Optical flow evaluation results |
Statistics:
Average
SD
R0.5
R1.0
R2.0
A90
A95
A99
Error type: endpoint angle interpolation normalized interpolation |
R1.0 normalized interpolation error |
avg. |
Mequon (Hidden texture) im0 GT im1 |
Schefflera (Hidden texture) im0 GT im1 |
Urban (Synthetic) im0 GT im1 |
Teddy (Stereo) im0 GT im1 |
Backyard (High-speed camera) im0 GT im1 |
Basketball (High-speed camera) im0 GT im1 |
Dumptruck (High-speed camera) im0 GT im1 |
Evergreen (High-speed camera) im0 GT im1 | ||||||||||||||||
rank | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | |
MDP-Flow2 [68] | 11.5 | 6.71 2 | 9.78 2 | 8.39 8 | 6.36 9 | 11.5 11 | 6.23 12 | 7.12 4 | 9.73 7 | 5.42 2 | 21.2 5 | 18.8 8 | 41.2 11 | 30.1 5 | 26.5 3 | 45.6 48 | 27.7 18 | 19.5 26 | 36.0 18 | 6.48 2 | 13.8 4 | 10.3 6 | 8.19 23 | 16.4 17 | 7.11 25 |
PMMST [114] | 13.0 | 6.84 9 | 9.80 3 | 8.50 16 | 6.74 26 | 11.7 13 | 6.36 29 | 7.10 3 | 9.45 3 | 5.41 1 | 21.1 1 | 18.6 2 | 41.1 3 | 30.2 9 | 26.5 3 | 45.7 60 | 27.3 2 | 18.1 4 | 36.0 18 | 6.51 4 | 13.9 9 | 10.3 6 | 8.22 26 | 16.5 24 | 7.15 37 |
PH-Flow [101] | 14.5 | 7.05 22 | 10.9 17 | 8.50 16 | 6.10 3 | 10.6 5 | 6.14 3 | 7.18 5 | 9.83 8 | 5.55 6 | 21.1 1 | 18.5 1 | 41.1 3 | 30.1 5 | 26.6 8 | 45.2 14 | 27.9 35 | 21.7 86 | 35.7 9 | 6.60 15 | 14.3 19 | 10.3 6 | 8.11 14 | 16.3 14 | 7.14 33 |
NNF-Local [87] | 14.5 | 6.74 4 | 10.1 6 | 8.30 2 | 5.97 1 | 10.3 2 | 6.14 3 | 7.09 2 | 9.63 6 | 5.44 3 | 21.7 31 | 20.5 61 | 41.2 11 | 30.3 19 | 26.6 8 | 45.4 23 | 27.9 35 | 20.6 62 | 36.0 18 | 6.51 4 | 13.8 4 | 10.3 6 | 8.04 9 | 16.1 6 | 7.10 23 |
CombBMOF [113] | 15.8 | 6.93 13 | 9.87 5 | 8.43 10 | 6.33 8 | 11.4 9 | 6.22 11 | 7.60 44 | 10.3 14 | 6.25 75 | 21.5 20 | 19.4 22 | 41.2 11 | 30.2 9 | 26.6 8 | 45.3 19 | 27.7 18 | 19.2 18 | 36.1 31 | 6.57 11 | 14.1 12 | 10.3 6 | 7.67 1 | 15.4 3 | 6.82 1 |
NN-field [71] | 18.9 | 6.88 10 | 10.8 13 | 8.48 13 | 5.99 2 | 10.3 2 | 6.13 2 | 7.65 53 | 9.54 4 | 5.81 32 | 21.9 46 | 21.0 82 | 41.4 24 | 30.2 9 | 26.6 8 | 45.4 23 | 27.8 28 | 20.0 43 | 36.0 18 | 6.48 2 | 13.7 3 | 10.3 6 | 8.02 8 | 16.1 6 | 7.04 16 |
IROF++ [58] | 24.1 | 7.07 25 | 11.3 27 | 8.50 16 | 6.64 21 | 12.0 18 | 6.19 7 | 7.54 36 | 10.6 27 | 5.84 37 | 21.2 5 | 18.6 2 | 41.5 28 | 30.2 9 | 26.9 20 | 45.0 9 | 27.6 15 | 18.8 10 | 36.2 38 | 6.69 32 | 14.7 30 | 10.5 57 | 8.17 20 | 16.4 17 | 7.33 72 |
Layers++ [37] | 24.5 | 7.17 36 | 11.1 22 | 8.79 57 | 6.14 6 | 10.3 2 | 6.41 33 | 7.34 16 | 10.3 14 | 5.69 19 | 21.3 10 | 19.0 14 | 41.3 17 | 30.5 35 | 27.1 39 | 45.4 23 | 28.1 53 | 21.1 74 | 36.2 38 | 6.51 4 | 13.8 4 | 10.2 1 | 8.20 24 | 16.4 17 | 7.13 31 |
nLayers [57] | 26.4 | 7.15 34 | 10.4 10 | 8.81 62 | 6.44 12 | 11.4 9 | 6.42 34 | 7.23 8 | 9.30 1 | 5.65 14 | 21.4 15 | 19.1 17 | 41.5 28 | 30.7 71 | 27.4 80 | 45.6 48 | 28.0 44 | 20.8 66 | 36.2 38 | 6.54 9 | 13.6 1 | 10.3 6 | 8.07 10 | 16.3 14 | 6.91 3 |
Sparse-NonSparse [56] | 27.9 | 7.09 26 | 11.3 27 | 8.57 24 | 6.53 16 | 11.8 16 | 6.21 9 | 7.40 24 | 10.5 24 | 5.64 10 | 21.5 20 | 19.0 14 | 41.7 45 | 30.4 26 | 27.0 29 | 45.4 23 | 28.3 65 | 21.5 83 | 36.4 61 | 6.66 24 | 14.4 21 | 10.3 6 | 8.23 27 | 16.6 29 | 7.09 21 |
ProbFlowFields [128] | 28.5 | 7.03 19 | 11.8 49 | 8.66 38 | 6.41 10 | 11.7 13 | 6.31 22 | 7.18 5 | 10.3 14 | 5.58 7 | 21.7 31 | 19.6 29 | 41.8 53 | 30.7 71 | 27.1 39 | 46.0 104 | 27.9 35 | 20.5 54 | 36.2 38 | 6.51 4 | 13.8 4 | 10.3 6 | 7.80 3 | 15.6 4 | 7.14 33 |
2DHMM-SAS [92] | 28.6 | 7.27 48 | 12.0 56 | 8.59 28 | 7.82 63 | 14.2 53 | 6.36 29 | 7.25 9 | 10.5 24 | 5.74 25 | 21.4 15 | 18.7 5 | 41.4 24 | 30.3 19 | 26.9 20 | 45.2 14 | 27.9 35 | 20.3 47 | 36.0 18 | 6.64 22 | 14.4 21 | 10.3 6 | 8.41 45 | 17.0 43 | 7.08 18 |
NNF-EAC [103] | 29.6 | 7.35 56 | 11.1 22 | 8.79 57 | 6.92 36 | 12.5 30 | 6.29 20 | 7.52 35 | 10.1 11 | 5.76 28 | 21.8 38 | 19.5 27 | 42.8 101 | 30.2 9 | 26.6 8 | 45.4 23 | 27.5 7 | 18.9 12 | 36.0 18 | 6.58 13 | 14.2 15 | 10.4 32 | 8.32 34 | 16.8 33 | 7.19 46 |
AGIF+OF [85] | 29.7 | 7.11 28 | 11.3 27 | 8.46 11 | 6.68 25 | 12.2 22 | 6.27 16 | 7.38 21 | 10.1 11 | 5.71 23 | 21.2 5 | 18.6 2 | 41.1 3 | 30.8 83 | 27.6 100 | 45.4 23 | 28.3 65 | 22.2 104 | 36.0 18 | 6.67 25 | 14.0 11 | 10.3 6 | 8.34 37 | 17.0 43 | 6.91 3 |
FlowFields [110] | 29.9 | 7.02 18 | 11.5 35 | 8.54 22 | 6.66 24 | 12.5 30 | 6.44 38 | 7.45 28 | 11.5 52 | 5.64 10 | 21.9 46 | 20.5 61 | 41.8 53 | 30.6 51 | 27.0 29 | 45.5 38 | 27.7 18 | 20.2 45 | 36.0 18 | 6.59 14 | 14.3 19 | 10.4 32 | 8.00 7 | 16.2 11 | 7.08 18 |
FlowFields+ [130] | 30.6 | 6.99 15 | 11.4 33 | 8.47 12 | 6.61 20 | 12.3 24 | 6.48 41 | 7.42 26 | 11.5 52 | 5.67 17 | 21.7 31 | 20.3 49 | 41.6 35 | 30.7 71 | 27.2 54 | 45.6 48 | 27.8 28 | 20.3 47 | 36.1 31 | 6.61 16 | 14.4 21 | 10.4 32 | 7.99 6 | 16.2 11 | 7.03 14 |
FMOF [94] | 30.7 | 7.36 58 | 12.0 56 | 8.73 48 | 6.42 11 | 11.3 8 | 6.30 21 | 7.63 49 | 10.4 19 | 6.02 65 | 21.8 38 | 19.9 36 | 41.2 11 | 30.5 35 | 27.0 29 | 45.4 23 | 28.0 44 | 20.5 54 | 36.1 31 | 6.51 4 | 13.8 4 | 10.2 1 | 8.30 32 | 16.7 32 | 7.12 27 |
S2F-IF [123] | 31.4 | 7.01 17 | 11.5 35 | 8.48 13 | 6.57 17 | 12.2 22 | 6.42 34 | 7.40 24 | 11.2 48 | 5.64 10 | 21.6 24 | 20.1 42 | 41.3 17 | 30.7 71 | 27.3 68 | 45.7 60 | 27.8 28 | 20.4 51 | 36.1 31 | 6.71 37 | 14.9 45 | 10.4 32 | 7.98 5 | 16.1 6 | 7.04 16 |
LSM [39] | 32.8 | 7.17 36 | 11.8 49 | 8.58 26 | 6.64 21 | 12.1 21 | 6.17 6 | 7.49 31 | 10.9 39 | 5.69 19 | 21.6 24 | 19.6 29 | 41.6 35 | 30.5 35 | 27.1 39 | 45.4 23 | 28.3 65 | 21.6 84 | 36.3 52 | 6.68 30 | 14.6 26 | 10.2 1 | 8.35 40 | 16.9 41 | 7.03 14 |
LME [70] | 33.4 | 6.72 3 | 9.86 4 | 8.36 5 | 6.97 38 | 12.4 27 | 7.40 85 | 7.51 33 | 11.8 60 | 5.70 22 | 21.3 10 | 19.2 20 | 41.3 17 | 31.0 104 | 27.6 100 | 46.6 116 | 27.8 28 | 20.5 54 | 36.0 18 | 6.45 1 | 13.6 1 | 10.2 1 | 8.08 13 | 16.3 14 | 7.12 27 |
TV-L1-MCT [64] | 34.5 | 7.50 77 | 12.5 84 | 8.79 57 | 7.19 44 | 13.4 42 | 6.37 31 | 7.28 11 | 10.6 27 | 5.80 31 | 21.4 15 | 18.8 8 | 41.3 17 | 30.5 35 | 27.1 39 | 45.1 12 | 27.9 35 | 18.6 7 | 36.6 74 | 6.72 41 | 15.0 52 | 10.4 32 | 7.92 4 | 15.9 5 | 7.20 49 |
WLIF-Flow [93] | 34.6 | 6.99 15 | 11.0 19 | 8.48 13 | 6.76 27 | 12.4 27 | 6.39 32 | 7.38 21 | 10.3 14 | 5.68 18 | 21.4 15 | 18.8 8 | 41.9 66 | 30.4 26 | 26.9 20 | 45.9 94 | 28.8 98 | 21.9 93 | 36.9 93 | 6.56 10 | 13.9 9 | 10.3 6 | 8.34 37 | 16.8 33 | 7.15 37 |
ComponentFusion [96] | 34.7 | 6.91 11 | 10.8 13 | 8.55 23 | 6.49 15 | 12.0 18 | 6.10 1 | 7.49 31 | 11.2 48 | 5.72 24 | 21.3 10 | 19.4 22 | 41.2 11 | 30.6 51 | 27.2 54 | 45.8 76 | 27.8 28 | 19.6 29 | 36.2 38 | 6.92 74 | 16.4 92 | 10.4 32 | 8.43 49 | 17.0 43 | 7.16 40 |
COFM [59] | 35.8 | 7.04 20 | 10.7 12 | 8.70 42 | 6.60 18 | 11.9 17 | 6.35 26 | 7.26 10 | 9.93 9 | 5.63 9 | 21.2 5 | 18.8 8 | 41.0 1 | 30.4 26 | 27.3 68 | 44.9 8 | 27.7 18 | 22.6 107 | 35.1 2 | 6.86 67 | 14.7 30 | 11.2 111 | 8.67 76 | 17.2 60 | 7.78 108 |
MDP-Flow [26] | 36.5 | 6.83 7 | 10.8 13 | 8.50 16 | 6.65 23 | 12.4 27 | 6.51 46 | 7.46 29 | 10.6 27 | 5.88 45 | 22.1 65 | 20.6 66 | 41.7 45 | 30.4 26 | 26.8 15 | 45.6 48 | 28.2 62 | 21.9 93 | 36.3 52 | 6.69 32 | 14.8 39 | 10.4 32 | 8.15 16 | 16.6 29 | 7.10 23 |
HAST [109] | 37.7 | 6.97 14 | 10.2 7 | 8.69 40 | 6.46 13 | 11.6 12 | 6.26 15 | 7.72 61 | 11.1 44 | 5.97 58 | 21.1 1 | 18.7 5 | 41.1 3 | 30.5 35 | 27.5 87 | 44.8 6 | 28.2 62 | 22.8 113 | 35.5 5 | 6.76 50 | 15.2 60 | 10.4 32 | 8.81 82 | 18.0 91 | 6.96 8 |
RNLOD-Flow [121] | 37.8 | 7.12 30 | 11.5 35 | 8.64 35 | 7.38 50 | 14.0 50 | 6.35 26 | 7.55 37 | 11.2 48 | 5.83 36 | 21.3 10 | 19.0 14 | 41.1 3 | 30.5 35 | 27.2 54 | 45.4 23 | 28.3 65 | 21.6 84 | 36.2 38 | 6.62 17 | 14.2 15 | 10.4 32 | 8.70 79 | 17.7 80 | 7.02 12 |
OFLAF [77] | 38.3 | 6.81 6 | 10.2 7 | 8.40 9 | 6.10 3 | 10.7 6 | 6.21 9 | 7.36 18 | 10.6 27 | 5.54 5 | 21.1 1 | 18.8 8 | 41.0 1 | 30.8 83 | 27.4 80 | 45.7 60 | 28.1 53 | 21.9 93 | 36.0 18 | 7.02 85 | 16.1 89 | 10.4 32 | 8.90 89 | 18.1 98 | 7.16 40 |
Ramp [62] | 40.5 | 7.31 54 | 12.1 60 | 8.78 54 | 6.60 18 | 12.0 18 | 6.27 16 | 7.36 18 | 10.4 19 | 5.65 14 | 21.3 10 | 18.9 13 | 41.4 24 | 30.5 35 | 27.1 39 | 45.4 23 | 28.7 93 | 22.3 105 | 36.6 74 | 6.73 46 | 14.9 45 | 10.3 6 | 8.55 63 | 17.3 65 | 7.26 60 |
Second-order prior [8] | 41.2 | 7.30 53 | 11.3 27 | 8.90 71 | 8.52 81 | 15.6 80 | 6.74 60 | 8.32 94 | 13.6 103 | 6.42 88 | 21.8 38 | 20.0 39 | 41.5 28 | 30.1 5 | 26.5 3 | 45.5 38 | 27.5 7 | 19.0 13 | 36.0 18 | 6.67 25 | 14.6 26 | 10.3 6 | 8.25 28 | 16.8 33 | 7.11 25 |
PGM-C [120] | 42.6 | 7.19 41 | 12.1 60 | 8.72 45 | 6.82 29 | 12.9 34 | 6.62 52 | 7.67 57 | 12.2 72 | 5.78 30 | 21.9 46 | 20.9 76 | 41.8 53 | 30.6 51 | 27.1 39 | 45.8 76 | 27.7 18 | 19.5 26 | 36.2 38 | 6.65 23 | 14.7 30 | 10.3 6 | 8.20 24 | 16.6 29 | 7.31 67 |
Kuang [131] | 43.0 | 7.12 30 | 12.1 60 | 8.58 26 | 7.14 41 | 13.5 44 | 6.50 44 | 7.77 67 | 12.1 68 | 5.94 53 | 21.9 46 | 20.7 69 | 41.7 45 | 30.6 51 | 27.1 39 | 45.6 48 | 27.5 7 | 19.6 29 | 35.9 13 | 6.84 60 | 15.6 74 | 10.5 57 | 8.07 10 | 16.4 17 | 7.14 33 |
DeepFlow2 [108] | 43.3 | 7.28 49 | 11.3 27 | 8.88 68 | 7.68 59 | 14.4 60 | 6.94 73 | 7.58 42 | 12.3 76 | 5.88 45 | 21.9 46 | 20.2 46 | 41.7 45 | 30.5 35 | 26.8 15 | 45.9 94 | 27.5 7 | 18.0 3 | 36.4 61 | 6.67 25 | 14.6 26 | 10.4 32 | 8.18 21 | 16.4 17 | 7.31 67 |
Aniso. Huber-L1 [22] | 43.5 | 7.61 84 | 12.2 69 | 9.19 83 | 8.99 91 | 15.7 83 | 7.12 79 | 7.73 63 | 11.0 42 | 5.86 42 | 21.8 38 | 20.0 39 | 41.6 35 | 30.2 9 | 26.6 8 | 45.5 38 | 27.4 4 | 19.6 29 | 35.7 9 | 6.68 30 | 14.6 26 | 10.3 6 | 8.34 37 | 16.8 33 | 7.29 66 |
Classic+NL [31] | 43.5 | 7.44 71 | 12.3 72 | 8.86 65 | 6.78 28 | 12.3 24 | 6.28 18 | 7.32 15 | 10.4 19 | 5.69 19 | 21.6 24 | 19.4 22 | 41.8 53 | 30.5 35 | 27.1 39 | 45.5 38 | 28.6 88 | 21.8 89 | 36.6 74 | 6.72 41 | 14.7 30 | 10.3 6 | 8.50 57 | 17.2 60 | 7.24 57 |
FC-2Layers-FF [74] | 43.7 | 7.22 44 | 11.9 54 | 8.70 42 | 6.10 3 | 10.2 1 | 6.47 40 | 7.31 14 | 10.5 24 | 5.64 10 | 21.4 15 | 19.1 17 | 41.6 35 | 30.7 71 | 27.5 87 | 45.6 48 | 28.6 88 | 22.7 110 | 36.4 61 | 6.77 52 | 15.0 52 | 10.3 6 | 8.57 65 | 17.2 60 | 7.20 49 |
SRR-TVOF-NL [91] | 44.5 | 7.42 69 | 11.5 35 | 8.86 65 | 7.79 61 | 14.8 68 | 7.08 77 | 7.62 46 | 11.5 52 | 5.85 40 | 21.7 31 | 19.6 29 | 41.1 3 | 30.3 19 | 27.1 39 | 45.2 14 | 27.5 7 | 20.5 54 | 35.3 4 | 6.72 41 | 14.8 39 | 10.4 32 | 8.97 96 | 18.3 104 | 7.17 43 |
DF-Auto [115] | 45.7 | 7.54 79 | 11.1 22 | 9.32 91 | 8.42 77 | 14.5 63 | 8.82 96 | 7.35 17 | 10.3 14 | 5.65 14 | 22.0 57 | 20.2 46 | 41.5 28 | 30.4 26 | 26.7 14 | 45.8 76 | 27.5 7 | 18.7 9 | 36.1 31 | 6.82 59 | 15.3 64 | 10.5 57 | 8.43 49 | 17.1 51 | 7.20 49 |
FESL [72] | 46.0 | 7.36 58 | 11.7 44 | 8.65 37 | 6.82 29 | 12.6 32 | 6.33 24 | 7.51 33 | 10.7 34 | 5.89 48 | 21.6 24 | 19.6 29 | 41.3 17 | 30.9 100 | 27.5 87 | 45.7 60 | 28.4 77 | 22.1 102 | 36.2 38 | 6.70 35 | 14.8 39 | 10.2 1 | 8.59 66 | 17.4 71 | 7.08 18 |
CPM-Flow [116] | 46.2 | 7.21 42 | 12.2 69 | 8.71 44 | 6.83 32 | 12.9 34 | 6.65 54 | 7.61 45 | 11.7 57 | 5.88 45 | 22.2 71 | 21.4 95 | 41.8 53 | 30.6 51 | 27.1 39 | 45.8 76 | 27.9 35 | 19.1 14 | 36.6 74 | 6.67 25 | 14.7 30 | 10.3 6 | 8.16 18 | 16.5 24 | 7.34 75 |
Classic+CPF [83] | 46.5 | 7.22 44 | 11.6 40 | 8.52 21 | 6.90 35 | 12.6 32 | 6.28 18 | 7.37 20 | 10.6 27 | 5.76 28 | 21.2 5 | 18.7 5 | 41.1 3 | 31.1 108 | 27.9 108 | 45.5 38 | 28.7 93 | 23.1 117 | 36.3 52 | 6.92 74 | 15.3 64 | 10.3 6 | 8.75 81 | 17.9 87 | 6.99 9 |
S2D-Matching [84] | 48.2 | 7.37 61 | 12.3 72 | 8.80 60 | 7.62 58 | 14.2 53 | 6.43 37 | 7.28 11 | 10.4 19 | 5.74 25 | 21.6 24 | 19.1 17 | 42.2 80 | 30.6 51 | 27.3 68 | 45.4 23 | 28.6 88 | 22.5 106 | 36.4 61 | 6.76 50 | 14.5 24 | 10.3 6 | 8.46 52 | 17.0 43 | 7.32 69 |
DeepFlow [86] | 48.4 | 7.21 42 | 11.0 19 | 8.88 68 | 7.79 61 | 14.3 56 | 7.33 84 | 7.64 51 | 12.6 84 | 5.95 55 | 22.1 65 | 20.1 42 | 42.0 71 | 30.6 51 | 26.8 15 | 46.1 107 | 28.0 44 | 17.9 2 | 37.2 100 | 6.57 11 | 14.1 12 | 10.4 32 | 8.07 10 | 16.2 11 | 7.32 69 |
IROF-TV [53] | 48.5 | 7.33 55 | 12.3 72 | 8.82 63 | 6.83 32 | 12.3 24 | 6.23 12 | 7.70 60 | 12.9 92 | 5.93 52 | 21.5 20 | 19.5 27 | 42.0 71 | 30.8 83 | 27.3 68 | 45.9 94 | 27.5 7 | 20.2 45 | 35.6 6 | 6.75 48 | 15.1 58 | 10.5 57 | 8.18 21 | 16.4 17 | 7.37 79 |
EPPM w/o HM [88] | 48.5 | 6.77 5 | 10.4 10 | 8.32 3 | 7.00 39 | 13.4 42 | 6.16 5 | 8.19 87 | 13.6 103 | 6.26 76 | 21.7 31 | 20.3 49 | 41.5 28 | 30.5 35 | 27.2 54 | 45.5 38 | 28.5 83 | 21.8 89 | 36.5 69 | 6.84 60 | 15.6 74 | 10.6 82 | 8.41 45 | 17.1 51 | 6.95 7 |
p-harmonic [29] | 48.9 | 7.04 20 | 11.3 27 | 8.62 33 | 8.81 85 | 15.8 85 | 6.98 74 | 7.76 66 | 13.1 94 | 6.18 73 | 22.4 81 | 20.7 69 | 41.9 66 | 30.5 35 | 27.0 29 | 45.5 38 | 27.8 28 | 19.2 18 | 36.4 61 | 6.71 37 | 15.1 58 | 10.3 6 | 8.29 31 | 16.8 33 | 7.12 27 |
Brox et al. [5] | 49.1 | 7.28 49 | 11.4 33 | 8.76 50 | 7.86 64 | 14.6 65 | 6.92 72 | 8.03 81 | 13.1 94 | 6.34 82 | 21.9 46 | 19.9 36 | 41.4 24 | 30.6 51 | 27.0 29 | 45.8 76 | 27.7 18 | 19.5 26 | 36.2 38 | 6.80 55 | 15.4 69 | 10.4 32 | 8.16 18 | 16.5 24 | 7.19 46 |
Efficient-NL [60] | 49.5 | 7.28 49 | 11.6 40 | 8.61 32 | 7.24 47 | 13.3 41 | 6.35 26 | 8.21 89 | 10.8 36 | 6.39 87 | 21.7 31 | 19.6 29 | 41.2 11 | 30.4 26 | 27.0 29 | 45.3 19 | 28.3 65 | 22.8 113 | 35.6 6 | 6.86 67 | 15.6 74 | 10.4 32 | 9.10 102 | 18.3 104 | 7.14 33 |
SepConv-v1 [127] | 50.2 | 4.07 1 | 8.88 1 | 4.61 1 | 6.87 34 | 13.0 38 | 7.47 86 | 6.42 1 | 9.58 5 | 9.25 124 | 23.4 105 | 20.0 39 | 44.0 109 | 30.2 9 | 26.3 2 | 45.7 60 | 27.9 35 | 16.5 1 | 37.4 106 | 7.61 113 | 15.6 74 | 12.9 129 | 7.71 2 | 13.8 1 | 9.78 128 |
EpicFlow [102] | 52.9 | 7.18 38 | 12.0 56 | 8.72 45 | 7.42 52 | 14.4 60 | 6.72 59 | 7.68 58 | 12.1 68 | 5.92 51 | 22.1 65 | 21.1 86 | 42.0 71 | 30.7 71 | 27.1 39 | 45.8 76 | 27.5 7 | 19.9 41 | 35.9 13 | 6.79 54 | 15.2 60 | 10.4 32 | 8.40 44 | 17.1 51 | 7.33 72 |
DPOF [18] | 53.3 | 7.58 83 | 13.2 102 | 9.07 74 | 6.27 7 | 11.0 7 | 6.54 48 | 8.10 85 | 10.6 27 | 6.27 78 | 22.0 57 | 20.5 61 | 41.9 66 | 30.2 9 | 26.8 15 | 45.4 23 | 28.0 44 | 21.2 76 | 35.8 12 | 6.84 60 | 15.0 52 | 10.7 92 | 8.62 70 | 17.4 71 | 7.26 60 |
PMF [73] | 53.6 | 6.83 7 | 10.3 9 | 8.37 6 | 6.96 37 | 13.1 39 | 6.19 7 | 7.86 72 | 13.1 94 | 6.03 66 | 21.5 20 | 19.4 22 | 41.3 17 | 31.0 104 | 27.7 104 | 45.8 76 | 28.7 93 | 20.5 54 | 37.2 100 | 6.80 55 | 15.0 52 | 10.5 57 | 8.87 85 | 18.2 101 | 7.00 10 |
ComplOF-FED-GPU [35] | 54.4 | 7.23 46 | 11.8 49 | 8.72 45 | 7.20 45 | 13.9 47 | 6.62 52 | 8.43 97 | 12.6 84 | 6.45 89 | 21.9 46 | 20.8 75 | 42.3 82 | 30.4 26 | 26.9 20 | 45.4 23 | 27.7 18 | 20.1 44 | 36.1 31 | 6.86 67 | 15.4 69 | 10.5 57 | 8.55 63 | 17.3 65 | 7.28 65 |
Sparse Occlusion [54] | 55.5 | 7.37 61 | 12.3 72 | 8.87 67 | 8.04 68 | 15.3 74 | 6.48 41 | 7.58 42 | 10.8 36 | 5.87 43 | 22.0 57 | 20.4 55 | 41.5 28 | 30.6 51 | 27.2 54 | 45.5 38 | 28.3 65 | 21.8 89 | 36.4 61 | 6.80 55 | 15.3 64 | 10.3 6 | 8.74 80 | 17.7 80 | 7.18 45 |
TC/T-Flow [76] | 55.9 | 7.37 61 | 11.8 49 | 8.59 28 | 7.31 48 | 14.0 50 | 6.42 34 | 7.47 30 | 11.1 44 | 5.81 32 | 21.8 38 | 20.5 61 | 41.7 45 | 30.8 83 | 27.5 87 | 45.7 60 | 28.1 53 | 20.9 67 | 36.2 38 | 7.03 87 | 16.0 87 | 10.6 82 | 8.62 70 | 17.6 77 | 7.13 31 |
CLG-TV [48] | 57.5 | 7.52 78 | 12.3 72 | 9.14 78 | 8.67 83 | 15.8 85 | 7.11 78 | 7.97 78 | 12.7 88 | 6.26 76 | 22.1 65 | 20.3 49 | 42.0 71 | 30.5 35 | 26.9 20 | 45.7 60 | 27.6 15 | 19.1 14 | 36.2 38 | 6.71 37 | 14.9 45 | 10.4 32 | 8.53 62 | 17.3 65 | 7.24 57 |
AggregFlow [97] | 57.5 | 7.71 90 | 12.6 86 | 9.11 76 | 7.50 55 | 13.9 47 | 7.06 75 | 7.19 7 | 9.98 10 | 5.53 4 | 21.9 46 | 20.4 55 | 41.6 35 | 30.8 83 | 27.3 68 | 46.1 107 | 29.0 102 | 19.7 37 | 37.9 113 | 6.75 48 | 14.7 30 | 10.5 57 | 8.32 34 | 16.8 33 | 7.40 83 |
SuperFlow [81] | 57.7 | 7.43 70 | 11.5 35 | 9.30 88 | 8.55 82 | 14.8 68 | 9.15 99 | 7.91 76 | 12.0 65 | 6.31 79 | 22.1 65 | 19.9 36 | 42.0 71 | 30.7 71 | 27.2 54 | 45.9 94 | 27.3 2 | 18.4 6 | 35.9 13 | 6.86 67 | 15.8 81 | 10.6 82 | 8.15 16 | 16.5 24 | 7.16 40 |
RFlow [90] | 58.0 | 7.24 47 | 12.1 60 | 8.90 71 | 8.42 77 | 15.6 80 | 6.49 43 | 7.72 61 | 12.2 72 | 6.01 64 | 22.0 57 | 20.6 66 | 41.7 45 | 30.4 26 | 27.1 39 | 45.7 60 | 27.4 4 | 19.8 40 | 35.6 6 | 6.84 60 | 15.9 85 | 10.5 57 | 8.91 92 | 18.0 91 | 7.47 88 |
SIOF [67] | 58.5 | 7.66 87 | 12.6 86 | 9.09 75 | 9.45 99 | 16.6 98 | 8.48 93 | 7.65 53 | 11.9 63 | 5.98 59 | 21.9 46 | 20.1 42 | 41.8 53 | 30.0 3 | 26.5 3 | 45.3 19 | 28.1 53 | 19.7 37 | 36.6 74 | 6.63 20 | 14.7 30 | 10.5 57 | 8.82 83 | 17.9 87 | 7.46 85 |
TCOF [69] | 58.5 | 7.36 58 | 12.1 60 | 8.68 39 | 9.41 98 | 16.6 98 | 7.17 81 | 7.38 21 | 10.7 34 | 5.61 8 | 21.8 38 | 20.4 55 | 41.8 53 | 30.4 26 | 27.0 29 | 45.6 48 | 28.1 53 | 21.8 89 | 35.9 13 | 6.85 64 | 15.7 80 | 10.4 32 | 9.30 111 | 19.0 115 | 7.61 102 |
IAOF [50] | 59.0 | 8.70 111 | 12.9 95 | 10.3 108 | 12.4 121 | 19.2 125 | 9.77 110 | 7.74 64 | 12.0 65 | 6.21 74 | 22.8 91 | 20.2 46 | 42.0 71 | 30.2 9 | 26.5 3 | 45.5 38 | 27.7 18 | 19.6 29 | 36.1 31 | 6.67 25 | 15.0 52 | 10.3 6 | 8.41 45 | 17.1 51 | 7.12 27 |
TC-Flow [46] | 59.3 | 7.18 38 | 11.8 49 | 8.78 54 | 7.46 54 | 14.6 65 | 6.77 64 | 7.86 72 | 12.6 84 | 5.89 48 | 21.8 38 | 20.3 49 | 41.9 66 | 30.7 71 | 27.4 80 | 45.7 60 | 28.3 65 | 21.0 70 | 36.6 74 | 6.73 46 | 14.8 39 | 10.5 57 | 8.51 58 | 17.3 65 | 7.24 57 |
OAR-Flow [125] | 60.8 | 7.45 73 | 11.7 44 | 8.98 73 | 7.57 57 | 14.4 60 | 6.91 71 | 7.62 46 | 12.4 79 | 5.82 34 | 21.6 24 | 20.3 49 | 41.6 35 | 30.9 100 | 27.5 87 | 45.8 76 | 28.0 44 | 20.5 54 | 36.4 61 | 6.97 82 | 15.6 74 | 10.5 57 | 8.46 52 | 17.1 51 | 7.34 75 |
ALD-Flow [66] | 62.6 | 7.54 79 | 12.1 60 | 9.14 78 | 7.43 53 | 14.3 56 | 6.85 68 | 7.66 56 | 12.5 82 | 5.87 43 | 21.8 38 | 20.4 55 | 42.3 82 | 30.8 83 | 27.4 80 | 45.9 94 | 28.1 53 | 19.9 41 | 36.6 74 | 6.62 17 | 14.2 15 | 10.5 57 | 8.68 77 | 17.5 76 | 7.46 85 |
OFH [38] | 63.2 | 7.39 66 | 12.1 60 | 8.88 68 | 8.07 69 | 15.0 72 | 6.66 56 | 8.03 81 | 13.8 106 | 5.96 57 | 21.9 46 | 21.1 86 | 42.1 77 | 30.5 35 | 27.3 68 | 45.4 23 | 27.8 28 | 20.4 51 | 36.2 38 | 7.11 89 | 16.4 92 | 10.5 57 | 8.61 69 | 17.6 77 | 7.19 46 |
SVFilterOh [111] | 63.6 | 7.18 38 | 10.9 17 | 8.76 50 | 6.48 14 | 11.7 13 | 6.45 39 | 7.62 46 | 10.2 13 | 5.99 61 | 21.7 31 | 19.4 22 | 42.5 92 | 31.3 112 | 28.0 112 | 46.6 116 | 28.6 88 | 22.0 98 | 36.5 69 | 6.92 74 | 14.1 12 | 11.4 115 | 8.97 96 | 17.8 85 | 8.09 113 |
MLDP_OF [89] | 64.9 | 7.10 27 | 11.2 26 | 8.64 35 | 7.33 49 | 13.7 45 | 6.31 22 | 7.44 27 | 10.9 39 | 5.75 27 | 22.0 57 | 19.8 35 | 42.3 82 | 30.6 51 | 27.3 68 | 46.2 113 | 31.0 125 | 22.6 107 | 40.0 125 | 6.93 77 | 15.2 60 | 11.0 105 | 8.65 74 | 17.4 71 | 7.79 110 |
CostFilter [40] | 65.1 | 6.91 11 | 11.1 22 | 8.37 6 | 6.82 29 | 12.9 34 | 6.25 14 | 7.99 79 | 13.9 107 | 6.10 68 | 21.9 46 | 20.6 66 | 41.7 45 | 31.1 108 | 27.9 108 | 45.9 94 | 29.8 113 | 20.3 47 | 39.1 121 | 6.94 79 | 15.8 81 | 10.6 82 | 8.82 83 | 18.1 98 | 7.09 21 |
Modified CLG [34] | 65.7 | 7.63 85 | 11.6 40 | 9.65 94 | 10.7 109 | 17.2 106 | 10.7 114 | 8.25 91 | 14.3 112 | 6.60 95 | 22.4 81 | 21.1 86 | 41.8 53 | 30.6 51 | 26.9 20 | 45.8 76 | 27.7 18 | 19.2 18 | 36.3 52 | 6.69 32 | 14.9 45 | 10.4 32 | 8.41 45 | 17.0 43 | 7.35 78 |
Fusion [6] | 66.0 | 7.13 33 | 12.3 72 | 8.60 31 | 7.18 43 | 13.1 39 | 6.56 49 | 7.63 49 | 10.9 39 | 6.13 71 | 22.5 87 | 21.1 86 | 41.5 28 | 30.7 71 | 28.2 115 | 44.3 2 | 28.1 53 | 23.8 120 | 35.2 3 | 7.22 98 | 17.9 105 | 10.6 82 | 9.64 118 | 19.9 122 | 7.32 69 |
F-TV-L1 [15] | 67.0 | 8.24 101 | 13.1 99 | 9.92 101 | 9.28 94 | 16.3 93 | 7.48 87 | 8.00 80 | 13.2 99 | 6.35 84 | 22.3 77 | 20.9 76 | 42.3 82 | 29.9 2 | 26.9 20 | 44.8 6 | 27.9 35 | 19.4 23 | 36.5 69 | 6.87 71 | 15.4 69 | 10.5 57 | 8.46 52 | 16.8 33 | 7.58 98 |
FlowNet2 [122] | 67.9 | 9.30 114 | 14.6 115 | 10.5 111 | 8.42 77 | 14.6 65 | 9.24 103 | 8.03 81 | 12.5 82 | 6.14 72 | 22.2 71 | 21.9 104 | 41.8 53 | 30.9 100 | 27.5 87 | 45.8 76 | 28.0 44 | 20.5 54 | 36.0 18 | 6.72 41 | 14.9 45 | 10.4 32 | 8.31 33 | 16.9 41 | 7.01 11 |
Complementary OF [21] | 68.8 | 7.11 28 | 12.1 60 | 8.50 16 | 7.17 42 | 14.0 50 | 6.58 51 | 8.76 106 | 12.0 65 | 6.55 92 | 22.3 77 | 21.4 95 | 42.6 98 | 30.6 51 | 27.5 87 | 45.2 14 | 28.1 53 | 20.9 67 | 36.4 61 | 7.15 92 | 16.7 96 | 10.5 57 | 9.09 101 | 18.7 111 | 7.38 80 |
SimpleFlow [49] | 69.0 | 7.37 61 | 12.4 81 | 8.74 49 | 7.88 65 | 14.3 56 | 6.50 44 | 8.59 102 | 11.5 52 | 6.51 90 | 21.6 24 | 19.3 21 | 41.8 53 | 30.6 51 | 27.3 68 | 45.5 38 | 28.5 83 | 22.9 115 | 36.2 38 | 7.66 116 | 20.5 124 | 10.8 100 | 8.89 88 | 18.2 101 | 7.15 37 |
LDOF [28] | 69.5 | 8.08 96 | 12.3 72 | 9.79 99 | 8.94 89 | 14.9 70 | 9.18 101 | 8.23 90 | 13.5 102 | 6.52 91 | 22.3 77 | 21.1 86 | 42.4 90 | 30.6 51 | 27.0 29 | 45.8 76 | 27.9 35 | 18.8 10 | 36.6 74 | 6.77 52 | 15.3 64 | 10.4 32 | 8.44 51 | 17.1 51 | 7.38 80 |
ROF-ND [107] | 70.4 | 7.46 74 | 11.0 19 | 8.77 52 | 7.96 66 | 15.4 75 | 6.76 63 | 7.55 37 | 11.0 42 | 5.85 40 | 23.3 102 | 23.5 121 | 41.6 35 | 30.6 51 | 27.1 39 | 45.8 76 | 28.3 65 | 22.7 110 | 35.9 13 | 7.52 110 | 17.5 102 | 11.4 115 | 9.25 109 | 18.7 111 | 7.27 62 |
Local-TV-L1 [65] | 70.5 | 8.46 107 | 12.6 86 | 10.4 109 | 9.68 100 | 16.0 91 | 8.93 98 | 7.56 41 | 11.2 48 | 5.84 37 | 23.1 98 | 20.4 55 | 46.0 121 | 30.6 51 | 27.1 39 | 45.9 94 | 30.1 118 | 19.1 14 | 39.9 124 | 6.72 41 | 14.9 45 | 10.5 57 | 8.13 15 | 16.1 6 | 7.58 98 |
TF+OM [100] | 70.8 | 7.41 67 | 12.1 60 | 9.19 83 | 7.21 46 | 12.9 34 | 7.83 88 | 7.55 37 | 12.3 76 | 5.82 34 | 22.2 71 | 21.0 82 | 41.9 66 | 30.8 83 | 27.5 87 | 46.0 104 | 28.3 65 | 20.5 54 | 36.8 88 | 6.97 82 | 16.3 90 | 10.5 57 | 8.65 74 | 17.3 65 | 7.75 106 |
TriFlow [95] | 71.6 | 7.77 92 | 13.7 110 | 9.28 86 | 8.98 90 | 15.7 83 | 9.30 105 | 7.65 53 | 12.4 79 | 5.84 37 | 22.0 57 | 20.9 76 | 41.1 3 | 30.9 100 | 27.7 104 | 45.7 60 | 28.4 77 | 21.3 80 | 36.3 52 | 6.85 64 | 15.5 73 | 10.4 32 | 8.69 78 | 17.4 71 | 7.23 56 |
Classic++ [32] | 72.2 | 7.49 76 | 12.5 84 | 9.11 76 | 8.07 69 | 15.2 73 | 6.67 57 | 7.89 75 | 12.6 84 | 6.04 67 | 22.3 77 | 20.7 69 | 42.2 80 | 30.6 51 | 27.2 54 | 45.7 60 | 29.0 102 | 21.0 70 | 37.6 108 | 6.81 58 | 15.2 60 | 10.5 57 | 8.62 70 | 17.4 71 | 7.46 85 |
Occlusion-TV-L1 [63] | 72.2 | 7.44 71 | 12.3 72 | 9.14 78 | 8.91 88 | 16.5 96 | 6.85 68 | 7.83 70 | 12.8 89 | 6.32 80 | 22.6 90 | 21.5 99 | 42.5 92 | 30.5 35 | 26.9 20 | 45.8 76 | 28.4 77 | 19.6 29 | 37.1 97 | 7.15 92 | 14.8 39 | 10.7 92 | 8.51 58 | 17.1 51 | 7.34 75 |
Nguyen [33] | 72.6 | 9.74 117 | 12.6 86 | 12.4 121 | 12.3 119 | 18.6 121 | 11.1 115 | 8.27 93 | 14.8 114 | 6.69 97 | 23.4 105 | 21.7 101 | 41.8 53 | 30.3 19 | 26.8 15 | 45.3 19 | 27.4 4 | 19.6 29 | 35.7 9 | 7.24 100 | 18.3 108 | 10.5 57 | 8.37 42 | 17.0 43 | 7.22 55 |
2D-CLG [1] | 72.7 | 8.44 105 | 12.3 72 | 10.6 113 | 11.9 117 | 18.0 116 | 12.3 121 | 8.94 109 | 13.9 107 | 7.33 112 | 23.1 98 | 21.2 93 | 41.3 17 | 30.5 35 | 26.9 20 | 45.8 76 | 27.6 15 | 19.2 18 | 36.2 38 | 7.14 90 | 17.2 100 | 10.5 57 | 8.37 42 | 16.5 24 | 7.20 49 |
FlowNetS+ft+v [112] | 73.7 | 7.81 94 | 11.7 44 | 9.63 93 | 9.77 102 | 16.8 100 | 9.16 100 | 8.06 84 | 13.4 101 | 6.36 85 | 22.1 65 | 20.7 69 | 42.1 77 | 30.8 83 | 27.4 80 | 45.8 76 | 27.7 18 | 19.4 23 | 36.3 52 | 7.01 84 | 16.4 92 | 10.5 57 | 8.51 58 | 17.2 60 | 7.33 72 |
Aniso-Texture [82] | 74.2 | 7.16 35 | 11.6 40 | 8.78 54 | 8.84 86 | 16.5 96 | 6.86 70 | 8.38 96 | 11.8 60 | 5.99 61 | 22.4 81 | 21.4 95 | 42.5 92 | 31.0 104 | 27.5 87 | 46.0 104 | 29.0 102 | 24.2 123 | 36.7 83 | 6.70 35 | 14.7 30 | 10.3 6 | 8.90 89 | 18.0 91 | 7.27 62 |
Shiralkar [42] | 74.5 | 7.48 75 | 12.8 93 | 8.80 60 | 9.00 92 | 15.8 85 | 6.65 54 | 8.52 100 | 16.1 118 | 6.84 101 | 23.4 105 | 22.3 108 | 41.6 35 | 30.0 3 | 27.0 29 | 44.5 3 | 28.7 93 | 21.1 74 | 37.1 97 | 7.49 108 | 18.7 116 | 10.6 82 | 8.64 73 | 17.7 80 | 6.93 5 |
Adaptive [20] | 75.6 | 7.71 90 | 13.2 102 | 9.21 85 | 9.40 97 | 16.8 100 | 7.07 76 | 7.87 74 | 12.4 79 | 6.12 69 | 22.0 57 | 20.3 49 | 41.8 53 | 30.7 71 | 27.3 68 | 45.6 48 | 28.4 77 | 20.7 65 | 36.8 88 | 6.95 81 | 16.0 87 | 10.4 32 | 8.87 85 | 17.9 87 | 7.55 95 |
CNN-flow-warp+ref [117] | 77.7 | 7.35 56 | 10.8 13 | 9.30 88 | 8.87 87 | 16.2 92 | 8.14 92 | 8.60 103 | 14.1 110 | 6.62 96 | 23.7 109 | 21.9 104 | 42.7 99 | 30.8 83 | 27.3 68 | 45.9 94 | 28.0 44 | 19.1 14 | 36.7 83 | 7.37 103 | 18.5 113 | 10.6 82 | 8.33 36 | 16.8 33 | 7.27 62 |
CRTflow [80] | 78.1 | 7.69 88 | 12.6 86 | 9.28 86 | 8.45 80 | 15.5 77 | 6.81 66 | 8.55 101 | 14.0 109 | 7.29 110 | 22.4 81 | 20.7 69 | 43.8 108 | 30.7 71 | 27.2 54 | 45.7 60 | 28.1 53 | 19.6 29 | 36.7 83 | 6.87 71 | 15.8 81 | 10.6 82 | 8.59 66 | 17.2 60 | 7.65 103 |
Black & Anandan [4] | 78.2 | 8.54 108 | 12.8 93 | 10.2 107 | 10.9 111 | 17.3 109 | 9.40 106 | 9.06 111 | 13.6 103 | 6.99 105 | 22.9 95 | 21.3 94 | 41.7 45 | 30.7 71 | 27.2 54 | 45.9 94 | 28.0 44 | 18.6 7 | 36.7 83 | 6.93 77 | 15.9 85 | 10.4 32 | 8.46 52 | 17.0 43 | 7.20 49 |
HBpMotionGpu [43] | 78.9 | 9.39 115 | 14.6 115 | 11.3 117 | 11.7 116 | 18.9 123 | 11.5 118 | 7.55 37 | 11.1 44 | 6.00 63 | 23.3 102 | 22.3 108 | 43.5 107 | 30.3 19 | 27.2 54 | 45.2 14 | 28.7 93 | 20.9 67 | 37.1 97 | 6.62 17 | 14.2 15 | 10.5 57 | 8.99 98 | 17.8 85 | 8.04 112 |
StereoOF-V1MT [119] | 79.3 | 7.65 86 | 13.5 107 | 8.77 52 | 8.69 84 | 15.9 90 | 6.52 47 | 9.43 116 | 15.4 116 | 7.23 108 | 24.4 115 | 22.3 108 | 43.2 105 | 30.5 35 | 27.2 54 | 45.0 9 | 28.9 100 | 21.2 76 | 37.2 100 | 7.77 119 | 19.4 119 | 11.0 105 | 8.26 30 | 16.4 17 | 6.93 5 |
GraphCuts [14] | 79.4 | 8.65 110 | 14.1 114 | 9.83 100 | 8.28 73 | 14.2 53 | 9.28 104 | 9.89 120 | 10.6 27 | 7.38 113 | 23.0 96 | 21.1 86 | 42.5 92 | 30.3 19 | 27.3 68 | 44.7 5 | 27.2 1 | 21.4 81 | 34.7 1 | 7.42 106 | 17.8 103 | 11.0 105 | 9.32 112 | 18.9 113 | 7.66 104 |
HBM-GC [105] | 80.8 | 7.91 95 | 12.6 86 | 9.75 97 | 7.51 56 | 13.9 47 | 6.80 65 | 7.29 13 | 9.43 2 | 5.94 53 | 22.0 57 | 19.7 34 | 42.3 82 | 32.1 122 | 28.6 119 | 48.0 123 | 30.0 115 | 24.6 126 | 37.8 110 | 7.14 90 | 14.8 39 | 11.6 118 | 8.95 95 | 17.7 80 | 8.28 115 |
Steered-L1 [118] | 81.0 | 7.06 24 | 12.2 69 | 8.59 28 | 7.40 51 | 14.3 56 | 6.83 67 | 8.48 99 | 11.7 57 | 6.69 97 | 22.8 91 | 20.9 76 | 42.7 99 | 31.2 111 | 28.1 113 | 45.8 76 | 28.3 65 | 21.2 76 | 36.7 83 | 7.25 101 | 17.8 103 | 10.9 101 | 9.00 99 | 18.3 104 | 7.58 98 |
CBF [12] | 81.0 | 7.41 67 | 11.9 54 | 9.31 90 | 8.07 69 | 14.9 70 | 7.14 80 | 7.69 59 | 11.1 44 | 5.95 55 | 22.8 91 | 20.7 69 | 45.1 117 | 30.8 83 | 27.3 68 | 47.0 120 | 28.2 62 | 20.6 62 | 36.5 69 | 7.17 95 | 16.6 95 | 11.2 111 | 9.16 106 | 17.9 87 | 8.83 122 |
TriangleFlow [30] | 81.1 | 7.79 93 | 13.0 98 | 9.16 82 | 8.36 76 | 15.5 77 | 6.69 58 | 8.20 88 | 11.9 63 | 6.59 93 | 22.5 87 | 21.0 82 | 42.5 92 | 30.1 5 | 27.0 29 | 45.0 9 | 28.9 100 | 22.6 107 | 36.5 69 | 7.42 106 | 18.3 108 | 11.0 105 | 9.49 115 | 19.3 117 | 7.47 88 |
Correlation Flow [75] | 81.6 | 7.05 22 | 11.7 44 | 8.32 3 | 8.29 74 | 15.6 80 | 6.56 49 | 7.64 51 | 10.8 36 | 5.89 48 | 22.2 71 | 20.1 42 | 42.9 103 | 31.7 116 | 27.7 104 | 49.9 128 | 29.6 110 | 23.8 120 | 37.2 100 | 7.62 114 | 19.0 118 | 11.3 113 | 9.22 108 | 18.6 110 | 7.51 94 |
IAOF2 [51] | 83.4 | 8.43 104 | 13.6 109 | 9.76 98 | 9.86 104 | 17.4 110 | 8.67 94 | 7.74 64 | 12.2 72 | 6.33 81 | 23.1 98 | 21.7 101 | 42.3 82 | 31.0 104 | 27.9 108 | 45.6 48 | 28.5 83 | 21.0 70 | 36.6 74 | 6.71 37 | 15.0 52 | 10.3 6 | 9.14 104 | 18.4 107 | 7.49 92 |
BriefMatch [124] | 85.4 | 7.38 65 | 12.0 56 | 8.85 64 | 7.71 60 | 14.5 63 | 7.86 89 | 8.77 107 | 11.7 57 | 7.25 109 | 24.2 113 | 22.0 106 | 46.2 122 | 30.8 83 | 27.4 80 | 46.1 107 | 31.8 128 | 21.7 86 | 41.3 128 | 6.85 64 | 15.3 64 | 10.7 92 | 8.51 58 | 17.1 51 | 7.56 97 |
SegOF [10] | 87.1 | 8.16 99 | 12.4 81 | 10.1 106 | 9.10 93 | 15.5 77 | 8.83 97 | 9.48 117 | 14.1 110 | 7.46 115 | 22.8 91 | 23.0 119 | 41.6 35 | 30.8 83 | 27.4 80 | 45.8 76 | 28.3 65 | 22.0 98 | 36.3 52 | 7.83 120 | 21.5 126 | 11.0 105 | 8.46 52 | 17.1 51 | 7.17 43 |
BlockOverlap [61] | 87.9 | 8.81 112 | 12.4 81 | 11.1 116 | 10.0 106 | 15.8 85 | 10.6 113 | 7.84 71 | 10.4 19 | 6.59 93 | 23.3 102 | 20.4 55 | 46.3 123 | 31.9 119 | 27.9 108 | 48.8 126 | 30.3 121 | 19.7 37 | 39.8 123 | 7.08 88 | 14.5 24 | 11.7 121 | 8.35 40 | 16.1 6 | 8.60 120 |
TV-L1-improved [17] | 88.4 | 7.55 81 | 12.9 95 | 9.15 81 | 9.36 95 | 16.9 102 | 7.19 82 | 8.63 104 | 12.2 72 | 6.92 103 | 22.2 71 | 21.0 82 | 42.3 82 | 30.8 83 | 27.5 87 | 45.6 48 | 28.5 83 | 21.4 81 | 36.8 88 | 7.38 104 | 18.6 115 | 10.7 92 | 8.94 93 | 18.0 91 | 7.75 106 |
Dynamic MRF [7] | 88.8 | 7.29 52 | 13.1 99 | 8.69 40 | 8.20 72 | 16.3 93 | 6.74 60 | 9.18 113 | 16.4 121 | 7.22 107 | 24.5 117 | 23.1 120 | 44.4 111 | 30.3 19 | 27.2 54 | 45.1 12 | 29.2 107 | 23.4 118 | 37.2 100 | 7.64 115 | 19.8 122 | 10.7 92 | 9.14 104 | 18.0 91 | 7.48 91 |
AdaConv-v1 [126] | 89.3 | 9.81 119 | 13.9 112 | 11.6 119 | 12.1 118 | 17.6 112 | 16.0 127 | 11.4 127 | 16.1 118 | 13.1 129 | 26.5 126 | 24.4 126 | 45.3 118 | 28.4 1 | 24.4 1 | 44.6 4 | 28.4 77 | 18.1 4 | 37.7 109 | 7.74 118 | 16.3 90 | 13.1 130 | 8.25 28 | 15.1 2 | 10.1 129 |
LocallyOriented [52] | 89.6 | 8.08 96 | 13.1 99 | 9.72 95 | 9.73 101 | 17.0 104 | 7.88 90 | 8.34 95 | 12.8 89 | 6.34 82 | 23.0 96 | 22.1 107 | 43.0 104 | 30.6 51 | 27.2 54 | 45.6 48 | 30.0 115 | 21.9 93 | 38.5 119 | 7.02 85 | 15.8 81 | 10.5 57 | 9.05 100 | 18.4 107 | 7.39 82 |
SPSA-learn [13] | 90.4 | 8.28 102 | 12.9 95 | 9.95 103 | 9.92 105 | 16.3 93 | 9.49 107 | 9.15 112 | 12.8 89 | 7.30 111 | 23.1 98 | 20.5 61 | 41.6 35 | 30.8 83 | 27.5 87 | 45.7 60 | 28.0 44 | 20.4 51 | 36.3 52 | 8.81 130 | 27.1 131 | 11.8 122 | 10.0 123 | 21.0 126 | 7.20 49 |
Rannacher [23] | 91.1 | 7.69 88 | 13.2 102 | 9.32 91 | 9.37 96 | 16.9 102 | 7.28 83 | 8.67 105 | 13.0 93 | 6.91 102 | 22.2 71 | 21.1 86 | 42.4 90 | 30.8 83 | 27.5 87 | 45.7 60 | 28.5 83 | 21.2 76 | 36.9 93 | 7.35 102 | 18.5 113 | 10.7 92 | 8.90 89 | 18.0 91 | 7.78 108 |
Ad-TV-NDC [36] | 91.5 | 10.8 122 | 13.9 112 | 13.4 122 | 11.6 115 | 17.6 112 | 11.2 116 | 7.77 67 | 12.3 76 | 6.12 69 | 24.0 111 | 21.6 100 | 44.4 111 | 31.1 108 | 27.6 100 | 46.1 107 | 29.0 102 | 19.3 22 | 38.0 114 | 6.87 71 | 15.4 69 | 10.5 57 | 8.59 66 | 17.0 43 | 7.71 105 |
ACK-Prior [27] | 92.5 | 7.12 30 | 11.7 44 | 8.57 24 | 7.08 40 | 13.8 46 | 6.34 25 | 8.81 108 | 11.8 60 | 6.69 97 | 22.5 87 | 21.4 95 | 42.3 82 | 32.6 126 | 29.3 125 | 48.2 124 | 30.7 124 | 25.6 128 | 38.1 116 | 7.95 123 | 18.8 117 | 12.0 123 | 10.8 128 | 21.8 128 | 8.53 119 |
Horn & Schunck [3] | 93.2 | 8.45 106 | 13.3 106 | 10.0 104 | 11.4 114 | 18.1 118 | 9.84 111 | 9.65 118 | 16.1 118 | 7.89 118 | 24.6 118 | 22.8 113 | 42.8 101 | 30.6 51 | 27.2 54 | 45.6 48 | 28.3 65 | 19.4 23 | 36.8 88 | 7.41 105 | 18.0 107 | 10.6 82 | 8.94 93 | 17.7 80 | 7.55 95 |
UnFlow [129] | 95.4 | 9.13 113 | 15.0 119 | 10.7 114 | 10.9 111 | 18.1 118 | 9.23 102 | 9.21 114 | 16.9 124 | 7.18 106 | 22.4 81 | 21.8 103 | 41.8 53 | 30.8 83 | 27.6 100 | 45.9 94 | 28.6 88 | 22.7 110 | 36.0 18 | 6.94 79 | 15.6 74 | 10.5 57 | 10.0 123 | 19.3 117 | 7.47 88 |
TI-DOFE [24] | 95.8 | 11.8 124 | 14.7 117 | 14.8 125 | 13.9 126 | 20.3 127 | 13.5 125 | 9.26 115 | 16.5 123 | 7.69 117 | 25.1 120 | 22.8 113 | 43.3 106 | 30.2 9 | 27.1 39 | 45.4 23 | 28.4 77 | 19.6 29 | 36.8 88 | 7.22 98 | 17.2 100 | 10.7 92 | 9.21 107 | 18.1 98 | 7.59 101 |
StereoFlow [44] | 96.0 | 13.8 127 | 20.2 130 | 14.0 123 | 14.1 127 | 21.3 130 | 12.0 120 | 7.79 69 | 13.3 100 | 5.98 59 | 22.4 81 | 20.9 76 | 42.1 77 | 33.7 130 | 32.3 130 | 46.1 107 | 30.5 122 | 31.8 131 | 36.3 52 | 6.63 20 | 14.7 30 | 10.4 32 | 9.98 122 | 21.0 126 | 7.42 84 |
Filter Flow [19] | 101.1 | 8.30 103 | 13.2 102 | 10.0 104 | 10.8 110 | 17.1 105 | 11.7 119 | 7.96 77 | 12.1 68 | 6.38 86 | 23.7 109 | 20.9 76 | 44.5 115 | 31.5 115 | 28.2 115 | 46.7 118 | 28.8 98 | 21.0 70 | 37.3 105 | 7.15 92 | 17.0 99 | 10.7 92 | 9.54 117 | 18.9 113 | 8.47 118 |
NL-TV-NCC [25] | 103.2 | 7.56 82 | 12.7 92 | 8.62 33 | 8.00 67 | 15.4 75 | 6.74 60 | 8.46 98 | 13.1 94 | 6.70 100 | 24.2 113 | 24.0 124 | 45.0 116 | 32.8 127 | 28.3 118 | 52.0 131 | 29.4 108 | 24.1 122 | 36.9 93 | 7.89 122 | 17.9 105 | 12.4 127 | 10.1 125 | 19.9 122 | 8.92 123 |
Bartels [41] | 104.5 | 8.10 98 | 13.8 111 | 9.94 102 | 8.35 75 | 15.8 85 | 8.75 95 | 8.11 86 | 12.1 68 | 6.97 104 | 24.1 112 | 22.7 111 | 47.6 125 | 32.4 123 | 27.8 107 | 51.1 130 | 35.4 130 | 23.0 116 | 46.5 130 | 7.18 96 | 14.9 45 | 12.3 126 | 9.36 114 | 18.0 91 | 9.76 127 |
SILK [79] | 104.5 | 9.77 118 | 15.1 120 | 11.8 120 | 12.3 119 | 18.7 122 | 11.2 116 | 10.3 121 | 16.4 121 | 8.14 120 | 25.2 121 | 22.8 113 | 45.9 120 | 30.8 83 | 27.5 87 | 45.7 60 | 30.6 123 | 20.3 47 | 40.1 126 | 7.19 97 | 16.8 97 | 10.9 101 | 8.87 85 | 17.6 77 | 7.50 93 |
SLK [47] | 109.7 | 11.4 123 | 15.4 121 | 14.4 124 | 12.4 121 | 18.0 116 | 12.6 122 | 10.9 125 | 17.6 126 | 8.85 122 | 27.8 127 | 25.2 127 | 46.6 124 | 30.6 51 | 28.1 113 | 43.6 1 | 29.0 102 | 21.9 93 | 37.0 96 | 8.25 126 | 22.4 127 | 11.3 113 | 9.33 113 | 18.5 109 | 7.91 111 |
GroupFlow [9] | 110.2 | 10.1 121 | 16.9 125 | 11.3 117 | 10.4 108 | 17.8 115 | 10.0 112 | 10.8 124 | 17.5 125 | 9.21 123 | 23.6 108 | 23.9 122 | 42.5 92 | 31.9 119 | 29.3 125 | 46.2 113 | 30.1 118 | 24.5 125 | 37.8 110 | 7.55 111 | 18.4 111 | 10.6 82 | 9.52 116 | 19.8 121 | 6.89 2 |
Heeger++ [104] | 110.4 | 9.81 119 | 17.3 126 | 10.4 109 | 11.3 113 | 17.2 106 | 9.67 108 | 13.6 129 | 23.8 130 | 10.2 127 | 26.3 124 | 22.8 113 | 44.4 111 | 31.8 118 | 28.9 124 | 46.3 115 | 29.6 110 | 22.0 98 | 37.5 107 | 8.17 125 | 19.8 122 | 10.9 101 | 9.10 102 | 18.2 101 | 7.02 12 |
Learning Flow [11] | 112.3 | 8.21 100 | 14.8 118 | 9.74 96 | 9.78 103 | 17.6 112 | 8.11 91 | 9.68 119 | 15.5 117 | 7.56 116 | 25.0 119 | 24.3 125 | 45.4 119 | 31.9 119 | 28.7 122 | 47.3 121 | 29.4 108 | 22.0 98 | 37.8 110 | 7.49 108 | 18.3 108 | 10.9 101 | 10.2 126 | 20.2 124 | 8.31 116 |
2bit-BM-tele [98] | 113.1 | 8.61 109 | 13.5 107 | 10.5 111 | 10.0 106 | 17.5 111 | 9.73 109 | 8.26 92 | 11.5 52 | 7.40 114 | 24.4 115 | 22.7 111 | 48.1 126 | 32.5 125 | 28.6 119 | 50.2 129 | 34.7 129 | 24.3 124 | 44.9 129 | 9.35 131 | 26.2 130 | 13.8 131 | 9.25 109 | 17.3 65 | 10.2 130 |
FFV1MT [106] | 117.0 | 9.53 116 | 16.7 124 | 10.7 114 | 12.6 124 | 18.2 120 | 12.8 123 | 13.3 128 | 23.5 129 | 10.5 128 | 26.3 124 | 22.8 113 | 44.4 111 | 31.4 114 | 28.2 115 | 46.1 107 | 29.8 113 | 20.6 62 | 38.1 116 | 8.32 128 | 20.5 124 | 11.0 105 | 10.5 127 | 20.4 125 | 8.45 117 |
Adaptive flow [45] | 119.6 | 13.2 126 | 15.9 122 | 16.2 127 | 14.2 128 | 19.9 126 | 16.4 128 | 9.02 110 | 13.1 94 | 8.03 119 | 26.0 123 | 22.8 113 | 48.6 127 | 32.4 123 | 29.4 127 | 47.9 122 | 30.1 118 | 24.6 126 | 38.0 114 | 7.55 111 | 16.9 98 | 12.2 124 | 9.85 119 | 19.5 119 | 8.97 126 |
FOLKI [16] | 119.8 | 15.0 129 | 17.4 127 | 19.4 129 | 14.3 129 | 20.9 129 | 14.4 126 | 10.7 123 | 19.2 128 | 9.99 126 | 29.8 129 | 26.8 128 | 53.1 130 | 31.3 112 | 28.6 119 | 45.8 76 | 30.0 115 | 21.7 86 | 38.9 120 | 7.85 121 | 19.4 119 | 11.6 118 | 9.85 119 | 19.2 116 | 8.80 121 |
Pyramid LK [2] | 122.1 | 16.3 130 | 16.1 123 | 21.6 130 | 16.0 130 | 20.3 127 | 18.2 130 | 16.7 130 | 15.3 115 | 14.3 130 | 35.7 131 | 36.7 131 | 56.5 131 | 32.8 127 | 31.2 129 | 45.7 60 | 29.7 112 | 22.1 102 | 38.2 118 | 8.31 127 | 23.1 128 | 11.6 118 | 11.8 129 | 25.0 129 | 8.22 114 |
PGAM+LK [55] | 123.1 | 12.7 125 | 18.1 128 | 15.3 126 | 12.4 121 | 19.1 124 | 13.0 124 | 11.1 126 | 18.6 127 | 9.33 125 | 29.2 128 | 27.5 129 | 51.6 129 | 31.7 116 | 28.8 123 | 46.7 118 | 31.3 126 | 23.7 119 | 40.1 126 | 7.67 117 | 19.4 119 | 11.4 115 | 9.89 121 | 19.5 119 | 8.95 124 |
HCIC-L [99] | 123.8 | 18.0 131 | 18.7 129 | 23.1 131 | 12.7 125 | 17.2 106 | 17.0 129 | 10.5 122 | 14.4 113 | 8.49 121 | 25.7 122 | 23.9 122 | 44.3 110 | 33.2 129 | 29.8 128 | 49.0 127 | 31.7 127 | 26.4 130 | 39.2 122 | 7.98 124 | 18.4 111 | 12.4 127 | 12.4 130 | 25.3 131 | 8.96 125 |
Periodicity [78] | 129.7 | 14.9 128 | 20.8 131 | 18.2 128 | 20.1 131 | 22.0 131 | 21.5 131 | 17.7 131 | 26.4 131 | 16.1 131 | 29.8 129 | 34.8 130 | 49.7 128 | 35.4 131 | 34.2 131 | 48.7 125 | 37.1 131 | 25.8 129 | 47.4 131 | 8.68 129 | 23.6 129 | 12.2 124 | 13.3 131 | 25.1 130 | 11.6 131 |
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. |