Optical flow evaluation results |
Statistics:
Average
SD
R0.5
R1.0
R2.0
A50
A75
A95
Error type: endpoint angle interpolation normalized interpolation |
A50 endpoint error |
avg. |
Army (Hidden texture) GT im0 im1 |
Mequon (Hidden texture) GT im0 im1 |
Schefflera (Hidden texture) GT im0 im1 |
Wooden (Hidden texture) GT im0 im1 |
Grove (Synthetic) GT im0 im1 |
Urban (Synthetic) GT im0 im1 |
Yosemite (Synthetic) GT im0 im1 |
Teddy (Stereo) GT im0 im1 | ||||||||||||||||
rank | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | all | disc | untext | |
NN-field [71] | 13.0 | 0.04 5 | 0.06 2 | 0.04 5 | 0.08 12 | 0.25 29 | 0.08 10 | 0.10 1 | 0.11 9 | 0.12 3 | 0.04 4 | 0.13 13 | 0.04 5 | 0.13 5 | 0.22 5 | 0.10 9 | 0.07 19 | 0.13 3 | 0.07 16 | 0.09 61 | 0.09 19 | 0.17 51 | 0.14 1 | 0.28 24 | 0.11 1 |
ComponentFusion [96] | 13.2 | 0.03 1 | 0.07 8 | 0.03 1 | 0.07 4 | 0.20 8 | 0.07 4 | 0.11 6 | 0.11 9 | 0.12 3 | 0.03 1 | 0.17 38 | 0.03 1 | 0.14 10 | 0.25 15 | 0.10 9 | 0.07 19 | 0.28 77 | 0.06 5 | 0.07 23 | 0.10 45 | 0.10 10 | 0.16 4 | 0.27 14 | 0.15 3 |
NNF-Local [87] | 13.3 | 0.04 5 | 0.06 2 | 0.03 1 | 0.07 4 | 0.22 16 | 0.07 4 | 0.10 1 | 0.11 9 | 0.11 1 | 0.05 24 | 0.13 13 | 0.04 5 | 0.14 10 | 0.23 10 | 0.11 13 | 0.08 45 | 0.15 13 | 0.09 63 | 0.07 23 | 0.08 3 | 0.14 38 | 0.14 1 | 0.27 14 | 0.12 2 |
ALD-Flow [66] | 13.5 | 0.04 5 | 0.07 8 | 0.04 5 | 0.07 4 | 0.19 4 | 0.07 4 | 0.11 6 | 0.11 9 | 0.12 3 | 0.04 4 | 0.12 9 | 0.04 5 | 0.13 5 | 0.25 15 | 0.09 4 | 0.05 3 | 0.16 18 | 0.06 5 | 0.06 5 | 0.08 3 | 0.22 77 | 0.18 29 | 0.34 43 | 0.20 51 |
TC/T-Flow [76] | 13.8 | 0.03 1 | 0.07 8 | 0.03 1 | 0.06 1 | 0.25 29 | 0.06 1 | 0.11 6 | 0.12 26 | 0.12 3 | 0.03 1 | 0.12 9 | 0.03 1 | 0.13 5 | 0.25 15 | 0.08 1 | 0.05 3 | 0.17 24 | 0.05 3 | 0.05 2 | 0.07 2 | 0.22 77 | 0.18 29 | 0.32 39 | 0.19 44 |
WLIF-Flow [93] | 14.5 | 0.04 5 | 0.07 8 | 0.05 18 | 0.08 12 | 0.24 19 | 0.09 22 | 0.11 6 | 0.11 9 | 0.13 17 | 0.05 24 | 0.13 13 | 0.04 5 | 0.15 15 | 0.24 14 | 0.11 13 | 0.07 19 | 0.16 18 | 0.07 16 | 0.07 23 | 0.09 19 | 0.16 45 | 0.16 4 | 0.23 1 | 0.15 3 |
RNLOD-Flow [121] | 14.9 | 0.04 5 | 0.06 2 | 0.04 5 | 0.07 4 | 0.24 19 | 0.07 4 | 0.11 6 | 0.10 3 | 0.12 3 | 0.04 4 | 0.11 5 | 0.04 5 | 0.13 5 | 0.22 5 | 0.09 4 | 0.07 19 | 0.16 18 | 0.07 16 | 0.09 61 | 0.11 68 | 0.23 86 | 0.16 4 | 0.25 3 | 0.15 3 |
TC-Flow [46] | 15.2 | 0.04 5 | 0.08 27 | 0.04 5 | 0.06 1 | 0.18 2 | 0.06 1 | 0.11 6 | 0.11 9 | 0.12 3 | 0.04 4 | 0.10 3 | 0.04 5 | 0.13 5 | 0.26 20 | 0.09 4 | 0.06 8 | 0.20 42 | 0.07 16 | 0.05 2 | 0.08 3 | 0.20 70 | 0.18 29 | 0.34 43 | 0.20 51 |
MDP-Flow2 [68] | 16.2 | 0.05 34 | 0.08 27 | 0.05 18 | 0.07 4 | 0.19 4 | 0.07 4 | 0.11 6 | 0.11 9 | 0.12 3 | 0.05 24 | 0.12 9 | 0.05 39 | 0.15 15 | 0.23 10 | 0.11 13 | 0.08 45 | 0.14 6 | 0.07 16 | 0.07 23 | 0.09 19 | 0.12 20 | 0.18 29 | 0.25 3 | 0.16 8 |
nLayers [57] | 16.7 | 0.04 5 | 0.05 1 | 0.04 5 | 0.11 70 | 0.24 19 | 0.12 81 | 0.11 6 | 0.10 3 | 0.13 17 | 0.04 4 | 0.08 1 | 0.04 5 | 0.12 4 | 0.18 3 | 0.09 4 | 0.07 19 | 0.12 1 | 0.06 5 | 0.08 44 | 0.09 19 | 0.15 41 | 0.17 13 | 0.27 14 | 0.17 17 |
OAR-Flow [125] | 18.0 | 0.04 5 | 0.09 40 | 0.05 18 | 0.07 4 | 0.24 19 | 0.07 4 | 0.11 6 | 0.13 38 | 0.13 17 | 0.04 4 | 0.11 5 | 0.04 5 | 0.16 27 | 0.30 41 | 0.10 9 | 0.04 1 | 0.15 13 | 0.04 1 | 0.05 2 | 0.06 1 | 0.14 38 | 0.18 29 | 0.36 48 | 0.21 57 |
OFLAF [77] | 18.1 | 0.05 34 | 0.07 8 | 0.05 18 | 0.08 12 | 0.20 8 | 0.08 10 | 0.10 1 | 0.09 2 | 0.12 3 | 0.05 24 | 0.11 5 | 0.04 5 | 0.11 1 | 0.17 1 | 0.09 4 | 0.09 62 | 0.15 13 | 0.08 45 | 0.07 23 | 0.08 3 | 0.18 57 | 0.19 45 | 0.26 7 | 0.19 44 |
Layers++ [37] | 18.8 | 0.04 5 | 0.07 8 | 0.05 18 | 0.10 48 | 0.24 19 | 0.11 68 | 0.11 6 | 0.10 3 | 0.13 17 | 0.04 4 | 0.08 1 | 0.04 5 | 0.11 1 | 0.18 3 | 0.08 1 | 0.07 19 | 0.14 6 | 0.07 16 | 0.09 61 | 0.10 45 | 0.19 65 | 0.17 13 | 0.25 3 | 0.17 17 |
LME [70] | 20.0 | 0.05 34 | 0.07 8 | 0.04 5 | 0.08 12 | 0.19 4 | 0.08 10 | 0.11 6 | 0.12 26 | 0.13 17 | 0.05 24 | 0.18 48 | 0.05 39 | 0.16 27 | 0.26 20 | 0.11 13 | 0.07 19 | 0.18 32 | 0.07 16 | 0.07 23 | 0.09 19 | 0.13 27 | 0.18 29 | 0.27 14 | 0.16 8 |
NNF-EAC [103] | 20.0 | 0.05 34 | 0.08 27 | 0.05 18 | 0.08 12 | 0.20 8 | 0.08 10 | 0.11 6 | 0.11 9 | 0.12 3 | 0.05 24 | 0.14 17 | 0.05 39 | 0.15 15 | 0.23 10 | 0.11 13 | 0.08 45 | 0.14 6 | 0.08 45 | 0.07 23 | 0.08 3 | 0.13 27 | 0.19 45 | 0.28 24 | 0.17 17 |
HAST [109] | 20.2 | 0.03 1 | 0.06 2 | 0.03 1 | 0.06 1 | 0.18 2 | 0.06 1 | 0.10 1 | 0.08 1 | 0.11 1 | 0.04 4 | 0.10 3 | 0.03 1 | 0.11 1 | 0.17 1 | 0.08 1 | 0.08 45 | 0.17 24 | 0.08 45 | 0.09 61 | 0.12 82 | 0.36 119 | 0.18 29 | 0.24 2 | 0.21 57 |
Efficient-NL [60] | 20.5 | 0.04 5 | 0.06 2 | 0.04 5 | 0.09 22 | 0.30 56 | 0.09 22 | 0.11 6 | 0.11 9 | 0.13 17 | 0.04 4 | 0.14 17 | 0.04 5 | 0.14 10 | 0.23 10 | 0.11 13 | 0.07 19 | 0.16 18 | 0.07 16 | 0.09 61 | 0.09 19 | 0.20 70 | 0.18 29 | 0.28 24 | 0.18 34 |
IROF++ [58] | 20.6 | 0.04 5 | 0.07 8 | 0.05 18 | 0.09 22 | 0.30 56 | 0.10 39 | 0.11 6 | 0.12 26 | 0.13 17 | 0.05 24 | 0.16 30 | 0.04 5 | 0.15 15 | 0.26 20 | 0.12 24 | 0.07 19 | 0.21 46 | 0.07 16 | 0.07 23 | 0.09 19 | 0.09 3 | 0.17 13 | 0.28 24 | 0.17 17 |
AGIF+OF [85] | 20.8 | 0.04 5 | 0.07 8 | 0.04 5 | 0.10 48 | 0.33 65 | 0.10 39 | 0.11 6 | 0.12 26 | 0.13 17 | 0.04 4 | 0.17 38 | 0.04 5 | 0.15 15 | 0.25 15 | 0.11 13 | 0.07 19 | 0.17 24 | 0.07 16 | 0.07 23 | 0.10 45 | 0.16 45 | 0.16 4 | 0.26 7 | 0.16 8 |
Classic+CPF [83] | 21.5 | 0.04 5 | 0.07 8 | 0.05 18 | 0.09 22 | 0.31 59 | 0.10 39 | 0.11 6 | 0.12 26 | 0.13 17 | 0.04 4 | 0.16 30 | 0.04 5 | 0.16 27 | 0.26 20 | 0.13 30 | 0.07 19 | 0.16 18 | 0.07 16 | 0.07 23 | 0.09 19 | 0.23 86 | 0.16 4 | 0.26 7 | 0.16 8 |
PH-Flow [101] | 22.7 | 0.04 5 | 0.08 27 | 0.05 18 | 0.09 22 | 0.29 50 | 0.09 22 | 0.11 6 | 0.12 26 | 0.13 17 | 0.04 4 | 0.17 38 | 0.04 5 | 0.15 15 | 0.26 20 | 0.13 30 | 0.07 19 | 0.19 37 | 0.07 16 | 0.08 44 | 0.08 3 | 0.23 86 | 0.17 13 | 0.27 14 | 0.16 8 |
Sparse-NonSparse [56] | 22.9 | 0.04 5 | 0.08 27 | 0.05 18 | 0.09 22 | 0.28 41 | 0.10 39 | 0.11 6 | 0.12 26 | 0.13 17 | 0.04 4 | 0.17 38 | 0.04 5 | 0.16 27 | 0.28 31 | 0.12 24 | 0.07 19 | 0.17 24 | 0.07 16 | 0.08 44 | 0.08 3 | 0.22 77 | 0.17 13 | 0.26 7 | 0.17 17 |
COFM [59] | 24.0 | 0.03 1 | 0.07 8 | 0.04 5 | 0.07 4 | 0.21 11 | 0.08 10 | 0.11 6 | 0.10 3 | 0.13 17 | 0.03 1 | 0.13 13 | 0.03 1 | 0.17 36 | 0.28 31 | 0.19 73 | 0.07 19 | 0.14 6 | 0.07 16 | 0.06 5 | 0.09 19 | 0.22 77 | 0.23 78 | 0.36 48 | 0.29 88 |
FC-2Layers-FF [74] | 25.6 | 0.04 5 | 0.07 8 | 0.05 18 | 0.09 22 | 0.27 37 | 0.10 39 | 0.11 6 | 0.10 3 | 0.14 47 | 0.05 24 | 0.12 9 | 0.04 5 | 0.14 10 | 0.22 5 | 0.12 24 | 0.08 45 | 0.15 13 | 0.08 45 | 0.10 81 | 0.10 45 | 0.23 86 | 0.17 13 | 0.26 7 | 0.17 17 |
LSM [39] | 25.6 | 0.04 5 | 0.07 8 | 0.05 18 | 0.09 22 | 0.28 41 | 0.10 39 | 0.11 6 | 0.11 9 | 0.13 17 | 0.05 24 | 0.16 30 | 0.04 5 | 0.15 15 | 0.27 29 | 0.13 30 | 0.08 45 | 0.17 24 | 0.08 45 | 0.09 61 | 0.09 19 | 0.23 86 | 0.17 13 | 0.26 7 | 0.17 17 |
2DHMM-SAS [92] | 26.5 | 0.04 5 | 0.08 27 | 0.05 18 | 0.09 22 | 0.31 59 | 0.09 22 | 0.11 6 | 0.12 26 | 0.14 47 | 0.04 4 | 0.17 38 | 0.04 5 | 0.16 27 | 0.28 31 | 0.13 30 | 0.07 19 | 0.21 46 | 0.07 16 | 0.08 44 | 0.08 3 | 0.23 86 | 0.17 13 | 0.28 24 | 0.17 17 |
Ramp [62] | 26.8 | 0.04 5 | 0.08 27 | 0.05 18 | 0.09 22 | 0.28 41 | 0.10 39 | 0.11 6 | 0.11 9 | 0.14 47 | 0.05 24 | 0.16 30 | 0.04 5 | 0.16 27 | 0.28 31 | 0.13 30 | 0.08 45 | 0.17 24 | 0.08 45 | 0.08 44 | 0.08 3 | 0.22 77 | 0.17 13 | 0.27 14 | 0.17 17 |
FMOF [94] | 27.3 | 0.04 5 | 0.07 8 | 0.05 18 | 0.10 48 | 0.32 62 | 0.10 39 | 0.11 6 | 0.11 9 | 0.14 47 | 0.04 4 | 0.16 30 | 0.04 5 | 0.15 15 | 0.26 20 | 0.13 30 | 0.07 19 | 0.15 13 | 0.07 16 | 0.09 61 | 0.10 45 | 0.25 100 | 0.17 13 | 0.29 34 | 0.16 8 |
PMMST [114] | 28.0 | 0.06 61 | 0.09 40 | 0.06 60 | 0.09 22 | 0.24 19 | 0.10 39 | 0.11 6 | 0.12 26 | 0.13 17 | 0.06 60 | 0.11 5 | 0.06 67 | 0.14 10 | 0.22 5 | 0.10 9 | 0.08 45 | 0.14 6 | 0.07 16 | 0.07 23 | 0.08 3 | 0.12 20 | 0.19 45 | 0.29 34 | 0.18 34 |
Classic+NL [31] | 29.2 | 0.04 5 | 0.07 8 | 0.05 18 | 0.09 22 | 0.29 50 | 0.10 39 | 0.11 6 | 0.11 9 | 0.14 47 | 0.05 24 | 0.15 24 | 0.04 5 | 0.15 15 | 0.26 20 | 0.13 30 | 0.08 45 | 0.18 32 | 0.08 45 | 0.10 81 | 0.10 45 | 0.23 86 | 0.17 13 | 0.27 14 | 0.17 17 |
ProbFlowFields [128] | 30.0 | 0.05 34 | 0.14 75 | 0.05 18 | 0.09 22 | 0.26 34 | 0.09 22 | 0.12 41 | 0.14 44 | 0.14 47 | 0.04 4 | 0.17 38 | 0.04 5 | 0.17 36 | 0.29 38 | 0.13 30 | 0.06 8 | 0.17 24 | 0.06 5 | 0.06 5 | 0.09 19 | 0.13 27 | 0.19 45 | 0.36 48 | 0.20 51 |
TV-L1-MCT [64] | 30.6 | 0.04 5 | 0.07 8 | 0.04 5 | 0.10 48 | 0.33 65 | 0.11 68 | 0.11 6 | 0.11 9 | 0.14 47 | 0.05 24 | 0.16 30 | 0.04 5 | 0.17 36 | 0.29 38 | 0.16 59 | 0.08 45 | 0.21 46 | 0.08 45 | 0.07 23 | 0.09 19 | 0.11 16 | 0.18 29 | 0.28 24 | 0.18 34 |
S2D-Matching [84] | 31.3 | 0.04 5 | 0.07 8 | 0.05 18 | 0.09 22 | 0.28 41 | 0.10 39 | 0.11 6 | 0.11 9 | 0.14 47 | 0.05 24 | 0.15 24 | 0.05 39 | 0.15 15 | 0.26 20 | 0.13 30 | 0.08 45 | 0.18 32 | 0.08 45 | 0.10 81 | 0.10 45 | 0.24 95 | 0.17 13 | 0.27 14 | 0.18 34 |
FESL [72] | 32.3 | 0.04 5 | 0.06 2 | 0.05 18 | 0.10 48 | 0.34 71 | 0.10 39 | 0.11 6 | 0.11 9 | 0.13 17 | 0.05 24 | 0.14 17 | 0.05 39 | 0.17 36 | 0.26 20 | 0.16 59 | 0.08 45 | 0.14 6 | 0.08 45 | 0.09 61 | 0.11 68 | 0.20 70 | 0.17 13 | 0.28 24 | 0.18 34 |
SimpleFlow [49] | 32.6 | 0.04 5 | 0.08 27 | 0.05 18 | 0.10 48 | 0.31 59 | 0.11 68 | 0.12 41 | 0.13 38 | 0.14 47 | 0.05 24 | 0.17 38 | 0.04 5 | 0.16 27 | 0.28 31 | 0.15 50 | 0.08 45 | 0.17 24 | 0.08 45 | 0.08 44 | 0.08 3 | 0.17 51 | 0.17 13 | 0.27 14 | 0.17 17 |
Occlusion-TV-L1 [63] | 35.3 | 0.05 34 | 0.09 40 | 0.05 18 | 0.09 22 | 0.27 37 | 0.09 22 | 0.12 41 | 0.15 48 | 0.13 17 | 0.05 24 | 0.18 48 | 0.05 39 | 0.21 57 | 0.37 61 | 0.19 73 | 0.06 8 | 0.20 42 | 0.07 16 | 0.07 23 | 0.09 19 | 0.09 3 | 0.19 45 | 0.43 66 | 0.19 44 |
Classic++ [32] | 35.6 | 0.04 5 | 0.08 27 | 0.05 18 | 0.09 22 | 0.25 29 | 0.10 39 | 0.11 6 | 0.13 38 | 0.14 47 | 0.05 24 | 0.17 38 | 0.04 5 | 0.17 36 | 0.35 52 | 0.13 30 | 0.08 45 | 0.25 62 | 0.08 45 | 0.09 61 | 0.10 45 | 0.24 95 | 0.18 29 | 0.32 39 | 0.17 17 |
Adaptive [20] | 36.9 | 0.04 5 | 0.08 27 | 0.04 5 | 0.09 22 | 0.29 50 | 0.09 22 | 0.12 41 | 0.15 48 | 0.13 17 | 0.05 24 | 0.19 52 | 0.04 5 | 0.32 108 | 0.47 89 | 0.28 100 | 0.06 8 | 0.19 37 | 0.05 3 | 0.09 61 | 0.11 68 | 0.18 57 | 0.16 4 | 0.28 24 | 0.16 8 |
SVFilterOh [111] | 38.0 | 0.05 34 | 0.07 8 | 0.05 18 | 0.10 48 | 0.25 29 | 0.10 39 | 0.12 41 | 0.11 9 | 0.13 17 | 0.06 60 | 0.20 58 | 0.05 39 | 0.15 15 | 0.22 5 | 0.11 13 | 0.09 62 | 0.16 18 | 0.08 45 | 0.12 103 | 0.16 117 | 0.32 114 | 0.16 4 | 0.26 7 | 0.16 8 |
Correlation Flow [75] | 38.2 | 0.05 34 | 0.09 40 | 0.06 60 | 0.08 12 | 0.24 19 | 0.08 10 | 0.11 6 | 0.13 38 | 0.12 3 | 0.05 24 | 0.17 38 | 0.05 39 | 0.17 36 | 0.28 31 | 0.12 24 | 0.11 79 | 0.25 62 | 0.11 77 | 0.08 44 | 0.09 19 | 0.25 100 | 0.19 45 | 0.29 34 | 0.19 44 |
MDP-Flow [26] | 38.4 | 0.05 34 | 0.11 53 | 0.06 60 | 0.09 22 | 0.24 19 | 0.10 39 | 0.12 41 | 0.14 44 | 0.13 17 | 0.05 24 | 0.21 62 | 0.05 39 | 0.19 47 | 0.32 43 | 0.16 59 | 0.07 19 | 0.25 62 | 0.07 16 | 0.07 23 | 0.10 45 | 0.11 16 | 0.19 45 | 0.40 58 | 0.18 34 |
IROF-TV [53] | 39.5 | 0.05 34 | 0.08 27 | 0.05 18 | 0.10 48 | 0.32 62 | 0.10 39 | 0.11 6 | 0.12 26 | 0.14 47 | 0.06 60 | 0.21 62 | 0.05 39 | 0.24 76 | 0.35 52 | 0.22 88 | 0.10 70 | 0.28 77 | 0.09 63 | 0.06 5 | 0.08 3 | 0.08 2 | 0.17 13 | 0.27 14 | 0.17 17 |
BriefMatch [124] | 40.2 | 0.05 34 | 0.10 46 | 0.05 18 | 0.08 12 | 0.23 18 | 0.08 10 | 0.10 1 | 0.10 3 | 0.12 3 | 0.05 24 | 0.14 17 | 0.05 39 | 0.18 44 | 0.34 49 | 0.11 13 | 0.14 97 | 0.34 88 | 0.15 102 | 0.10 81 | 0.12 82 | 0.29 109 | 0.16 4 | 0.38 53 | 0.17 17 |
AggregFlow [97] | 40.7 | 0.05 34 | 0.07 8 | 0.05 18 | 0.10 48 | 0.38 86 | 0.11 68 | 0.13 64 | 0.16 54 | 0.15 72 | 0.06 60 | 0.16 30 | 0.06 67 | 0.15 15 | 0.30 41 | 0.11 13 | 0.06 8 | 0.12 1 | 0.07 16 | 0.07 23 | 0.08 3 | 0.10 10 | 0.25 84 | 0.41 60 | 0.30 93 |
OFH [38] | 41.3 | 0.06 61 | 0.10 46 | 0.07 77 | 0.08 12 | 0.21 11 | 0.08 10 | 0.11 6 | 0.15 48 | 0.12 3 | 0.04 4 | 0.14 17 | 0.04 5 | 0.19 47 | 0.37 61 | 0.16 59 | 0.09 62 | 0.32 85 | 0.10 72 | 0.06 5 | 0.11 68 | 0.16 45 | 0.20 57 | 0.44 69 | 0.22 61 |
Aniso-Texture [82] | 42.8 | 0.04 5 | 0.08 27 | 0.04 5 | 0.10 48 | 0.21 11 | 0.11 68 | 0.11 6 | 0.12 26 | 0.14 47 | 0.04 4 | 0.18 48 | 0.04 5 | 0.19 47 | 0.32 43 | 0.13 30 | 0.10 70 | 0.18 32 | 0.10 72 | 0.15 116 | 0.15 113 | 0.27 106 | 0.18 29 | 0.28 24 | 0.19 44 |
DeepFlow2 [108] | 43.2 | 0.06 61 | 0.12 59 | 0.07 77 | 0.09 22 | 0.28 41 | 0.09 22 | 0.12 41 | 0.18 64 | 0.14 47 | 0.05 24 | 0.22 66 | 0.05 39 | 0.16 27 | 0.32 43 | 0.11 13 | 0.06 8 | 0.19 37 | 0.07 16 | 0.07 23 | 0.08 3 | 0.19 65 | 0.24 82 | 0.47 74 | 0.28 83 |
TV-L1-improved [17] | 48.4 | 0.04 5 | 0.09 40 | 0.04 5 | 0.08 12 | 0.24 19 | 0.08 10 | 0.12 41 | 0.15 48 | 0.13 17 | 0.05 24 | 0.18 48 | 0.04 5 | 0.22 64 | 0.40 72 | 0.13 30 | 0.15 104 | 0.41 103 | 0.17 111 | 0.11 95 | 0.13 92 | 0.25 100 | 0.18 29 | 0.38 53 | 0.18 34 |
PMF [73] | 48.9 | 0.05 34 | 0.08 27 | 0.05 18 | 0.10 48 | 0.25 29 | 0.10 39 | 0.13 64 | 0.14 44 | 0.15 72 | 0.06 60 | 0.14 17 | 0.06 67 | 0.17 36 | 0.29 38 | 0.13 30 | 0.09 62 | 0.30 83 | 0.09 63 | 0.14 110 | 0.15 113 | 0.30 110 | 0.16 4 | 0.25 3 | 0.15 3 |
RFlow [90] | 49.0 | 0.06 61 | 0.13 65 | 0.07 77 | 0.10 48 | 0.21 11 | 0.10 39 | 0.12 41 | 0.17 59 | 0.13 17 | 0.05 24 | 0.21 62 | 0.05 39 | 0.20 53 | 0.37 61 | 0.14 46 | 0.08 45 | 0.23 56 | 0.08 45 | 0.08 44 | 0.09 19 | 0.20 70 | 0.22 69 | 0.39 55 | 0.24 71 |
Steered-L1 [118] | 50.2 | 0.06 61 | 0.11 53 | 0.06 60 | 0.07 4 | 0.16 1 | 0.08 10 | 0.11 6 | 0.13 38 | 0.12 3 | 0.05 24 | 0.17 38 | 0.05 39 | 0.19 47 | 0.36 54 | 0.14 46 | 0.09 62 | 0.28 77 | 0.09 63 | 0.12 103 | 0.13 92 | 0.46 123 | 0.21 61 | 0.47 74 | 0.23 67 |
Kuang [131] | 52.0 | 0.05 34 | 0.17 92 | 0.05 18 | 0.09 22 | 0.43 98 | 0.09 22 | 0.13 64 | 0.29 100 | 0.14 47 | 0.05 24 | 0.28 78 | 0.05 39 | 0.28 94 | 0.59 108 | 0.19 73 | 0.07 19 | 0.34 88 | 0.07 16 | 0.06 5 | 0.09 19 | 0.13 27 | 0.19 45 | 0.46 72 | 0.19 44 |
S2F-IF [123] | 52.5 | 0.05 34 | 0.18 97 | 0.05 18 | 0.10 48 | 0.40 93 | 0.10 39 | 0.13 64 | 0.26 92 | 0.15 72 | 0.05 24 | 0.30 83 | 0.04 5 | 0.26 88 | 0.52 100 | 0.19 73 | 0.06 8 | 0.24 59 | 0.06 5 | 0.06 5 | 0.09 19 | 0.13 27 | 0.21 61 | 0.54 85 | 0.22 61 |
PGM-C [120] | 54.5 | 0.05 34 | 0.17 92 | 0.05 18 | 0.11 70 | 0.39 89 | 0.11 68 | 0.13 64 | 0.24 85 | 0.15 72 | 0.05 24 | 0.31 86 | 0.05 39 | 0.24 76 | 0.48 93 | 0.19 73 | 0.06 8 | 0.20 42 | 0.06 5 | 0.06 5 | 0.09 19 | 0.13 27 | 0.21 61 | 0.56 91 | 0.23 67 |
DeepFlow [86] | 54.8 | 0.06 61 | 0.13 65 | 0.09 99 | 0.10 48 | 0.29 50 | 0.10 39 | 0.12 41 | 0.20 73 | 0.15 72 | 0.06 60 | 0.23 70 | 0.06 67 | 0.17 36 | 0.34 49 | 0.12 24 | 0.07 19 | 0.23 56 | 0.07 16 | 0.07 23 | 0.08 3 | 0.19 65 | 0.26 93 | 0.55 88 | 0.33 97 |
CPM-Flow [116] | 54.8 | 0.05 34 | 0.17 92 | 0.05 18 | 0.11 70 | 0.39 89 | 0.11 68 | 0.13 64 | 0.23 84 | 0.15 72 | 0.05 24 | 0.31 86 | 0.05 39 | 0.24 76 | 0.48 93 | 0.19 73 | 0.06 8 | 0.19 37 | 0.06 5 | 0.06 5 | 0.09 19 | 0.13 27 | 0.22 69 | 0.57 96 | 0.23 67 |
EpicFlow [102] | 55.2 | 0.05 34 | 0.17 92 | 0.05 18 | 0.11 70 | 0.40 93 | 0.11 68 | 0.13 64 | 0.24 85 | 0.15 72 | 0.05 24 | 0.31 86 | 0.05 39 | 0.24 76 | 0.48 93 | 0.19 73 | 0.06 8 | 0.21 46 | 0.06 5 | 0.06 5 | 0.09 19 | 0.13 27 | 0.22 69 | 0.56 91 | 0.23 67 |
Sparse Occlusion [54] | 55.8 | 0.06 61 | 0.10 46 | 0.05 18 | 0.11 70 | 0.28 41 | 0.12 81 | 0.12 41 | 0.15 48 | 0.13 17 | 0.06 60 | 0.19 52 | 0.05 39 | 0.21 57 | 0.33 47 | 0.15 50 | 0.11 79 | 0.22 51 | 0.09 63 | 0.17 124 | 0.16 117 | 0.22 77 | 0.19 45 | 0.32 39 | 0.17 17 |
TF+OM [100] | 57.1 | 0.06 61 | 0.09 40 | 0.05 18 | 0.11 70 | 0.19 4 | 0.12 81 | 0.12 41 | 0.12 26 | 0.15 72 | 0.08 86 | 0.14 17 | 0.08 87 | 0.16 27 | 0.27 29 | 0.15 50 | 0.11 79 | 0.18 32 | 0.12 85 | 0.10 81 | 0.13 92 | 0.22 77 | 0.22 69 | 0.45 70 | 0.25 76 |
Second-order prior [8] | 57.5 | 0.05 34 | 0.13 65 | 0.06 60 | 0.09 22 | 0.34 71 | 0.08 10 | 0.13 64 | 0.27 95 | 0.14 47 | 0.04 4 | 0.15 24 | 0.04 5 | 0.25 85 | 0.49 97 | 0.13 30 | 0.10 70 | 0.58 113 | 0.08 45 | 0.12 103 | 0.12 82 | 0.25 100 | 0.19 45 | 0.47 74 | 0.18 34 |
MLDP_OF [89] | 58.0 | 0.07 86 | 0.15 80 | 0.07 77 | 0.10 48 | 0.27 37 | 0.10 39 | 0.12 41 | 0.17 59 | 0.13 17 | 0.06 60 | 0.19 52 | 0.05 39 | 0.23 70 | 0.37 61 | 0.15 50 | 0.11 79 | 0.24 59 | 0.12 85 | 0.09 61 | 0.11 68 | 0.28 107 | 0.18 29 | 0.30 37 | 0.20 51 |
FlowFields+ [130] | 58.4 | 0.05 34 | 0.19 102 | 0.05 18 | 0.11 70 | 0.42 96 | 0.11 68 | 0.14 86 | 0.29 100 | 0.15 72 | 0.05 24 | 0.32 89 | 0.04 5 | 0.27 91 | 0.54 103 | 0.21 83 | 0.05 3 | 0.27 70 | 0.06 5 | 0.06 5 | 0.10 45 | 0.12 20 | 0.21 61 | 0.56 91 | 0.22 61 |
FlowFields [110] | 59.3 | 0.05 34 | 0.18 97 | 0.05 18 | 0.11 70 | 0.42 96 | 0.11 68 | 0.14 86 | 0.28 97 | 0.15 72 | 0.05 24 | 0.32 89 | 0.05 39 | 0.27 91 | 0.54 103 | 0.21 83 | 0.05 3 | 0.27 70 | 0.06 5 | 0.06 5 | 0.10 45 | 0.12 20 | 0.21 61 | 0.56 91 | 0.21 57 |
TriangleFlow [30] | 59.5 | 0.05 34 | 0.10 46 | 0.06 60 | 0.10 48 | 0.33 65 | 0.09 22 | 0.13 64 | 0.19 69 | 0.13 17 | 0.04 4 | 0.15 24 | 0.04 5 | 0.31 105 | 0.58 107 | 0.22 88 | 0.14 97 | 0.37 95 | 0.15 102 | 0.08 44 | 0.13 92 | 0.16 45 | 0.21 61 | 0.42 62 | 0.24 71 |
Rannacher [23] | 61.5 | 0.06 61 | 0.12 59 | 0.07 77 | 0.11 70 | 0.30 56 | 0.11 68 | 0.13 64 | 0.19 69 | 0.15 72 | 0.06 60 | 0.27 76 | 0.05 39 | 0.22 64 | 0.41 76 | 0.15 50 | 0.11 79 | 0.34 88 | 0.10 72 | 0.09 61 | 0.10 45 | 0.18 57 | 0.18 29 | 0.37 51 | 0.18 34 |
CostFilter [40] | 61.7 | 0.06 61 | 0.10 46 | 0.06 60 | 0.11 70 | 0.27 37 | 0.11 68 | 0.13 64 | 0.16 54 | 0.15 72 | 0.08 86 | 0.15 24 | 0.09 93 | 0.18 44 | 0.32 43 | 0.13 30 | 0.10 70 | 0.36 93 | 0.10 72 | 0.14 110 | 0.17 122 | 0.33 117 | 0.15 3 | 0.32 39 | 0.15 3 |
LDOF [28] | 61.8 | 0.06 61 | 0.14 75 | 0.07 77 | 0.10 48 | 0.36 79 | 0.10 39 | 0.13 64 | 0.25 90 | 0.14 47 | 0.06 60 | 0.33 91 | 0.05 39 | 0.22 64 | 0.41 76 | 0.22 88 | 0.07 19 | 0.25 62 | 0.07 16 | 0.07 23 | 0.10 45 | 0.13 27 | 0.26 93 | 0.62 101 | 0.35 98 |
FlowNetS+ft+v [112] | 61.8 | 0.05 34 | 0.12 59 | 0.06 60 | 0.10 48 | 0.33 65 | 0.10 39 | 0.13 64 | 0.22 81 | 0.16 85 | 0.05 24 | 0.27 76 | 0.05 39 | 0.25 85 | 0.46 88 | 0.20 80 | 0.07 19 | 0.27 70 | 0.07 16 | 0.10 81 | 0.12 82 | 0.18 57 | 0.22 69 | 0.52 82 | 0.27 80 |
ComplOF-FED-GPU [35] | 61.8 | 0.07 86 | 0.15 80 | 0.08 90 | 0.08 12 | 0.26 34 | 0.08 10 | 0.12 41 | 0.18 64 | 0.12 3 | 0.07 78 | 0.15 24 | 0.07 82 | 0.22 64 | 0.45 84 | 0.16 59 | 0.12 90 | 0.40 100 | 0.12 85 | 0.09 61 | 0.10 45 | 0.23 86 | 0.21 61 | 0.47 74 | 0.24 71 |
ACK-Prior [27] | 62.0 | 0.07 86 | 0.11 53 | 0.07 77 | 0.09 22 | 0.22 16 | 0.09 22 | 0.12 41 | 0.13 38 | 0.13 17 | 0.07 78 | 0.19 52 | 0.06 67 | 0.20 53 | 0.34 49 | 0.14 46 | 0.13 94 | 0.29 81 | 0.12 85 | 0.11 95 | 0.11 68 | 0.42 122 | 0.25 84 | 0.40 58 | 0.28 83 |
CombBMOF [113] | 62.0 | 0.06 61 | 0.14 75 | 0.05 18 | 0.11 70 | 0.35 78 | 0.10 39 | 0.12 41 | 0.16 54 | 0.13 17 | 0.06 60 | 0.22 66 | 0.06 67 | 0.24 76 | 0.39 68 | 0.14 46 | 0.14 97 | 0.33 86 | 0.15 102 | 0.11 95 | 0.13 92 | 0.23 86 | 0.18 29 | 0.35 47 | 0.17 17 |
Complementary OF [21] | 62.1 | 0.07 86 | 0.15 80 | 0.08 90 | 0.09 22 | 0.21 11 | 0.09 22 | 0.12 41 | 0.16 54 | 0.14 47 | 0.08 86 | 0.16 30 | 0.08 87 | 0.20 53 | 0.38 65 | 0.17 68 | 0.11 79 | 0.33 86 | 0.11 77 | 0.07 23 | 0.10 45 | 0.18 57 | 0.26 93 | 0.56 91 | 0.35 98 |
TCOF [69] | 63.2 | 0.06 61 | 0.12 59 | 0.07 77 | 0.12 85 | 0.34 71 | 0.12 81 | 0.14 86 | 0.21 78 | 0.16 85 | 0.09 95 | 0.19 52 | 0.10 96 | 0.24 76 | 0.45 84 | 0.12 24 | 0.07 19 | 0.13 3 | 0.08 45 | 0.11 95 | 0.12 82 | 0.12 20 | 0.20 57 | 0.37 51 | 0.18 34 |
EPPM w/o HM [88] | 63.4 | 0.06 61 | 0.16 87 | 0.06 60 | 0.10 48 | 0.34 71 | 0.09 22 | 0.13 64 | 0.22 81 | 0.13 17 | 0.06 60 | 0.20 58 | 0.07 82 | 0.23 70 | 0.36 54 | 0.16 59 | 0.12 90 | 0.40 100 | 0.12 85 | 0.09 61 | 0.12 82 | 0.37 120 | 0.18 29 | 0.34 43 | 0.17 17 |
Aniso. Huber-L1 [22] | 63.6 | 0.05 34 | 0.11 53 | 0.05 18 | 0.15 93 | 0.38 86 | 0.17 94 | 0.13 64 | 0.20 73 | 0.17 89 | 0.06 60 | 0.33 91 | 0.06 67 | 0.21 57 | 0.36 54 | 0.16 59 | 0.09 62 | 0.22 51 | 0.09 63 | 0.11 95 | 0.11 68 | 0.18 57 | 0.19 45 | 0.34 43 | 0.20 51 |
F-TV-L1 [15] | 64.5 | 0.09 100 | 0.17 92 | 0.12 107 | 0.12 85 | 0.34 71 | 0.13 88 | 0.12 41 | 0.18 64 | 0.14 47 | 0.08 86 | 0.24 71 | 0.08 87 | 0.26 88 | 0.42 80 | 0.21 83 | 0.08 45 | 0.25 62 | 0.08 45 | 0.08 44 | 0.10 45 | 0.18 57 | 0.17 13 | 0.31 38 | 0.16 8 |
Brox et al. [5] | 65.1 | 0.06 61 | 0.15 80 | 0.08 90 | 0.11 70 | 0.33 65 | 0.12 81 | 0.13 64 | 0.21 78 | 0.14 47 | 0.05 24 | 0.28 78 | 0.05 39 | 0.28 94 | 0.44 81 | 0.43 113 | 0.07 19 | 0.27 70 | 0.07 16 | 0.08 44 | 0.10 45 | 0.09 3 | 0.26 93 | 0.63 102 | 0.39 105 |
CRTflow [80] | 66.5 | 0.06 61 | 0.14 75 | 0.06 60 | 0.10 48 | 0.29 50 | 0.10 39 | 0.12 41 | 0.20 73 | 0.13 17 | 0.06 60 | 0.21 62 | 0.06 67 | 0.20 53 | 0.38 65 | 0.15 50 | 0.24 121 | 0.50 108 | 0.27 119 | 0.08 44 | 0.11 68 | 0.17 51 | 0.25 84 | 0.54 85 | 0.32 95 |
LocallyOriented [52] | 67.1 | 0.05 34 | 0.10 46 | 0.05 18 | 0.12 85 | 0.46 100 | 0.12 81 | 0.14 86 | 0.24 85 | 0.16 85 | 0.07 78 | 0.20 58 | 0.06 67 | 0.23 70 | 0.44 81 | 0.17 68 | 0.08 45 | 0.22 51 | 0.09 63 | 0.09 61 | 0.11 68 | 0.17 51 | 0.23 78 | 0.46 72 | 0.26 79 |
SIOF [67] | 67.5 | 0.07 86 | 0.11 53 | 0.07 77 | 0.10 48 | 0.28 41 | 0.10 39 | 0.15 97 | 0.17 59 | 0.20 97 | 0.08 86 | 0.19 52 | 0.09 93 | 0.24 76 | 0.41 76 | 0.21 83 | 0.11 79 | 0.22 51 | 0.11 77 | 0.09 61 | 0.09 19 | 0.14 38 | 0.25 84 | 0.43 66 | 0.28 83 |
DPOF [18] | 68.0 | 0.06 61 | 0.16 87 | 0.05 18 | 0.11 70 | 0.37 82 | 0.10 39 | 0.13 64 | 0.20 73 | 0.15 72 | 0.07 78 | 0.28 78 | 0.06 67 | 0.21 57 | 0.41 76 | 0.15 50 | 0.09 62 | 0.25 62 | 0.09 63 | 0.09 61 | 0.10 45 | 0.48 124 | 0.25 84 | 0.45 70 | 0.29 88 |
NL-TV-NCC [25] | 68.3 | 0.06 61 | 0.11 53 | 0.06 60 | 0.11 70 | 0.34 71 | 0.10 39 | 0.12 41 | 0.15 48 | 0.13 17 | 0.07 78 | 0.22 66 | 0.06 67 | 0.27 91 | 0.45 84 | 0.15 50 | 0.15 104 | 0.36 93 | 0.13 95 | 0.10 81 | 0.14 103 | 0.22 77 | 0.22 69 | 0.41 60 | 0.22 61 |
DF-Auto [115] | 69.6 | 0.06 61 | 0.13 65 | 0.05 18 | 0.20 99 | 0.49 102 | 0.24 99 | 0.14 86 | 0.26 92 | 0.21 99 | 0.08 86 | 0.33 91 | 0.08 87 | 0.21 57 | 0.36 54 | 0.24 94 | 0.05 3 | 0.13 3 | 0.06 5 | 0.10 81 | 0.13 92 | 0.09 3 | 0.28 102 | 0.54 85 | 0.40 107 |
CBF [12] | 70.1 | 0.06 61 | 0.13 65 | 0.07 77 | 0.19 97 | 0.36 79 | 0.25 101 | 0.12 41 | 0.17 59 | 0.14 47 | 0.05 24 | 0.24 71 | 0.05 39 | 0.23 70 | 0.39 68 | 0.18 70 | 0.10 70 | 0.22 51 | 0.11 77 | 0.14 110 | 0.14 103 | 0.25 100 | 0.22 69 | 0.42 62 | 0.24 71 |
SRR-TVOF-NL [91] | 70.1 | 0.06 61 | 0.13 65 | 0.06 60 | 0.10 48 | 0.33 65 | 0.10 39 | 0.13 64 | 0.22 81 | 0.14 47 | 0.06 60 | 0.25 73 | 0.05 39 | 0.24 76 | 0.40 72 | 0.15 50 | 0.10 70 | 0.29 81 | 0.09 63 | 0.15 116 | 0.14 103 | 0.24 95 | 0.27 99 | 0.42 62 | 0.31 94 |
Dynamic MRF [7] | 70.3 | 0.07 86 | 0.16 87 | 0.08 90 | 0.09 22 | 0.28 41 | 0.09 22 | 0.12 41 | 0.19 69 | 0.14 47 | 0.06 60 | 0.26 74 | 0.06 67 | 0.28 94 | 0.53 101 | 0.22 88 | 0.13 94 | 0.56 112 | 0.14 98 | 0.08 44 | 0.09 19 | 0.24 95 | 0.22 69 | 0.55 88 | 0.27 80 |
Local-TV-L1 [65] | 70.4 | 0.08 98 | 0.15 80 | 0.11 103 | 0.19 97 | 0.39 89 | 0.22 97 | 0.14 86 | 0.25 90 | 0.17 89 | 0.08 86 | 0.39 98 | 0.09 93 | 0.18 44 | 0.33 47 | 0.13 30 | 0.07 19 | 0.21 46 | 0.07 16 | 0.06 5 | 0.08 3 | 0.18 57 | 0.28 102 | 0.63 102 | 0.49 112 |
SuperFlow [81] | 71.1 | 0.05 34 | 0.12 59 | 0.05 18 | 0.15 93 | 0.37 82 | 0.18 95 | 0.13 64 | 0.19 69 | 0.19 95 | 0.10 97 | 0.37 97 | 0.11 98 | 0.23 70 | 0.36 54 | 0.41 111 | 0.06 8 | 0.26 68 | 0.07 16 | 0.10 81 | 0.11 68 | 0.15 41 | 0.26 93 | 0.58 97 | 0.35 98 |
CNN-flow-warp+ref [117] | 71.6 | 0.06 61 | 0.16 87 | 0.07 77 | 0.13 89 | 0.38 86 | 0.15 90 | 0.14 86 | 0.26 92 | 0.18 94 | 0.06 60 | 0.42 99 | 0.06 67 | 0.26 88 | 0.45 84 | 0.35 107 | 0.07 19 | 0.27 70 | 0.07 16 | 0.06 5 | 0.09 19 | 0.10 10 | 0.28 102 | 0.66 107 | 0.37 103 |
CLG-TV [48] | 73.8 | 0.06 61 | 0.12 59 | 0.06 60 | 0.17 96 | 0.36 79 | 0.21 96 | 0.14 86 | 0.20 73 | 0.19 95 | 0.08 86 | 0.47 105 | 0.08 87 | 0.23 70 | 0.40 72 | 0.21 83 | 0.10 70 | 0.27 70 | 0.11 77 | 0.10 81 | 0.11 68 | 0.13 27 | 0.20 57 | 0.39 55 | 0.21 57 |
Fusion [6] | 73.8 | 0.05 34 | 0.18 97 | 0.06 60 | 0.09 22 | 0.29 50 | 0.09 22 | 0.13 64 | 0.18 64 | 0.14 47 | 0.05 24 | 0.28 78 | 0.05 39 | 0.29 99 | 0.44 81 | 0.30 103 | 0.14 97 | 0.38 97 | 0.15 102 | 0.14 110 | 0.15 113 | 0.19 65 | 0.34 114 | 0.51 81 | 0.41 108 |
Bartels [41] | 74.4 | 0.07 86 | 0.10 46 | 0.08 90 | 0.12 85 | 0.24 19 | 0.14 89 | 0.13 64 | 0.14 44 | 0.17 89 | 0.10 97 | 0.20 58 | 0.11 98 | 0.21 57 | 0.39 68 | 0.20 80 | 0.11 79 | 0.26 68 | 0.16 107 | 0.11 95 | 0.13 92 | 0.32 114 | 0.19 45 | 0.39 55 | 0.22 61 |
ROF-ND [107] | 74.9 | 0.07 86 | 0.13 65 | 0.07 77 | 0.13 89 | 0.26 34 | 0.12 81 | 0.12 41 | 0.17 59 | 0.13 17 | 0.08 86 | 0.30 83 | 0.07 82 | 0.22 64 | 0.38 65 | 0.18 70 | 0.16 109 | 0.30 83 | 0.12 85 | 0.14 110 | 0.14 103 | 0.30 110 | 0.22 69 | 0.48 78 | 0.20 51 |
HBM-GC [105] | 78.4 | 0.10 104 | 0.13 65 | 0.10 101 | 0.16 95 | 0.28 41 | 0.16 92 | 0.15 97 | 0.16 54 | 0.17 89 | 0.12 100 | 0.22 66 | 0.12 100 | 0.19 47 | 0.25 15 | 0.16 59 | 0.14 97 | 0.19 37 | 0.13 95 | 0.15 116 | 0.16 117 | 0.30 110 | 0.25 84 | 0.28 24 | 0.25 76 |
Learning Flow [11] | 78.5 | 0.06 61 | 0.13 65 | 0.06 60 | 0.14 91 | 0.37 82 | 0.15 90 | 0.13 64 | 0.21 78 | 0.17 89 | 0.07 78 | 0.34 94 | 0.06 67 | 0.28 94 | 0.55 105 | 0.27 98 | 0.11 79 | 0.38 97 | 0.11 77 | 0.10 81 | 0.12 82 | 0.17 51 | 0.21 61 | 0.50 79 | 0.22 61 |
p-harmonic [29] | 79.6 | 0.07 86 | 0.20 103 | 0.08 90 | 0.11 70 | 0.34 71 | 0.11 68 | 0.14 86 | 0.28 97 | 0.16 85 | 0.06 60 | 0.46 104 | 0.06 67 | 0.30 100 | 0.47 89 | 0.29 101 | 0.11 79 | 0.40 100 | 0.11 77 | 0.10 81 | 0.12 82 | 0.17 51 | 0.20 57 | 0.42 62 | 0.19 44 |
TriFlow [95] | 85.3 | 0.07 86 | 0.13 65 | 0.06 60 | 0.14 91 | 0.32 62 | 0.16 92 | 0.14 86 | 0.18 64 | 0.20 97 | 0.10 97 | 0.28 78 | 0.10 96 | 0.22 64 | 0.36 54 | 0.24 94 | 0.11 79 | 0.20 42 | 0.11 77 | 0.18 125 | 0.14 103 | 1.18 128 | 0.30 107 | 0.58 97 | 0.37 103 |
StereoOF-V1MT [119] | 86.1 | 0.07 86 | 0.24 109 | 0.08 90 | 0.09 22 | 0.56 106 | 0.09 22 | 0.15 97 | 0.49 111 | 0.14 47 | 0.07 78 | 0.47 105 | 0.07 82 | 0.43 113 | 0.82 117 | 0.33 104 | 0.13 94 | 0.91 119 | 0.13 95 | 0.08 44 | 0.11 68 | 0.19 65 | 0.26 93 | 0.95 117 | 0.28 83 |
Shiralkar [42] | 86.7 | 0.06 61 | 0.26 113 | 0.08 90 | 0.09 22 | 0.40 93 | 0.09 22 | 0.14 86 | 0.37 105 | 0.14 47 | 0.07 78 | 0.42 99 | 0.07 82 | 0.30 100 | 0.62 109 | 0.20 80 | 0.18 114 | 0.83 115 | 0.17 111 | 0.10 81 | 0.12 82 | 0.33 117 | 0.24 82 | 0.80 112 | 0.27 80 |
SegOF [10] | 87.5 | 0.08 98 | 0.15 80 | 0.06 60 | 0.36 113 | 0.64 111 | 0.43 115 | 0.18 104 | 0.34 104 | 0.35 114 | 0.15 106 | 0.42 99 | 0.13 101 | 0.42 112 | 0.68 112 | 0.72 119 | 0.10 70 | 0.50 108 | 0.12 85 | 0.06 5 | 0.10 45 | 0.09 3 | 0.23 78 | 0.55 88 | 0.24 71 |
Ad-TV-NDC [36] | 88.8 | 0.12 112 | 0.20 103 | 0.20 121 | 0.34 112 | 0.57 109 | 0.41 114 | 0.20 109 | 0.38 106 | 0.27 106 | 0.14 104 | 0.48 108 | 0.16 106 | 0.21 57 | 0.36 54 | 0.16 59 | 0.10 70 | 0.23 56 | 0.10 72 | 0.08 44 | 0.09 19 | 0.16 45 | 0.34 114 | 0.80 112 | 0.81 119 |
Modified CLG [34] | 91.2 | 0.10 104 | 0.23 107 | 0.10 101 | 0.29 109 | 0.56 106 | 0.37 110 | 0.20 109 | 0.52 112 | 0.29 111 | 0.19 113 | 0.72 112 | 0.20 111 | 0.28 94 | 0.53 101 | 0.33 104 | 0.07 19 | 0.37 95 | 0.08 45 | 0.07 23 | 0.10 45 | 0.11 16 | 0.33 111 | 0.83 114 | 0.63 116 |
StereoFlow [44] | 92.5 | 0.30 128 | 0.57 129 | 0.36 127 | 1.03 128 | 1.75 129 | 0.92 125 | 0.81 128 | 1.43 129 | 0.51 123 | 1.05 128 | 2.03 127 | 0.92 127 | 0.53 118 | 0.72 113 | 0.44 114 | 0.04 1 | 0.14 6 | 0.04 1 | 0.04 1 | 0.10 45 | 0.06 1 | 0.27 99 | 0.60 99 | 0.32 95 |
BlockOverlap [61] | 93.4 | 0.12 112 | 0.16 87 | 0.12 107 | 0.25 106 | 0.39 89 | 0.29 107 | 0.19 108 | 0.24 85 | 0.25 103 | 0.16 107 | 0.35 95 | 0.17 109 | 0.19 47 | 0.28 31 | 0.18 70 | 0.15 104 | 0.24 59 | 0.14 98 | 0.15 116 | 0.15 113 | 0.37 120 | 0.25 84 | 0.50 79 | 0.39 105 |
IAOF2 [51] | 95.6 | 0.07 86 | 0.14 75 | 0.07 77 | 0.21 100 | 0.43 98 | 0.25 101 | 0.17 100 | 0.24 85 | 0.22 100 | 0.40 119 | 0.75 114 | 0.66 123 | 0.31 105 | 0.47 89 | 0.27 98 | 0.15 104 | 0.34 88 | 0.16 107 | 0.14 110 | 0.13 92 | 0.20 70 | 0.25 84 | 0.52 82 | 0.29 88 |
HBpMotionGpu [43] | 97.8 | 0.09 100 | 0.15 80 | 0.09 99 | 0.32 111 | 0.49 102 | 0.38 112 | 0.17 100 | 0.32 103 | 0.28 109 | 0.14 104 | 0.35 95 | 0.14 103 | 0.25 85 | 0.40 72 | 0.22 88 | 0.14 97 | 0.27 70 | 0.14 98 | 0.15 116 | 0.13 92 | 0.24 95 | 0.29 106 | 0.60 99 | 0.44 110 |
SPSA-learn [13] | 98.5 | 0.10 104 | 0.26 113 | 0.12 107 | 0.24 105 | 0.50 104 | 0.28 106 | 0.18 104 | 0.40 107 | 0.27 106 | 0.12 100 | 0.55 110 | 0.14 103 | 0.30 100 | 0.47 89 | 0.38 109 | 0.12 90 | 0.41 103 | 0.14 98 | 0.09 61 | 0.11 68 | 0.15 41 | 0.34 114 | 0.64 105 | 0.64 117 |
Filter Flow [19] | 99.0 | 0.09 100 | 0.18 97 | 0.08 90 | 0.22 102 | 0.53 105 | 0.24 99 | 0.17 100 | 0.28 97 | 0.24 102 | 0.16 107 | 0.47 105 | 0.16 106 | 0.32 108 | 0.48 93 | 0.38 109 | 0.17 112 | 0.34 88 | 0.16 107 | 0.18 125 | 0.19 126 | 0.22 77 | 0.23 78 | 0.43 66 | 0.25 76 |
GroupFlow [9] | 99.3 | 0.10 104 | 0.25 111 | 0.14 114 | 0.40 115 | 1.06 120 | 0.44 117 | 0.23 114 | 0.80 118 | 0.37 116 | 0.12 100 | 0.45 102 | 0.13 101 | 0.52 117 | 1.05 123 | 0.24 94 | 0.22 119 | 0.89 118 | 0.25 118 | 0.06 5 | 0.09 19 | 0.09 3 | 0.31 110 | 0.93 116 | 0.43 109 |
Black & Anandan [4] | 99.4 | 0.10 104 | 0.25 111 | 0.15 115 | 0.23 103 | 0.56 106 | 0.26 103 | 0.18 104 | 0.45 109 | 0.26 105 | 0.13 103 | 0.68 111 | 0.15 105 | 0.31 105 | 0.49 97 | 0.34 106 | 0.12 90 | 0.49 107 | 0.12 85 | 0.11 95 | 0.13 92 | 0.10 10 | 0.30 107 | 0.63 102 | 0.48 111 |
2D-CLG [1] | 99.8 | 0.12 112 | 0.28 115 | 0.11 103 | 0.44 119 | 0.72 113 | 0.55 121 | 0.30 118 | 0.74 114 | 0.40 117 | 0.48 122 | 1.17 119 | 0.62 122 | 0.38 110 | 0.67 111 | 0.57 115 | 0.09 62 | 0.51 110 | 0.12 85 | 0.06 5 | 0.09 19 | 0.12 20 | 0.44 120 | 1.04 120 | 0.88 122 |
IAOF [50] | 101.2 | 0.09 100 | 0.18 97 | 0.12 107 | 0.27 108 | 0.48 101 | 0.34 109 | 0.18 104 | 0.44 108 | 0.25 103 | 0.18 111 | 0.53 109 | 0.25 115 | 0.30 100 | 0.50 99 | 0.26 97 | 0.14 97 | 0.53 111 | 0.12 85 | 0.12 103 | 0.11 68 | 0.20 70 | 0.30 107 | 0.65 106 | 0.60 115 |
2bit-BM-tele [98] | 101.4 | 0.15 120 | 0.20 103 | 0.19 120 | 0.23 103 | 0.37 82 | 0.26 103 | 0.20 109 | 0.27 95 | 0.23 101 | 0.18 111 | 0.30 83 | 0.19 110 | 0.24 76 | 0.39 68 | 0.22 88 | 0.20 116 | 0.38 97 | 0.23 116 | 0.16 121 | 0.17 122 | 0.50 126 | 0.25 84 | 0.52 82 | 0.35 98 |
GraphCuts [14] | 102.5 | 0.10 104 | 0.20 103 | 0.11 103 | 0.21 100 | 0.64 111 | 0.22 97 | 0.17 100 | 0.30 102 | 0.27 106 | 0.09 95 | 0.45 102 | 0.08 87 | 0.30 100 | 0.55 105 | 0.29 101 | 0.16 109 | 0.28 77 | 0.16 107 | 0.13 109 | 0.13 92 | 0.30 110 | 0.37 117 | 0.69 109 | 0.56 113 |
SILK [79] | 105.8 | 0.13 117 | 0.31 116 | 0.21 122 | 0.37 114 | 0.82 114 | 0.43 115 | 0.22 112 | 0.77 116 | 0.31 112 | 0.20 114 | 0.73 113 | 0.22 113 | 0.53 118 | 0.86 118 | 0.79 122 | 0.21 117 | 0.97 120 | 0.21 115 | 0.06 5 | 0.10 45 | 0.15 41 | 0.44 120 | 1.03 118 | 0.86 121 |
Nguyen [33] | 107.7 | 0.12 112 | 0.23 107 | 0.13 111 | 0.60 123 | 0.63 110 | 0.87 124 | 0.26 116 | 0.61 113 | 0.36 115 | 0.40 119 | 0.84 116 | 0.51 120 | 0.40 111 | 0.63 110 | 0.59 116 | 0.17 112 | 0.48 106 | 0.20 114 | 0.10 81 | 0.10 45 | 0.16 45 | 0.40 119 | 0.84 115 | 1.04 124 |
Horn & Schunck [3] | 108.8 | 0.11 110 | 0.35 118 | 0.16 117 | 0.26 107 | 0.87 116 | 0.27 105 | 0.22 112 | 0.83 120 | 0.28 109 | 0.17 110 | 0.87 117 | 0.22 113 | 0.48 115 | 0.78 116 | 0.70 118 | 0.15 104 | 1.07 122 | 0.15 102 | 0.12 103 | 0.14 103 | 0.10 10 | 0.47 122 | 1.26 123 | 0.84 120 |
FlowNet2 [122] | 109.3 | 0.13 117 | 0.24 109 | 0.13 111 | 0.40 115 | 0.83 115 | 0.44 117 | 0.27 117 | 0.46 110 | 0.45 120 | 0.16 107 | 0.26 74 | 0.16 106 | 0.47 114 | 0.75 115 | 0.35 107 | 0.18 114 | 0.42 105 | 0.19 113 | 0.16 121 | 0.18 124 | 0.25 100 | 0.28 102 | 0.66 107 | 0.28 83 |
Periodicity [78] | 112.0 | 0.14 119 | 0.32 117 | 0.11 103 | 0.29 109 | 1.16 122 | 0.31 108 | 0.51 125 | 0.74 114 | 0.66 125 | 0.49 123 | 1.53 123 | 0.43 118 | 1.23 129 | 2.67 129 | 0.95 125 | 0.41 125 | 3.18 131 | 0.37 124 | 0.08 44 | 0.14 103 | 0.09 3 | 0.48 123 | 2.09 128 | 0.79 118 |
UnFlow [129] | 112.2 | 0.23 126 | 0.39 121 | 0.16 117 | 0.53 122 | 1.01 118 | 0.54 120 | 0.46 123 | 1.10 124 | 0.46 121 | 0.21 115 | 0.79 115 | 0.20 111 | 0.77 124 | 1.12 125 | 0.85 123 | 0.27 122 | 0.87 117 | 0.28 120 | 0.12 103 | 0.14 103 | 0.11 16 | 0.27 99 | 0.77 111 | 0.35 98 |
TI-DOFE [24] | 114.0 | 0.22 125 | 0.42 124 | 0.38 128 | 0.80 126 | 1.16 122 | 0.99 126 | 0.46 123 | 1.15 128 | 0.59 124 | 0.72 125 | 1.38 122 | 0.96 128 | 0.53 118 | 0.86 118 | 0.76 121 | 0.16 109 | 1.12 123 | 0.23 116 | 0.09 61 | 0.12 82 | 0.10 10 | 0.75 126 | 1.45 126 | 1.48 126 |
SLK [47] | 114.1 | 0.11 110 | 0.50 128 | 0.17 119 | 0.77 125 | 1.27 126 | 1.02 127 | 0.30 118 | 1.11 125 | 0.44 119 | 1.08 129 | 1.23 121 | 1.28 129 | 0.77 124 | 1.07 124 | 1.27 127 | 0.21 117 | 1.22 124 | 0.28 120 | 0.07 23 | 0.14 103 | 0.12 20 | 0.80 128 | 1.44 125 | 1.95 127 |
Heeger++ [104] | 114.5 | 0.21 124 | 0.46 126 | 0.13 111 | 0.47 120 | 1.50 128 | 0.38 112 | 0.76 127 | 0.99 123 | 0.79 127 | 0.76 126 | 1.54 124 | 0.72 125 | 0.89 127 | 1.26 127 | 1.06 126 | 0.68 129 | 1.77 128 | 0.68 129 | 0.08 44 | 0.16 117 | 0.13 27 | 0.33 111 | 1.07 121 | 0.29 88 |
FOLKI [16] | 117.8 | 0.12 112 | 0.43 125 | 0.22 124 | 0.47 120 | 1.13 121 | 0.69 122 | 0.24 115 | 1.11 125 | 0.32 113 | 0.22 116 | 1.12 118 | 0.29 116 | 0.58 121 | 0.97 121 | 0.90 124 | 0.22 119 | 1.29 126 | 0.32 123 | 0.09 61 | 0.16 117 | 0.28 107 | 0.68 125 | 1.55 127 | 2.32 128 |
FFV1MT [106] | 117.8 | 0.18 122 | 0.39 121 | 0.15 115 | 0.41 117 | 1.31 127 | 0.37 110 | 0.81 128 | 1.12 127 | 1.00 128 | 0.77 127 | 2.09 128 | 0.75 126 | 0.87 126 | 1.24 126 | 1.40 129 | 0.64 128 | 1.64 127 | 0.64 128 | 0.11 95 | 0.14 103 | 0.20 70 | 0.33 111 | 1.07 121 | 0.29 88 |
Adaptive flow [45] | 120.1 | 0.25 127 | 0.35 118 | 0.33 126 | 0.64 124 | 0.92 117 | 0.73 123 | 0.36 121 | 0.82 119 | 0.47 122 | 0.39 118 | 1.17 119 | 0.43 118 | 0.48 115 | 0.74 114 | 0.42 112 | 0.38 124 | 0.86 116 | 0.38 125 | 0.33 128 | 0.26 128 | 1.16 127 | 0.38 118 | 0.75 110 | 0.56 113 |
PGAM+LK [55] | 122.5 | 0.17 121 | 0.48 127 | 0.23 125 | 0.43 118 | 1.18 124 | 0.50 119 | 0.32 120 | 0.94 122 | 0.41 118 | 0.37 117 | 2.27 129 | 0.38 117 | 0.64 122 | 1.03 122 | 0.73 120 | 0.32 123 | 1.28 125 | 0.28 120 | 0.23 127 | 0.23 127 | 0.49 125 | 0.61 124 | 1.42 124 | 1.13 125 |
HCIC-L [99] | 124.3 | 0.31 129 | 0.41 123 | 0.21 122 | 1.17 129 | 1.24 125 | 1.42 129 | 0.72 126 | 0.78 117 | 1.35 129 | 0.69 124 | 1.72 125 | 0.69 124 | 0.73 123 | 0.88 120 | 0.69 117 | 0.53 126 | 0.71 114 | 0.52 127 | 0.58 129 | 0.44 129 | 1.19 129 | 0.75 126 | 1.03 118 | 1.00 123 |
Pyramid LK [2] | 124.5 | 0.20 123 | 0.36 120 | 0.40 129 | 0.87 127 | 1.02 119 | 1.40 128 | 0.38 122 | 0.91 121 | 0.71 126 | 0.43 121 | 1.73 126 | 0.61 121 | 0.94 128 | 1.69 128 | 1.39 128 | 0.55 127 | 1.01 121 | 0.50 126 | 0.16 121 | 0.18 124 | 0.32 114 | 1.59 129 | 2.85 129 | 4.45 129 |
AdaConv-v1 [126] | 130.0 | 0.86 130 | 0.85 130 | 0.90 130 | 4.02 130 | 4.67 130 | 3.77 130 | 2.99 130 | 2.69 130 | 3.45 130 | 1.75 130 | 3.37 130 | 1.75 130 | 6.24 130 | 6.82 130 | 6.91 130 | 5.00 130 | 2.89 129 | 4.51 130 | 1.61 130 | 1.10 130 | 3.12 130 | 7.52 130 | 8.25 130 | 8.00 130 |
SepConv-v1 [127] | 130.0 | 0.86 130 | 0.85 130 | 0.90 130 | 4.02 130 | 4.67 130 | 3.77 130 | 2.99 130 | 2.69 130 | 3.45 130 | 1.75 130 | 3.37 130 | 1.75 130 | 6.24 130 | 6.82 130 | 6.91 130 | 5.00 130 | 2.89 129 | 4.51 130 | 1.61 130 | 1.10 130 | 3.12 130 | 7.52 130 | 8.25 130 | 8.00 130 |
Method | time* | frames | color | Reference and notes | |
[1] 2D-CLG | 844 | 2 | gray | The 2D-CLG method by Bruhn et al. as implemented by Stefan Roth. [A. Bruhn, J. Weickert, and C. Schnörr. Lucas/Kanade meets Horn/Schunck: combining local and global optic flow methods. IJCV 63(3), 2005.] Parameters were set to match the published performance on Yosemite sequence, which may not be optimal for other sequences. | |
[2] Pyramid LK | 12 | 2 | color | A modification of Bouguet's pyramidal implementation of Lucas-Kanade. | |
[3] Horn & Schunck | 49 | 2 | gray | A modern Matlab implementation of the Horn & Schunck method by Deqing Sun. Parameters set to optimize AAE on all training data. | |
[4] Black & Anandan | 328 | 2 | gray | A modern Matlab implementation of the Black & Anandan method by Deqing Sun. | |
[5] Brox et al. | 18 | 2 | color | T. Brox, A. Bruhn, N. Papenberg, and J. Weickert. High accuracy optical flow estimation based on a theory for warping. ECCV 2004. (Improved using separate robust functions as proposed in A. Bruhn and J. Weickert, Towards ultimate motion estimation, ICCV 2005; improved by training on the training set.) | |
[6] Fusion | 2,666 | 2 | color | V. Lempitsky, S. Roth, and C. Rother. Discrete-continuous optimization for optical flow estimation. CVPR 2008. | |
[7] Dynamic MRF | 366 | 2 | gray | B. Glocker, N. Paragios, N. Komodakis, G. Tziritas, and N. Navab. Optical flow estimation with uncertainties through dynamic MRFs. CVPR 2008. (Method improved since publication.) | |
[8] Second-order prior | 14 | 2 | gray | W. Trobin, T. Pock, D. Cremers, and H. Bischof. An unbiased second-order prior for high-accuracy motion estimation. DAGM 2008. (Method improved since publication; for details see W. Trobin, Ph.D. thesis, 2009.) | |
[9] GroupFlow | 600 | 2 | gray | X. Ren. Local Grouping for Optical Flow. CVPR 2008. | |
[10] SegOF | 60 | 2 | color | L. Xu, J. Chen, and J. Jia. Segmentation based variational model for accurate optical flow estimation. ECCV 2008. Code available. | |
[11] Learning Flow | 825 | 2 | gray | D. Sun, S. Roth, J.P. Lewis, and M. Black. Learning optical flow (SRF-LFC). ECCV 2008. | |
[12] CBF | 69 | 2 | color | W. Trobin, T. Pock, D. Cremers, and H. Bischof. Continuous energy minimization via repeated binary fusion. ECCV 2008. (Method improved since publication; for details see W. Trobin, Ph.D. thesis, 2009.) | |
[13] SPSA-learn | 200 | 2 | color | Y. Li and D. Huttenlocher. Learning for optical flow using stochastic optimization. ECCV 2008. | |
[14] GraphCuts | 1,200 | 2 | color | T. Cooke. Two applications of graph-cuts to image processing. DICTA 2008. | |
[15] F-TV-L1 | 8 | 2 | gray | A. Wedel, T. Pock, J. Braun, U. Franke, and D. Cremers. Duality TV-L1 flow with fundamental matrix prior. IVCNZ 2008. | |
[16] FOLKI | 1.4 | 2 | gray | G. Le Besnerais and F. Champagnat. Dense optical flow by iterative local window registration. ICIP 2005. | |
[17] TV-L1-improved | 2.9 | 2 | gray | A. Wedel, T. Pock, C. Zach, H. Bischof, and D. Cremers. An improved algorithm for TV-L1 optical flow computation. Proceedings of the Dagstuhl Visual Motion Analysis Workshop 2008. Code at GPU4Vision. | |
[18] DPOF | 287 | 2 | color | C. Lei and Y.-H. Yang. Optical flow estimation on coarse-to-fine region-trees using discrete optimization. ICCV 2009. (Method improved since publication.) | |
[19] Filter Flow | 34,000 | 2 | color | S. Seitz and S. Baker. Filter flow. ICCV 2009. | |
[20] Adaptive | 9.2 | 2 | gray | A. Wedel, D. Cremers, T. Pock, and H. Bischof. Structure- and motion-adaptive regularization for high accuracy optic flow. ICCV 2009. | |
[21] Complementary OF | 44 | 2 | color | H. Zimmer, A. Bruhn, J. Weickert, L. Valgaerts, A. Salgado, B. Rosenhahn, and H.-P. Seidel. Complementary optic flow. EMMCVPR 2009. | |
[22] Aniso. Huber-L1 | 2 | 2 | gray | M. Werlberger, W. Trobin, T. Pock, A. Wedel, D. Cremers, and H. Bischof. Anisotropic Huber-L1 optical flow. BMVC 2009. Code at GPU4Vision. | |
[23] Rannacher | 0.12 | 2 | gray | J. Rannacher. Realtime 3D motion estimation on graphics hardware. Bachelor thesis, Heidelberg University, 2009. | |
[24] TI-DOFE | 260 | 2 | gray | C. Cassisa, S. Simoens, and V. Prinet. Two-frame optical flow formulation in an unwarped multiresolution scheme. CIARP 2009. | |
[25] NL-TV-NCC | 20 | 2 | color | M. Werlberger, T. Pock, and H. Bischof. Motion estimation with non-local total variation regularization. CVPR 2010. | |
[26] MDP-Flow | 188 | 2 | color | L. Xu, J. Jia, and Y. Matsushita. Motion detail preserving optical flow estimation. CVPR 2010. | |
[27] ACK-Prior | 5872 | 2 | color | K. Lee, D. Kwon, I. Yun, and S. Lee. Optical flow estimation with adaptive convolution kernel prior on discrete framework. CVPR 2010. | |
[28] LDOF | 122 | 2 | color | T. Brox and J. Malik. Large displacement optical flow: descriptor matching in variational motion estimation. PAMI 33(3):500-513, 2011. | |
[29] p-harmonic | 565 | 2 | gray | J. Gai and R. Stevenson. Optical flow estimation with p-harmonic regularization. ICIP 2010. | |
[30] TriangleFlow | 4200 | 2 | gray | B. Glocker, H. Heibel, N. Navab, P. Kohli, and C. Rother. TriangleFlow: Optical flow with triangulation-based higher-order likelihoods. ECCV 2010. | |
[31] Classic+NL | 972 | 2 | color | D. Sun, S. Roth, and M. Black. Secrets of optical flow estimation and their principles. CVPR 2010. Matlab code. | |
[32] Classic++ | 486 | 2 | gray | A modern implementation of the classical formulation descended from Horn & Schunck and Black & Anandan; see D. Sun, S. Roth, and M. Black, Secrets of optical flow estimation and their principles, CVPR 2010. | |
[33] Nguyen | 33 | 2 | gray | D. Nguyen. Tuning optical flow estimation with image-driven functions. ICRA 2011. | |
[34] Modified CLG | 133 | 2 | gray | R. Fezzani, F. Champagnat, and G. Le Besnerais. Combined local global method for optic flow computation. EUSIPCO 2010. | |
[35] ComplOF-FED-GPU | 0.97 | 2 | color | P. Gwosdek, H. Zimmer, S. Grewenig, A. Bruhn, and J. Weickert. A highly efficient GPU implementation for variational optic flow based on the Euler-Lagrange framework. CVGPU Workshop 2010. | |
[36] Ad-TV-NDC | 35 | 2 | gray | M. Nawaz. Motion estimation with adaptive regularization and neighborhood dependent constraint. DICTA 2010. | |
[37] Layers++ | 18206 | 2 | color | D. Sun, E. Sudderth, and M. Black. Layered image motion with explicit occlusions, temporal consistency, and depth ordering. NIPS 2010. | |
[38] OFH | 620 | 3 | color | H. Zimmer, A. Bruhn, J. Weickert. Optic flow in harmony. IJCV 93(3) 2011. | |
[39] LSM | 1615 | 2 | color | K. Jia, X. Wang, and X. Tang. Optical flow estimation using learned sparse model. ICCV 2011. | |
[40] CostFilter | 55 | 2 | color | C. Rhemann, A. Hosni, M. Bleyer, C. Rother, and M. Gelautz. Fast cost-volume filtering for visual correspondence and beyond. CVPR 2011. | |
[41] Bartels | 0.15 | 2 | gray | C. Bartels and G. de Haan. Smoothness constraints in recursive search motion estimation for picture rate conversion. IEEE TCSVT 2010. Version improved since publication: mapped on GPU. | |
[42] Shiralkar | 600 | 2 | gray | M. Shiralkar and R. Schalkoff. A self organization-based optical flow estimator with GPU implementation. MVA 23(6):1229-1242. | |
[43] HBpMotionGpu | 1000 | 5 | gray | S. Grauer-Gray and C. Kambhamettu. Hierarchical belief propagation to reduce search space using CUDA for stereo and motion estimation. WACV 2009. (Method improved since publication.) | |
[44] StereoFlow | 7200 | 2 | color | G. Rosman, S. Shem-Tov, D. Bitton, T. Nir, G. Adiv, R. Kimmel, A. Feuer, and A. Bruckstein. Over-parameterized optical flow using a stereoscopic constraint. SSVM 2011:761-772. | |
[45] Adaptive flow | 121 | 2 | gray | T. Arici. Energy minimization based motion estimation using adaptive smoothness priors. Submitted to IEEE TIP 2011. | |
[46] TC-Flow | 2500 | 5 | color | S. Volz, A. Bruhn, L. Valgaerts, and H. Zimmer. Modeling temporal coherence for optical flow. ICCV 2011. | |
[47] SLK | 300 | 2 | gray | T. Corpetti and E. Mémin. Stochastic uncertainty models for the luminance consistency assumption. IEEE TIP 2011. | |
[48] CLG-TV | 29 | 2 | gray | M. Drulea. Total variation regularization of local-global optical flow. ITSC 2011. Matlab code. | |
[49] SimpleFlow | 1.7 | 2 | color | M. Tao, J. Bai, P. Kohli, S. Paris. SimpleFlow: a non-iterative, sublinear optical flow algorithm. EUROGRAPHICS 2012. | |
[50] IAOF | 57 | 2 | gray | D. Nguyen. Improving motion estimation using image-driven functions and hybrid scheme. PSIVT 2011. | |
[51] IAOF2 | 56 | 2 | gray | D. Nguyen. Enhancing the sharpness of flow field using image-driven functions with occlusion-aware filter. Submitted to TIP 2011. | |
[52] LocallyOriented | 9541 | 2 | gray | Y.Niu, A. Dick, and M. Brooks. Locally oriented optical flow computation. To appear in TIP 2012. | |
[53] IROF-TV | 261 | 2 | color | H. Rashwan, D. Puig, and M. Garcia. On improving the robustness of differential optical flow. ICCV 2011 Artemis workshop. | |
[54] Sparse Occlusion | 2312 | 2 | color | A. Ayvaci, M. Raptis, and S. Soatto. Sparse occlusion detection with optical flow. Submitted to IJCV 2011. | |
[55] PGAM+LK | 0.37 | 2 | gray | A. Alba, E. Arce-Santana, and M. Rivera. Optical flow estimation with prior models obtained from phase correlation. ISVC 2010. | |
[56] Sparse-NonSparse | 713 | 2 | color | L. Chen, J. Wang, and Y. Wu. Decomposing and regularizing sparse/non-sparse components for motion field estimation. Submitted to PAMI 2013. | |
[57] nLayers | 36150 | 4 | color | D. Sun, E. Sudderth, and M. Black. Layered segmentation and optical flow estimation over time. CVPR 2012. | |
[58] IROF++ | 187 | 2 | color | H. Rashwan, D. Puig, and M. Garcia. Variational optical flow estimation based on stick tensor voting. IEEE TIP 2013. | |
[59] COFM | 600 | 3 | color | M. Mozerov. Constrained optical flow estimation as a matching problem. IEEE TIP 2013. | |
[60] Efficient-NL | 400 | 2 | color | P. Krähenbühl and V. Koltun. Efficient nonlocal regularization for optical flow. ECCV 2012. | |
[61] BlockOverlap | 2 | 2 | gray | M. Santoro, G. AlRegib, and Y. Altunbasak. Motion estimation using block overlap minimization. Submitted to MMSP 2012. | |
[62] Ramp | 1200 | 2 | color | A. Singh and N. Ahuja. Exploiting ramp structures for improving optical flow estimation. ICPR 2012. | |
[63] Occlusion-TV-L1 | 538 | 3 | gray | C. Ballester, L. Garrido, V. Lazcano, and V. Caselles. A TV-L1 optical flow method with occlusion detection. DAGM-OAGM 2012. | |
[64] TV-L1-MCT | 90 | 2 | color | M. Mohamed and B. Mertsching. TV-L1 optical flow estimation with image details recovering based on modified census transform. ISVC 2012. | |
[65] Local-TV-L1 | 500 | 2 | gray | L. Raket. Local smoothness for global optical flow. ICIP 2012. | |
[66] ALD-Flow | 61 | 2 | color | M. Stoll, A. Bruhn, and S. Volz. Adaptive integration of feature matches into variational optic flow methods. ACCV 2012. | |
[67] SIOF | 234 | 2 | color | L. Xu, Z. Dai, and J. Jia. Scale invariant optical flow. ECCV 2012. | |
[68] MDP-Flow2 | 342 | 2 | color | L. Xu, J. Jia, and Y. Matsushita. Motion detail preserving optical flow estimation. PAMI 34(9):1744-1757, 2012. Code available. | |
[69] TCOF | 1421 | all | gray | J. Sanchez, A. Salgado, and N. Monzon. Optical flow estimation with consistent spatio-temporal coherence models. VISAPP 2013. | |
[70] LME | 476 | 2 | color | W. Li, D. Cosker, M. Brown, and R. Tang. Optical flow estimation using Laplacian mesh energy. CVPR 2013. | |
[71] NN-field | 362 | 2 | color | L. Chen, H. Jin, Z. Lin, S. Cohen, and Y. Wu. Large displacement optical flow from nearest neighbor fields. CVPR 2013. | |
[72] FESL | 3310 | 2 | color | W. Dong, G. Shi, X. Hu, and Y. Ma. Nonlocal sparse and low-rank regularization for optical flow estimation. Submitted to IEEE TIP 2013. | |
[73] PMF | 35 | 2 | color | J. Lu, H. Yang, D. Min, and M. Do. PatchMatch filter: efficient edge-aware filtering meets randomized search for fast correspondence field estimation. CVPR 2013. | |
[74] FC-2Layers-FF | 2662 | 4 | color | D. Sun, J. Wulff, E. Sudderth, H. Pfister, and M. Black. A fully-connected layered model of foreground and background flow. CVPR 2013. | |
[75] Correlation Flow | 290 | 2 | color | M. Drulea and S. Nedevschi. Motion estimation using the correlation transform. TIP 2013. Matlab code. | |
[76] TC/T-Flow | 341 | 5 | color | M. Stoll, S. Volz, and A. Bruhn. Joint trilateral filtering for multiframe optical flow. ICIP 2013. | |
[77] OFLAF | 1530 | 2 | color | T. Kim, H. Lee, and K. Lee. Optical flow via locally adaptive fusion of complementary data costs. ICCV 2013. | |
[78] Periodicity | 8000 | 4 | color | G. Khachaturov, S. Gonzalez-Brambila, and J. Gonzalez-Trejo. Periodicity-based computation of optical flow. Submitted to Computacion y Sistemas (CyS) 2013. | |
[79] SILK | 572 | 2 | gray | P. Zille, C. Xu, T. Corpetti, L. Shao. Observation models based on scale interactions for optical flow estimation. Submitted to IEEE TIP. | |
[80] CRTflow | 13 | 3 | color | O. Demetz, D. Hafner, and J. Weickert. The complete rank transform: a tool for accurate and morphologically invariant matching of structures. BMVC 2013. | |
[81] SuperFlow | 178 | 2 | color | Anonymous. Superpixel based optical flow estimation. ICCV 2013 submission 507. | |
[82] Aniso-Texture | 300 | 2 | color | Anonymous. Texture information-based optical flow estimation using an incremental multi-resolution approach. ITC-CSCC 2013 submission 267. | |
[83] Classic+CPF | 640 | 2 | gray | Z. Tu, R. Veltkamp, and N. van der Aa. A combined post-filtering method to improve accuracy of variational optical flow estimation. Submitted to Pattern Recognition 2013. | |
[84] S2D-Matching | 1200 | 2 | color | Anonymous. Locally affine sparse-to-dense matching for motion and occlusion estimation. ICCV 2013 submission 1479. | |
[85] AGIF+OF | 438 | 2 | gray | Z. Tu, R. Poppe, and R. Veltkamp. Adaptive guided image filter to warped interpolation image for variational optical flow computation. Submitted to Signal Processing 2015. | |
[86] DeepFlow | 13 | 2 | color | P. Weinzaepfel, J. Revaud, Z. Harchaoui, and C. Schmid. DeepFlow: large displacement optical flow with deep matching. ICCV 2013. | |
[87] NNF-Local | 673 | 2 | color | Z. Chen, H. Jin, Z. Lin, S. Cohen, and Y. Wu. Large displacement optical flow with nearest neighbor field. Submitted to PAMI 2014. | |
[88] EPPM w/o HM | 2.5 | 2 | color | L. Bao, Q. Yang, and H. Jin. Fast edge-preserving PatchMatch for large displacement optical flow. CVPR 2014. | |
[89] MLDP_OF | 165 | 2 | gray | M. Mohamed, H. Rashwan, B. Mertsching, M. Garcia, and D. Puig. Illumination-robust optical flow approach using local directional pattern. IEEE TCSVT 24(9):1499-1508, 2014. | |
[90] RFlow | 20 | 2 | gray | S. Ali, C. Daul, and W. Blondel. Robust and accurate optical flow estimation for weak texture and varying illumination condition: Application to cystoscopy. IPTA 2014. | |
[91] SRR-TVOF-NL | 32 | all | color | P. Pohl, M. Sirotenko, E. Tolstaya, and V. Bucha. Edge preserving motion estimation with occlusions correction for assisted 2D to 3D conversion. IS&T/SPIE Electronic Imaging 2014. | |
[92] 2DHMM-SAS | 157 | 2 | color | M.-C. Shih, R. Shenoy, and K. Rose. A two-dimensional hidden Markov model with spatially-adaptive states with application of optical flow. ICIP 2014 submission. | |
[93] WLIF-Flow | 700 | 2 | color | Z. Tu, R. Veltkamp, N. van der Aa, and C. Van Gemeren. Weighted local intensity fusion method for variational optical flow estimation. Submitted to TIP 2014. | |
[94] FMOF | 215 | 2 | color | N. Jith, A. Ramakanth, and V. Babu. Optical flow estimation using approximate nearest neighbor field fusion. ICASSP 2014. | |
[95] TriFlow | 150 | 2 | color | TriFlow. Optical flow with geometric occlusion estimation and fusion of multiple frames. ECCV 2014 submission 914. | |
[96] ComponentFusion | 6.5 | 2 | color | Anonymous. Fast optical flow by component fusion. ECCV 2014 submission 941. | |
[97] AggregFlow | 1642 | 2 | color | D. Fortun, P. Bouthemy, and C. Kervrann. Aggregation of local parametric candidates and exemplar-based occlusion handling for optical flow. Preprint arXiv:1407.5759. | |
[98] 2bit-BM-tele | 124 | 2 | gray | R. Xu and D. Taubman. Robust dense block-based motion estimation using a two-bit transform on a Laplacian pyramid. ICIP 2013. | |
[99] HCIC-L | 330 | 2 | color | Anonymous. Globally-optimal image correspondence using a hierarchical graphical model. NIPS 2014 submission 114. | |
[100] TF+OM | 600 | 2 | color | R. Kennedy and C. Taylor. Optical flow with geometric occlusion estimation and fusion of multiple frames. EMMCVPR 2015. | |
[101] PH-Flow | 800 | 2 | color | J. Yang and H. Li. Dense, accurate optical flow estimation with piecewise parametric model. CVPR 2015. | |
[102] EpicFlow | 16 | 2 | color | J. Revaud, P. Weinzaepfel, Z. Harchaoui, and C. Schmid. EpicFlow: edge-preserving interpolation of correspondences for optical flow. CVPR 2015. | |
[103] NNF-EAC | 380 | 2 | color | Anonymous. Variational method for joint optical flow estimation and edge-aware image restoration. CVPR 2015 submission 2336. | |
[104] Heeger++ | 6600 | 5 | gray | Anonymous. A context aware biologically inspired algorithm for optical flow (updated results). CVPR 2015 submission 2238. | |
[105] HBM-GC | 330 | 2 | color | A. Zheng and Y. Yuan. Motion estimation via hierarchical block matching and graph cut. Submitted to ICIP 2015. | |
[106] FFV1MT | 358 | 5 | gray | F. Solari, M. Chessa, N. Medathati, and P. Kornprobst. What can we expect from a V1-MT feedforward architecture for optical flow estimation? Submitted to Signal Processing: Image Communication 2015. | |
[107] ROF-ND | 4 | 2 | color | S. Ali, C. Daul, E. Galbrun, and W. Blondel. Illumination invariant large displacement optical flow using robust neighbourhood descriptors. Submitted to CVIU 2015. | |
[108] DeepFlow2 | 16 | 2 | color | J. Revaud, P. Weinzaepfel, Z. Harchaoui, and C. Schmid. Deep convolutional matching. Submitted to IJCV, 2015. | |
[109] HAST | 2667 | 2 | color | Anonymous. Highly accurate optical flow estimation on superpixel tree. ICCV 2015 submission 2221. | |
[110] FlowFields | 15 | 2 | color | C. Bailer, B. Taetz, and D. Stricker. Flow Fields: Dense unregularized correspondence fields for highly accurate large displacement optical flow estimation. ICCV 2015. | |
[111] SVFilterOh | 1.56 | 2 | color | Anonymous. Fast estimation of large displacement optical flow using PatchMatch and dominant motion patterns. CVPR 2016 submission 1788. | |
[112] FlowNetS+ft+v | 0.5 | 2 | color | Anonymous. Learning optical flow with convolutional neural networks. ICCV 2015 submission 235. | |
[113] CombBMOF | 51 | 2 | color | M. Brüggemann, R. Kays, P. Springer, and O. Erdler. Combined block-matching and adaptive differential motion estimation in a hierarchical multi-scale framework. ICGIP 2014. (Method improved since publication.) | |
[114] PMMST | 182 | 2 | color | F. Zhang, S. Xu, and X. Zhang. High accuracy correspondence field estimation via MST based patch matching. Submitted to TIP 2015. | |
[115] DF-Auto | 70 | 2 | color | N. Monzon, A. Salgado, and J. Sanchez. Regularization strategies for discontinuity-preserving optical flow methods. Submitted to TIP 2015. | |
[116] CPM-Flow | 3 | 2 | color | Anonymous. Efficient coarse-to-fine PatchMatch for large displacement optical flow. CVPR 2016 submission 241. | |
[117] CNN-flow-warp+ref | 1.4 | 3 | color | D. Teney and M. Hebert. Learning to extract motion from videos in convolutional neural networks. ArXiv 1601.07532, 2016. | |
[118] Steered-L1 | 804 | 2 | color | Anonymous. Optical flow estimation via steered-L1 norm. Submitted to WSCG 2016. | |
[119] StereoOF-V1MT | 343 | 2 | gray | Anonymous. Visual features for action-oriented tasks: a cortical-like model for disparity and optic flow computation. BMVC 2016 submission 132. | |
[120] PGM-C | 5 | 2 | color | Y. Li. Pyramidal gradient matching for optical flow estimation. Submitted to PAMI 2016. | |
[121] RNLOD-Flow | 1040 | 2 | gray | C. Zhang, Z. Chen, M. Wang, M. Li, and S. Jiang. Robust non-local TV-L1 optical flow estimation with occlusion detection. Submitted to TIP 2016. | |
[122] FlowNet2 | 0.091 | 2 | color | Anonymous. FlowNet 2.0: Evolution of optical flow estimation with deep networks. CVPR 2017 submission 900. | |
[123] S2F-IF | 20 | 2 | color | Anonymous. S2F-IF: Slow-to-fast interpolator flow. CVPR 2017 submission 765. | |
[124] BriefMatch | 0.068 | 2 | gray | G. Eilertsen, P.-E. Forssen, and J. Unger. Dense binary feature matching for real-time optical flow estimation. SCIA 2017 submission 62. | |
[125] OAR-Flow | 60 | 2 | color | Anonymous. Order-adaptive regularisation for variational optical flow: global, local and in between. SSVM 2017 submission 20. | |
[126] AdaConv-v1 | 2.8 | 2 | color | S. Niklaus, L. Mai, and F. Liu. (Interpolation results only.) Video frame interpolation via adaptive convolution. CVPR 2017. | |
[127] SepConv-v1 | 0.2 | 2 | color | S. Niklaus, L. Mai, and F. Liu. (Interpolation results only.) Video frame interpolation via adaptive separable convolution. ICCV 2017. | |
[128] ProbFlowFields | 37 | 2 | color | A. Wannenwetsch, M. Keuper, and S. Roth. ProbFlow: joint optical flow and uncertainty estimation. ICCV 2017. | |
[129] UnFlow | 0.12 | 2 | color | Anonymous. UnFlow: Unsupervised learning of optical flow with a bidirectional census loss. Submitted to AAAI 2018. | |
[130] FlowFields+ | 10.5 | 2 | color | C. Bailer, B. Taetz, and D. Stricker. Flow fields: Dense correspondence fields for highly accurate large displacement optical flow estimation. Submitted to PAMI 2017. | |
[131] Kuang | 9.9 | 2 | gray | F. Kuang. PatchMatch algorithms for motion estimation and stereo reconstruction. Master thesis, University of Stuttgart, 2017. |