.2 130 | 74.2 129 | 53.7 130 | 79.5 109 | 88.1 119 | 43.7 100 | 39.2 124 | 69.2 124 | 27.9 121 | 33.4 114 | 69.5 126 | 5.65 126 | 41.7 126 | 82.3 129 | 4.28 126 |
PGAM+LK [55] | 124.0 | 19.4 125 | 46.4 127 | 6.81 127 | 30.9 122 | 44.8 123 | 7.52 124 | 22.7 126 | 40.9 128 | 3.99 124 | 66.6 129 | 73.9 127 | 52.4 129 | 79.7 115 | 88.1 119 | 44.5 117 | 40.2 127 | 69.7 128 | 28.8 125 | 33.3 112 | 69.3 125 | 5.42 120 | 41.4 122 | 81.8 125 | 4.36 129 |
Periodicity [78] | 129.3 | 21.0 128 | 47.0 128 | 9.32 128 | 38.1 131 | 48.1 130 | 14.7 131 | 29.8 131 | 47.9 131 | 9.27 131 | 66.0 128 | 77.1 130 | 50.7 128 | 80.8 130 | 89.3 131 | 46.8 127 | 45.1 131 | 70.6 130 | 35.5 131 | 33.5 119 | 69.6 128 | 6.07 129 | 43.5 131 | 84.0 131 | 6.51 131 |
Method | time* | frames | color | Reference and notes | |
[1] 2D-CLG | 844 | 2 | gray | The 2D-CLG method by Bruhn et al. as implemented by Stefan Roth. [A. Bruhn, J. Weickert, and C. Schnörr. Lucas/Kanade meets Horn/Schunck: combining local and global optic flow methods. IJCV 63(3), 2005.] Parameters were set to match the published performance on Yosemite sequence, which may not be optimal for other sequences. | |
[2] Pyramid LK | 12 | 2 | color | A modification of Bouguet's pyramidal implementation of Lucas-Kanade. | |
[3] Horn & Schunck | 49 | 2 | gray | A modern Matlab implementation of the Horn & Schunck method by Deqing Sun. Parameters set to optimize AAE on all training data. | |
[4] Black & Anandan | 328 | 2 | gray | A modern Matlab implementation of the Black & Anandan method by Deqing Sun. | |
[5] Brox et al. | 18 | 2 | color | T. Brox, A. Bruhn, N. Papenberg, and J. Weickert. High accuracy optical flow estimation based on a theory for warping. ECCV 2004. (Improved using separate robust functions as proposed in A. Bruhn and J. Weickert, Towards ultimate motion estimation, ICCV 2005; improved by training on the training set.) | |
[6] Fusion | 2,666 | 2 | color | V. Lempitsky, S. Roth, and C. Rother. Discrete-continuous optimization for optical flow estimation. CVPR 2008. | |
[7] Dynamic MRF | 366 | 2 | gray | B. Glocker, N. Paragios, N. Komodakis, G. Tziritas, and N. Navab. Optical flow estimation with uncertainties through dynamic MRFs. CVPR 2008. (Method improved since publication.) | |
[8] Second-order prior | 14 | 2 | gray | W. Trobin, T. Pock, D. Cremers, and H. Bischof. An unbiased second-order prior for high-accuracy motion estimation. DAGM 2008. (Method improved since publication; for details see W. Trobin, Ph.D. thesis, 2009.) | |
[9] GroupFlow | 600 | 2 | gray | X. Ren. Local Grouping for Optical Flow. CVPR 2008. | |
[10] SegOF | 60 | 2 | color | L. Xu, J. Chen, and J. Jia. Segmentation based variational model for accurate optical flow estimation. ECCV 2008. Code available. | |
[11] Learning Flow | 825 | 2 | gray | D. Sun, S. Roth, J.P. Lewis, and M. Black. Learning optical flow (SRF-LFC). ECCV 2008. | |
[12] CBF | 69 | 2 | color | W. Trobin, T. Pock, D. Cremers, and H. Bischof. Continuous energy minimization via repeated binary fusion. ECCV 2008. (Method improved since publication; for details see W. Trobin, Ph.D. thesis, 2009.) | |
[13] SPSA-learn | 200 | 2 | color | Y. Li and D. Huttenlocher. Learning for optical flow using stochastic optimization. ECCV 2008. | |
[14] GraphCuts | 1,200 | 2 | color | T. Cooke. Two applications of graph-cuts to image processing. DICTA 2008. | |
[15] F-TV-L1 | 8 | 2 | gray | A. Wedel, T. Pock, J. Braun, U. Franke, and D. Cremers. Duality TV-L1 flow with fundamental matrix prior. IVCNZ 2008. | |
[16] FOLKI | 1.4 | 2 | gray | G. Le Besnerais and F. Champagnat. Dense optical flow by iterative local window registration. ICIP 2005. | |
[17] TV-L1-improved | 2.9 | 2 | gray | A. Wedel, T. Pock, C. Zach, H. Bischof, and D. Cremers. An improved algorithm for TV-L1 optical flow computation. Proceedings of the Dagstuhl Visual Motion Analysis Workshop 2008. Code at GPU4Vision. | |
[18] DPOF | 287 | 2 | color | C. Lei and Y.-H. Yang. Optical flow estimation on coarse-to-fine region-trees using discrete optimization. ICCV 2009. (Method improved since publication.) | |
[19] Filter Flow | 34,000 | 2 | color | S. Seitz and S. Baker. Filter flow. ICCV 2009. | |
[20] Adaptive | 9.2 | 2 | gray | A. Wedel, D. Cremers, T. Pock, and H. Bischof. Structure- and motion-adaptive regularization for high accuracy optic flow. ICCV 2009. | |
[21] Complementary OF | 44 | 2 | color | H. Zimmer, A. Bruhn, J. Weickert, L. Valgaerts, A. Salgado, B. Rosenhahn, and H.-P. Seidel. Complementary optic flow. EMMCVPR 2009. | |
[22] Aniso. Huber-L1 | 2 | 2 | gray | M. Werlberger, W. Trobin, T. Pock, A. Wedel, D. Cremers, and H. Bischof. Anisotropic Huber-L1 optical flow. BMVC 2009. Code at GPU4Vision. | |
[23] Rannacher | 0.12 | 2 | gray | J. Rannacher. Realtime 3D motion estimation on graphics hardware. Bachelor thesis, Heidelberg University, 2009. | |
[24] TI-DOFE | 260 | 2 | gray | C. Cassisa, S. Simoens, and V. Prinet. Two-frame optical flow formulation in an unwarped multiresolution scheme. CIARP 2009. | |
[25] NL-TV-NCC | 20 | 2 | color | M. Werlberger, T. Pock, and H. Bischof. Motion estimation with non-local total variation regularization. CVPR 2010. | |
[26] MDP-Flow | 188 | 2 | color | L. Xu, J. Jia, and Y. Matsushita. Motion detail preserving optical flow estimation. CVPR 2010. | |
[27] ACK-Prior | 5872 | 2 | color | K. Lee, D. Kwon, I. Yun, and S. Lee. Optical flow estimation with adaptive convolution kernel prior on discrete framework. CVPR 2010. | |
[28] LDOF | 122 | 2 | color | T. Brox and J. Malik. Large displacement optical flow: descriptor matching in variational motion estimation. PAMI 33(3):500-513, 2011. | |
[29] p-harmonic | 565 | 2 | gray | J. Gai and R. Stevenson. Optical flow estimation with p-harmonic regularization. ICIP 2010. | |
[30] TriangleFlow | 4200 | 2 | gray | B. Glocker, H. Heibel, N. Navab, P. Kohli, and C. Rother. TriangleFlow: Optical flow with triangulation-based higher-order likelihoods. ECCV 2010. | |
[31] Classic+NL | 972 | 2 | color | D. Sun, S. Roth, and M. Black. Secrets of optical flow estimation and their principles. CVPR 2010. Matlab code. | |
[32] Classic++ | 486 | 2 | gray | A modern implementation of the classical formulation descended from Horn & Schunck and Black & Anandan; see D. Sun, S. Roth, and M. Black, Secrets of optical flow estimation and their principles, CVPR 2010. | |
[33] Nguyen | 33 | 2 | gray | D. Nguyen. Tuning optical flow estimation with image-driven functions. ICRA 2011. | |
[34] Modified CLG | 133 | 2 | gray | R. Fezzani, F. Champagnat, and G. Le Besnerais. Combined local global method for optic flow computation. EUSIPCO 2010. | |
[35] ComplOF-FED-GPU | 0.97 | 2 | color | P. Gwosdek, H. Zimmer, S. Grewenig, A. Bruhn, and J. Weickert. A highly efficient GPU implementation for variational optic flow based on the Euler-Lagrange framework. CVGPU Workshop 2010. | |
[36] Ad-TV-NDC | 35 | 2 | gray | M. Nawaz. Motion estimation with adaptive regularization and neighborhood dependent constraint. DICTA 2010. | |
[37] Layers++ | 18206 | 2 | color | D. Sun, E. Sudderth, and M. Black. Layered image motion with explicit occlusions, temporal consistency, and depth ordering. NIPS 2010. | |
[38] OFH | 620 | 3 | color | H. Zimmer, A. Bruhn, J. Weickert. Optic flow in harmony. IJCV 93(3) 2011. | |
[39] LSM | 1615 | 2 | color | K. Jia, X. Wang, and X. Tang. Optical flow estimation using learned sparse model. ICCV 2011. | |
[40] CostFilter | 55 | 2 | color | C. Rhemann, A. Hosni, M. Bleyer, C. Rother, and M. Gelautz. Fast cost-volume filtering for visual correspondence and beyond. CVPR 2011. | |
[41] Bartels | 0.15 | 2 | gray | C. Bartels and G. de Haan. Smoothness constraints in recursive search motion estimation for picture rate conversion. IEEE TCSVT 2010. Version improved since publication: mapped on GPU. | |
[42] Shiralkar | 600 | 2 | gray | M. Shiralkar and R. Schalkoff. A self organization-based optical flow estimator with GPU implementation. MVA 23(6):1229-1242. | |
[43] HBpMotionGpu | 1000 | 5 | gray | S. Grauer-Gray and C. Kambhamettu. Hierarchical belief propagation to reduce search space using CUDA for stereo and motion estimation. WACV 2009. (Method improved since publication.) | |
[44] StereoFlow | 7200 | 2 | color | G. Rosman, S. Shem-Tov, D. Bitton, T. Nir, G. Adiv, R. Kimmel, A. Feuer, and A. Bruckstein. Over-parameterized optical flow using a stereoscopic constraint. SSVM 2011:761-772. | |
[45] Adaptive flow | 121 | 2 | gray | T. Arici. Energy minimization based motion estimation using adaptive smoothness priors. Submitted to IEEE TIP 2011. | |
[46] TC-Flow | 2500 | 5 | color | S. Volz, A. Bruhn, L. Valgaerts, and H. Zimmer. Modeling temporal coherence for optical flow. ICCV 2011. | |
[47] SLK | 300 | 2 | gray | T. Corpetti and E. Mémin. Stochastic uncertainty models for the luminance consistency assumption. IEEE TIP 2011. | |
[48] CLG-TV | 29 | 2 | gray | M. Drulea. Total variation regularization of local-global optical flow. ITSC 2011. Matlab code. | |
[49] SimpleFlow | 1.7 | 2 | color | M. Tao, J. Bai, P. Kohli, S. Paris. SimpleFlow: a non-iterative, sublinear optical flow algorithm. EUROGRAPHICS 2012. | |
[50] IAOF | 57 | 2 | gray | D. Nguyen. Improving motion estimation using image-driven functions and hybrid scheme. PSIVT 2011. | |
[51] IAOF2 | 56 | 2 | gray | D. Nguyen. Enhancing the sharpness of flow field using image-driven functions with occlusion-aware filter. Submitted to TIP 2011. | |
[52] LocallyOriented | 9541 | 2 | gray | Y.Niu, A. Dick, and M. Brooks. Locally oriented optical flow computation. To appear in TIP 2012. | |
[53] IROF-TV | 261 | 2 | color | H. Rashwan, D. Puig, and M. Garcia. On improving the robustness of differential optical flow. ICCV 2011 Artemis workshop. | |
[54] Sparse Occlusion | 2312 | 2 | color | A. Ayvaci, M. Raptis, and S. Soatto. Sparse occlusion detection with optical flow. Submitted to IJCV 2011. | |
[55] PGAM+LK | 0.37 | 2 | gray | A. Alba, E. Arce-Santana, and M. Rivera. Optical flow estimation with prior models obtained from phase correlation. ISVC 2010. | |
[56] Sparse-NonSparse | 713 | 2 | color | L. Chen, J. Wang, and Y. Wu. Decomposing and regularizing sparse/non-sparse components for motion field estimation. Submitted to PAMI 2013. | |
[57] nLayers | 36150 | 4 | color | D. Sun, E. Sudderth, and M. Black. Layered segmentation and optical flow estimation over time. CVPR 2012. | |
[58] IROF++ | 187 | 2 | color | H. Rashwan, D. Puig, and M. Garcia. Variational optical flow estimation based on stick tensor voting. IEEE TIP 2013. | |
[59] COFM | 600 | 3 | color | M. Mozerov. Constrained optical flow estimation as a matching problem. IEEE TIP 2013. | |
[60] Efficient-NL | 400 | 2 | color | P. Krähenbühl and V. Koltun. Efficient nonlocal regularization for optical flow. ECCV 2012. | |
[61] BlockOverlap | 2 | 2 | gray | M. Santoro, G. AlRegib, and Y. Altunbasak. Motion estimation using block overlap minimization. Submitted to MMSP 2012. | |
[62] Ramp | 1200 | 2 | color | A. Singh and N. Ahuja. Exploiting ramp structures for improving optical flow estimation. ICPR 2012. | |
[63] Occlusion-TV-L1 | 538 | 3 | gray | C. Ballester, L. Garrido, V. Lazcano, and V. Caselles. A TV-L1 optical flow method with occlusion detection. DAGM-OAGM 2012. | |
[64] TV-L1-MCT | 90 | 2 | color | M. Mohamed and B. Mertsching. TV-L1 optical flow estimation with image details recovering based on modified census transform. ISVC 2012. | |
[65] Local-TV-L1 | 500 | 2 | gray | L. Raket. Local smoothness for global optical flow. ICIP 2012. | |
[66] ALD-Flow | 61 | 2 | color | M. Stoll, A. Bruhn, and S. Volz. Adaptive integration of feature matches into variational optic flow methods. ACCV 2012. | |
[67] SIOF | 234 | 2 | color | L. Xu, Z. Dai, and J. Jia. Scale invariant optical flow. ECCV 2012. | |
[68] MDP-Flow2 | 342 | 2 | color | L. Xu, J. Jia, and Y. Matsushita. Motion detail preserving optical flow estimation. PAMI 34(9):1744-1757, 2012. Code available. | |
[69] TCOF | 1421 | all | gray | J. Sanchez, A. Salgado, and N. Monzon. Optical flow estimation with consistent spatio-temporal coherence models. VISAPP 2013. | |
[70] LME | 476 | 2 | color | W. Li, D. Cosker, M. Brown, and R. Tang. Optical flow estimation using Laplacian mesh energy. CVPR 2013. | |
[71] NN-field | 362 | 2 | color | L. Chen, H. Jin, Z. Lin, S. Cohen, and Y. Wu. Large displacement optical flow from nearest neighbor fields. CVPR 2013. | |
[72] FESL | 3310 | 2 | color | W. Dong, G. Shi, X. Hu, and Y. Ma. Nonlocal sparse and low-rank regularization for optical flow estimation. Submitted to IEEE TIP 2013. | |
[73] PMF | 35 | 2 | color | J. Lu, H. Yang, D. Min, and M. Do. PatchMatch filter: efficient edge-aware filtering meets randomized search for fast correspondence field estimation. CVPR 2013. | |
[74] FC-2Layers-FF | 2662 | 4 | color | D. Sun, J. Wulff, E. Sudderth, H. Pfister, and M. Black. A fully-connected layered model of foreground and background flow. CVPR 2013. | |
[75] Correlation Flow | 290 | 2 | color | M. Drulea and S. Nedevschi. Motion estimation using the correlation transform. TIP 2013. Matlab code. | |
[76] TC/T-Flow | 341 | 5 | color | M. Stoll, S. Volz, and A. Bruhn. Joint trilateral filtering for multiframe optical flow. ICIP 2013. | |
[77] OFLAF | 1530 | 2 | color | T. Kim, H. Lee, and K. Lee. Optical flow via locally adaptive fusion of complementary data costs. ICCV 2013. | |
[78] Periodicity | 8000 | 4 | color | G. Khachaturov, S. Gonzalez-Brambila, and J. Gonzalez-Trejo. Periodicity-based computation of optical flow. Submitted to Computacion y Sistemas (CyS) 2013. | |
[79] SILK | 572 | 2 | gray | P. Zille, C. Xu, T. Corpetti, L. Shao. Observation models based on scale interactions for optical flow estimation. Submitted to IEEE TIP. | |
[80] CRTflow | 13 | 3 | color | O. Demetz, D. Hafner, and J. Weickert. The complete rank transform: a tool for accurate and morphologically invariant matching of structures. BMVC 2013. | |
[81] SuperFlow | 178 | 2 | color | Anonymous. Superpixel based optical flow estimation. ICCV 2013 submission 507. | |
[82] Aniso-Texture | 300 | 2 | color | Anonymous. Texture information-based optical flow estimation using an incremental multi-resolution approach. ITC-CSCC 2013 submission 267. | |
[83] Classic+CPF | 640 | 2 | gray | Z. Tu, R. Veltkamp, and N. van der Aa. A combined post-filtering method to improve accuracy of variational optical flow estimation. Submitted to Pattern Recognition 2013. | |
[84] S2D-Matching | 1200 | 2 | color | Anonymous. Locally affine sparse-to-dense matching for motion and occlusion estimation. ICCV 2013 submission 1479. | |
[85] AGIF+OF | 438 | 2 | gray | Z. Tu, R. Poppe, and R. Veltkamp. Adaptive guided image filter to warped interpolation image for variational optical flow computation. Submitted to Signal Processing 2015. | |
[86] DeepFlow | 13 | 2 | color | P. Weinzaepfel, J. Revaud, Z. Harchaoui, and C. Schmid. DeepFlow: large displacement optical flow with deep matching. ICCV 2013. | |
[87] NNF-Local | 673 | 2 | color | Z. Chen, H. Jin, Z. Lin, S. Cohen, and Y. Wu. Large displacement optical flow with nearest neighbor field. Submitted to PAMI 2014. | |
[88] EPPM w/o HM | 2.5 | 2 | color | L. Bao, Q. Yang, and H. Jin. Fast edge-preserving PatchMatch for large displacement optical flow. CVPR 2014. | |
[89] MLDP_OF | 165 | 2 | gray | M. Mohamed, H. Rashwan, B. Mertsching, M. Garcia, and D. Puig. Illumination-robust optical flow approach using local directional pattern. IEEE TCSVT 24(9):1499-1508, 2014. | |
[90] RFlow | 20 | 2 | gray | S. Ali, C. Daul, and W. Blondel. Robust and accurate optical flow estimation for weak texture and varying illumination condition: Application to cystoscopy. IPTA 2014. | |
[91] SRR-TVOF-NL | 32 | all | color | P. Pohl, M. Sirotenko, E. Tolstaya, and V. Bucha. Edge preserving motion estimation with occlusions correction for assisted 2D to 3D conversion. IS&T/SPIE Electronic Imaging 2014. | |
[92] 2DHMM-SAS | 157 | 2 | color | M.-C. Shih, R. Shenoy, and K. Rose. A two-dimensional hidden Markov model with spatially-adaptive states with application of optical flow. ICIP 2014 submission. | |
[93] WLIF-Flow | 700 | 2 | color | Z. Tu, R. Veltkamp, N. van der Aa, and C. Van Gemeren. Weighted local intensity fusion method for variational optical flow estimation. Submitted to TIP 2014. | |
[94] FMOF | 215 | 2 | color | N. Jith, A. Ramakanth, and V. Babu. Optical flow estimation using approximate nearest neighbor field fusion. ICASSP 2014. | |
[95] TriFlow | 150 | 2 | color | TriFlow. Optical flow with geometric occlusion estimation and fusion of multiple frames. ECCV 2014 submission 914. | |
[96] ComponentFusion | 6.5 | 2 | color | Anonymous. Fast optical flow by component fusion. ECCV 2014 submission 941. | |
[97] AggregFlow | 1642 | 2 | color | D. Fortun, P. Bouthemy, and C. Kervrann. Aggregation of local parametric candidates and exemplar-based occlusion handling for optical flow. Preprint arXiv:1407.5759. | |
[98] 2bit-BM-tele | 124 | 2 | gray | R. Xu and D. Taubman. Robust dense block-based motion estimation using a two-bit transform on a Laplacian pyramid. ICIP 2013. | |
[99] HCIC-L | 330 | 2 | color | Anonymous. Globally-optimal image correspondence using a hierarchical graphical model. NIPS 2014 submission 114. | |
[100] TF+OM | 600 | 2 | color | R. Kennedy and C. Taylor. Optical flow with geometric occlusion estimation and fusion of multiple frames. EMMCVPR 2015. | |
[101] PH-Flow | 800 | 2 | color | J. Yang and H. Li. Dense, accurate optical flow estimation with piecewise parametric model. CVPR 2015. | |
[102] EpicFlow | 16 | 2 | color | J. Revaud, P. Weinzaepfel, Z. Harchaoui, and C. Schmid. EpicFlow: edge-preserving interpolation of correspondences for optical flow. CVPR 2015. | |
[103] NNF-EAC | 380 | 2 | color | Anonymous. Variational method for joint optical flow estimation and edge-aware image restoration. CVPR 2015 submission 2336. | |
[104] Heeger++ | 6600 | 5 | gray | Anonymous. A context aware biologically inspired algorithm for optical flow (updated results). CVPR 2015 submission 2238. | |
[105] HBM-GC | 330 | 2 | color | A. Zheng and Y. Yuan. Motion estimation via hierarchical block matching and graph cut. Submitted to ICIP 2015. | |
[106] FFV1MT | 358 | 5 | gray | F. Solari, M. Chessa, N. Medathati, and P. Kornprobst. What can we expect from a V1-MT feedforward architecture for optical flow estimation? Submitted to Signal Processing: Image Communication 2015. | |
[107] ROF-ND | 4 | 2 | color | S. Ali, C. Daul, E. Galbrun, and W. Blondel. Illumination invariant large displacement optical flow using robust neighbourhood descriptors. Submitted to CVIU 2015. | |
[108] DeepFlow2 | 16 | 2 | color | J. Revaud, P. Weinzaepfel, Z. Harchaoui, and C. Schmid. Deep convolutional matching. Submitted to IJCV, 2015. | |
[109] HAST | 2667 | 2 | color | Anonymous. Highly accurate optical flow estimation on superpixel tree. ICCV 2015 submission 2221. | |
[110] FlowFields | 15 | 2 | color | C. Bailer, B. Taetz, and D. Stricker. Flow Fields: Dense unregularized correspondence fields for highly accurate large displacement optical flow estimation. ICCV 2015. | |
[111] SVFilterOh | 1.56 | 2 | color | Anonymous. Fast estimation of large displacement optical flow using PatchMatch and dominant motion patterns. CVPR 2016 submission 1788. | |
[112] FlowNetS+ft+v | 0.5 | 2 | color | Anonymous. Learning optical flow with convolutional neural networks. ICCV 2015 submission 235. | |
[113] CombBMOF | 51 | 2 | color | M. Brüggemann, R. Kays, P. Springer, and O. Erdler. Combined block-matching and adaptive differential motion estimation in a hierarchical multi-scale framework. ICGIP 2014. (Method improved since publication.) | |
[114] PMMST | 182 | 2 | color | F. Zhang, S. Xu, and X. Zhang. High accuracy correspondence field estimation via MST based patch matching. Submitted to TIP 2015. | |
[115] DF-Auto | 70 | 2 | color | N. Monzon, A. Salgado, and J. Sanchez. Regularization strategies for discontinuity-preserving optical flow methods. Submitted to TIP 2015. | |
[116] CPM-Flow | 3 | 2 | color | Anonymous. Efficient coarse-to-fine PatchMatch for large displacement optical flow. CVPR 2016 submission 241. | |
[117] CNN-flow-warp+ref | 1.4 | 3 | color | D. Teney and M. Hebert. Learning to extract motion from videos in convolutional neural networks. ArXiv 1601.07532, 2016. | |
[118] Steered-L1 | 804 | 2 | color | Anonymous. Optical flow estimation via steered-L1 norm. Submitted to WSCG 2016. | |
[119] StereoOF-V1MT | 343 | 2 | gray | Anonymous. Visual features for action-oriented tasks: a cortical-like model for disparity and optic flow computation. BMVC 2016 submission 132. | |
[120] PGM-C | 5 | 2 | color | Y. Li. Pyramidal gradient matching for optical flow estimation. Submitted to PAMI 2016. | |
[121] RNLOD-Flow | 1040 | 2 | gray | C. Zhang, Z. Chen, M. Wang, M. Li, and S. Jiang. Robust non-local TV-L1 optical flow estimation with occlusion detection. Submitted to TIP 2016. | |
[122] FlowNet2 | 0.091 | 2 | color | Anonymous. FlowNet 2.0: Evolution of optical flow estimation with deep networks. CVPR 2017 submission 900. | |
[123] S2F-IF | 20 | 2 | color | Anonymous. S2F-IF: Slow-to-fast interpolator flow. CVPR 2017 submission 765. | |
[124] BriefMatch | 0.068 | 2 | gray | G. Eilertsen, P.-E. Forssen, and J. Unger. Dense binary feature matching for real-time optical flow estimation. SCIA 2017 submission 62. | |
[125] OAR-Flow | 60 | 2 | color | Anonymous. Order-adaptive regularisation for variational optical flow: global, local and in between. SSVM 2017 submission 20. | |
[126] AdaConv-v1 | 2.8 | 2 | color | S. Niklaus, L. Mai, and F. Liu. (Interpolation results only.) Video frame interpolation via adaptive convolution. CVPR 2017. | |
[127] SepConv-v1 | 0.2 | 2 | color | S. Niklaus, L. Mai, and F. Liu. (Interpolation results only.) Video frame interpolation via adaptive separable convolution. ICCV 2017. | |
[128] ProbFlowFields | 37 | 2 | color | A. Wannenwetsch, M. Keuper, and S. Roth. ProbFlow: joint optical flow and uncertainty estimation. ICCV 2017. | |
[129] UnFlow | 0.12 | 2 | color | Anonymous. UnFlow: Unsupervised learning of optical flow with a bidirectional census loss. Submitted to AAAI 2018. | |
[130] FlowFields+ | 10.5 | 2 | color | C. Bailer, B. Taetz, and D. Stricker. Flow fields: Dense correspondence fields for highly accurate large displacement optical flow estimation. Submitted to PAMI 2017. | |
[131] Kuang | 9.9 | 2 | gray | F. Kuang. PatchMatch algorithms for motion estimation and stereo reconstruction. Master thesis, University of Stuttgart, 2017. |