Optical flow evaluation results Statistics:     Average   SD   R2.5   R5.0   R10.0   A50   A75   A95  
Error type:   endpoint   angle   interpolation   normalized interpolation  
Show images: below table   above table        
R2.5
angle
error
avg. Army
(Hidden texture)
GT   im0   im1
Mequon
(Hidden texture)
GT   im0   im1
Schefflera
(Hidden texture)
GT   im0   im1
Wooden
(Hidden texture)
GT   im0   im1
Grove
(Synthetic)
GT   im0   im1
Urban
(Synthetic)
GT   im0   im1
Yosemite
(Synthetic)
GT   im0   im1
Teddy
(Stereo)
GT   im0   im1
rank all disc untext all disc untext all disc untext all disc untext all disc untext all disc untext all disc untext all disc untext
NNF-Local [87]14.8 20.0 3 43.3 3 13.0 3 15.9 11 45.3 26 16.1 15 12.5 4 34.1 9 12.8 10 9.68 21 32.0 11 7.01 25 23.3 4 30.0 4 16.2 4 22.0 34 45.4 9 23.8 65 25.8 25 44.8 7 17.6 22 4.37 11 14.4 29 0.48 1
NN-field [71]15.8 22.1 7 44.0 5 14.6 4 18.4 24 47.3 39 19.2 29 12.5 4 32.9 6 14.0 15 6.57 2 28.2 4 3.57 2 23.4 5 30.1 5 15.9 3 17.9 16 36.8 1 15.8 4 35.9 60 51.5 35 27.4 59 4.58 14 15.3 33 0.55 2
TC/T-Flow [76]21.1 19.9 2 46.9 11 10.2 1 16.0 12 47.8 41 13.9 5 13.0 8 36.6 21 11.1 4 8.90 14 35.0 37 6.10 14 27.2 18 35.7 23 21.0 15 15.7 4 47.2 17 15.8 4 21.0 10 39.2 2 42.1 90 7.86 46 19.5 48 10.9 59
ComponentFusion [96]21.3 20.0 3 46.3 8 14.8 5 16.6 15 40.3 6 18.5 26 11.7 3 33.6 7 10.9 3 7.09 6 35.2 39 4.45 6 27.6 21 36.0 26 21.4 18 21.5 30 54.1 53 20.2 31 31.8 49 56.8 64 16.2 15 5.46 24 12.8 22 7.47 31
ALD-Flow [66]21.6 21.6 6 46.0 7 15.9 7 15.5 8 41.3 9 15.6 11 13.1 9 35.2 13 12.2 6 8.17 11 33.4 23 5.35 10 28.1 25 37.3 36 20.3 11 16.4 7 47.4 19 16.0 6 26.5 28 45.4 9 41.8 88 8.39 50 22.3 59 11.3 61
ProFlow_ROB [147]21.8 24.2 11 53.3 36 15.8 6 15.6 9 44.5 21 14.8 8 15.5 18 43.0 44 11.8 5 6.94 4 33.1 19 3.97 4 30.0 38 39.9 45 19.6 8 16.2 6 49.1 28 16.1 7 14.9 4 54.0 50 13.9 13 7.47 41 21.2 55 8.63 44
nLayers [57]22.6 22.7 9 40.3 2 18.4 10 27.2 93 45.9 30 30.4 94 15.7 24 35.4 15 21.4 63 8.12 10 26.6 2 6.21 15 22.5 2 29.0 3 15.5 2 19.9 21 40.8 4 17.8 10 31.3 45 52.6 44 16.9 20 4.26 8 11.2 9 5.84 7
OFLAF [77]23.3 29.6 36 47.4 13 24.8 20 17.8 20 40.9 7 18.3 24 11.6 2 26.9 2 13.2 12 11.5 41 29.3 8 8.97 62 23.7 6 30.9 6 16.5 5 22.2 38 41.5 5 19.3 21 30.2 40 50.6 30 32.9 67 5.94 29 13.8 24 8.44 41
RNLOD-Flow [121]23.8 20.7 5 43.8 4 19.4 11 18.1 21 45.9 30 17.0 20 12.7 6 34.1 9 12.3 7 7.38 7 28.7 6 4.87 7 26.5 14 34.9 19 20.9 13 20.3 22 46.2 12 20.2 31 43.1 86 60.3 76 47.5 107 4.98 21 12.7 21 6.55 16
WLIF-Flow [93]23.9 27.0 21 47.0 12 23.1 18 21.7 45 46.6 37 23.3 44 14.8 13 36.2 17 16.3 25 9.33 18 32.2 14 6.64 19 27.7 22 35.1 21 23.4 34 21.6 31 47.7 21 19.2 18 28.7 36 45.3 8 32.0 66 4.50 13 11.2 9 6.43 12
HAST [109]24.2 19.5 1 40.1 1 11.6 2 16.1 13 39.7 4 14.8 8 8.49 1 21.2 1 7.09 1 6.79 3 29.0 7 3.66 3 21.6 1 28.3 2 13.8 1 24.3 60 48.5 25 24.3 67 41.7 82 59.8 74 63.0 133 6.05 31 11.5 15 8.72 45
MDP-Flow2 [68]25.6 35.3 49 55.4 43 30.2 47 14.2 3 39.7 4 14.2 7 13.6 10 31.4 3 12.9 11 12.2 46 34.1 31 8.19 48 27.7 22 35.0 20 22.1 21 22.4 42 45.4 9 21.5 45 27.1 29 54.1 51 16.5 16 5.87 28 14.0 26 4.25 4
OAR-Flow [125]26.7 25.6 14 54.9 40 22.4 15 18.7 25 44.8 23 19.1 27 17.2 37 43.7 47 18.0 31 8.53 13 31.6 10 5.65 11 29.7 36 39.3 41 20.9 13 14.5 2 47.3 18 13.4 2 14.1 3 38.0 1 20.7 36 9.84 64 21.1 54 16.1 78
Layers++ [37]27.2 28.0 33 48.8 16 30.9 50 23.3 61 45.6 29 25.5 69 13.7 11 31.4 3 18.1 32 8.08 9 24.9 1 5.87 12 22.9 3 28.1 1 19.6 8 22.2 38 46.3 14 20.7 35 39.6 76 55.9 60 35.0 69 4.16 4 9.78 2 6.81 18
TC-Flow [46]27.6 24.4 12 52.2 32 22.6 16 11.8 2 38.8 3 11.2 1 12.7 6 35.7 16 9.84 2 9.88 23 34.9 35 7.49 33 28.9 32 38.6 40 21.2 17 20.6 24 52.1 38 21.6 46 22.6 20 47.1 13 36.3 72 9.05 56 21.5 57 12.4 66
LME [70]29.5 31.5 38 51.4 27 21.0 14 14.7 5 36.9 2 15.7 12 16.0 29 35.2 13 19.8 52 11.8 43 36.6 43 8.08 43 28.6 29 36.0 26 25.5 46 21.1 28 49.0 27 19.8 28 29.8 39 50.0 27 21.5 38 6.61 33 15.1 30 7.95 37
AGIF+OF [85]29.9 26.3 15 48.8 16 24.1 19 24.9 72 52.0 71 26.2 71 16.0 29 39.4 31 19.7 49 8.96 15 32.0 11 6.64 19 26.7 16 33.7 13 21.5 20 21.4 29 49.4 29 18.5 14 28.2 33 49.6 24 30.5 64 4.63 17 11.3 14 7.30 27
PH-Flow [101]31.0 26.7 17 51.6 28 25.6 27 21.7 45 49.4 52 23.8 49 15.5 18 37.6 24 19.5 45 10.2 28 33.8 27 7.44 32 26.4 12 33.7 13 21.0 15 22.0 34 50.3 32 20.4 33 38.8 70 48.4 18 45.4 97 4.26 8 11.2 9 6.38 11
FC-2Layers-FF [74]31.1 26.9 20 48.6 15 28.3 40 22.1 49 48.8 47 23.8 49 14.1 12 32.4 5 19.6 47 9.10 16 28.3 5 6.47 18 25.5 7 31.7 7 23.0 30 23.3 49 47.7 21 21.8 47 44.1 90 56.3 63 46.0 102 3.54 1 9.08 1 5.44 6
NNF-EAC [103]32.2 34.9 44 54.8 38 29.9 46 15.6 9 41.7 11 15.9 13 15.1 17 34.8 12 15.7 19 12.4 50 35.1 38 8.33 51 28.1 25 35.7 23 22.9 28 24.5 63 46.2 12 22.7 58 31.6 48 49.1 21 20.1 34 7.54 43 17.1 41 7.44 30
Classic+CPF [83]32.3 27.1 25 51.2 26 25.5 26 23.9 65 51.4 67 25.1 64 15.9 26 39.3 30 19.4 43 9.17 17 32.8 17 6.69 21 27.3 19 34.5 17 23.9 35 21.0 27 48.7 26 18.0 12 35.6 59 49.9 26 45.7 99 4.25 7 10.7 5 6.61 17
IROF++ [58]32.6 27.4 27 50.7 22 26.7 33 22.1 49 50.1 62 24.1 54 16.3 32 40.0 35 19.6 47 10.6 35 34.3 32 7.72 38 27.9 24 35.7 23 22.5 23 22.3 40 54.1 53 19.9 29 25.5 24 49.6 24 11.5 10 5.49 25 14.1 27 6.54 15
COFM [59]32.9 22.1 7 49.1 18 16.6 8 18.2 22 43.5 16 19.2 29 15.9 26 38.3 27 21.5 64 7.02 5 32.6 15 4.40 5 31.4 47 37.9 37 35.0 86 22.1 36 46.4 15 18.3 13 28.7 36 45.9 11 45.5 98 9.25 57 15.5 35 15.7 77
Sparse-NonSparse [56]33.2 26.8 19 51.9 31 26.8 34 22.2 54 49.0 50 24.6 60 15.6 21 39.4 31 19.0 38 9.46 20 33.6 25 7.07 27 29.0 33 37.2 35 24.3 40 21.7 33 50.4 33 19.5 26 34.0 52 45.8 10 41.6 86 4.44 12 10.9 6 6.96 22
FESL [72]34.2 27.0 21 46.3 8 31.2 52 26.0 84 51.8 70 26.9 76 15.8 25 37.0 22 19.5 45 7.89 8 30.7 9 5.17 9 26.6 15 33.8 15 22.6 24 20.3 22 45.8 11 19.3 21 39.9 77 61.2 81 35.6 70 5.04 22 12.5 20 6.48 14
Efficient-NL [60]34.8 23.3 10 44.7 6 17.6 9 24.6 71 51.6 69 25.4 68 15.0 16 36.2 17 17.5 29 9.92 24 33.1 19 6.94 23 26.4 12 34.0 16 20.3 11 27.2 73 49.4 29 22.6 53 37.6 65 50.5 29 37.1 73 6.98 38 16.2 37 8.16 38
JOF [141]35.1 25.3 13 46.6 10 23.0 17 24.1 67 49.9 60 26.7 73 14.8 13 34.3 11 21.0 61 8.34 12 32.1 13 5.95 13 26.1 11 32.9 12 22.8 27 22.8 45 49.4 29 22.2 49 41.8 83 60.0 75 56.2 124 4.17 5 10.9 6 6.44 13
LSM [39]36.4 26.5 16 50.8 23 27.0 36 22.1 49 49.4 52 24.4 58 15.6 21 38.2 26 19.4 43 10.1 25 33.0 18 7.43 31 28.6 29 36.3 28 24.7 41 22.8 45 50.5 34 20.9 39 40.3 79 49.3 22 45.8 101 4.76 19 12.0 17 6.93 21
PMMST [114]36.5 42.4 76 58.9 48 40.2 73 22.1 49 45.3 26 24.7 63 17.9 43 38.4 28 18.8 35 12.2 46 28.1 3 8.26 49 25.7 8 32.7 10 18.5 7 22.3 40 45.3 8 20.8 37 28.5 35 50.7 31 18.3 25 8.51 53 16.8 38 8.73 46
Ramp [62]37.9 27.0 21 52.3 33 26.1 31 22.3 56 49.1 51 24.6 60 15.6 21 37.8 25 19.8 52 10.5 34 33.6 25 7.61 35 28.6 29 36.6 30 24.0 36 23.6 52 51.2 35 21.8 47 38.2 66 44.5 5 46.1 103 4.86 20 12.1 18 7.13 25
Classic+NL [31]38.0 27.2 26 47.8 14 28.4 42 22.0 48 49.4 52 24.1 54 15.5 18 37.4 23 19.8 52 10.4 32 33.8 27 7.40 30 28.5 27 36.4 29 24.2 39 23.3 49 52.2 39 21.1 40 43.9 88 52.5 41 44.6 94 4.60 16 11.2 9 7.04 24
2DHMM-SAS [92]40.7 26.7 17 51.8 30 25.6 27 22.2 54 51.3 65 23.8 49 17.8 41 42.3 41 20.0 57 10.4 32 34.6 34 7.52 34 28.5 27 36.6 30 24.0 36 22.7 44 54.5 56 20.6 34 39.0 72 47.8 16 45.1 95 5.51 26 13.9 25 7.73 35
FMOF [94]41.3 27.8 32 50.4 21 27.1 38 26.2 86 52.5 75 27.4 79 15.9 26 36.4 20 21.7 65 9.71 22 32.7 16 6.98 24 27.3 19 34.6 18 24.1 38 23.7 54 47.6 20 19.7 27 38.4 67 55.4 57 48.9 110 5.80 27 14.1 27 7.00 23
SVFilterOh [111]42.9 39.6 66 54.6 37 39.8 72 23.8 64 44.7 22 24.0 52 17.3 38 33.8 8 18.8 35 10.2 28 36.4 42 4.93 8 25.9 10 32.1 8 19.6 8 24.7 64 46.7 16 22.9 61 52.0 114 76.3 132 59.2 127 4.23 6 10.9 6 5.01 5
S2D-Matching [84]43.6 27.7 29 50.2 20 28.4 42 22.1 49 48.8 47 24.3 56 16.5 34 40.0 35 19.3 41 10.6 35 33.8 27 7.78 39 29.1 35 36.7 33 25.0 43 24.4 62 53.0 49 22.6 53 47.0 101 53.6 47 50.8 115 4.58 14 11.2 9 7.54 32
ProbFlowFields [128]44.0 33.9 40 68.5 73 30.9 50 20.3 35 45.3 26 22.1 39 19.2 52 44.8 50 22.9 70 11.7 42 39.3 52 8.91 61 31.0 45 40.1 46 23.3 32 16.8 8 47.8 23 19.2 18 23.7 23 55.9 60 23.9 47 8.39 50 22.4 60 9.82 55
SimpleFlow [49]44.5 28.6 35 51.6 28 29.5 45 25.0 73 51.3 65 28.2 85 18.6 48 43.0 44 23.3 71 10.1 25 33.5 24 7.04 26 29.9 37 37.9 37 26.2 48 27.8 78 52.2 39 24.0 66 35.1 57 47.9 17 29.7 63 4.66 18 12.4 19 6.92 20
PMF [73]45.9 37.7 59 58.3 46 27.3 39 19.4 28 45.0 25 18.3 24 16.2 31 39.9 34 14.0 15 13.6 64 35.7 40 8.03 42 25.8 9 32.5 9 17.1 6 30.2 86 57.0 68 31.6 89 58.9 125 74.7 124 55.9 123 3.95 3 10.3 3 6.15 10
Adaptive [20]46.8 27.0 21 52.6 34 19.8 12 21.9 47 47.8 41 22.6 41 20.5 57 47.8 59 19.7 49 10.3 30 39.9 56 6.31 17 45.6 126 52.6 123 51.3 128 17.4 9 48.2 24 13.9 3 34.8 56 56.9 66 19.8 32 6.04 30 15.2 31 7.54 32
TV-L1-MCT [64]46.9 27.5 28 49.4 19 27.0 36 26.5 87 52.8 80 27.8 84 16.8 35 39.1 29 21.8 67 10.6 35 33.8 27 7.82 41 30.3 40 38.1 39 28.7 64 24.7 64 53.4 51 23.4 63 27.4 30 52.5 41 19.4 29 7.28 40 15.3 33 11.4 63
Correlation Flow [75]47.1 36.6 56 55.3 42 34.4 61 16.6 15 44.4 20 14.8 8 18.3 47 42.9 43 12.7 9 12.4 50 39.7 54 8.46 55 32.2 50 40.6 50 25.2 44 29.0 81 54.2 55 29.7 85 39.1 73 52.1 39 47.4 106 6.53 32 16.1 36 6.88 19
IROF-TV [53]47.5 30.9 37 54.8 38 31.6 54 22.8 60 50.5 63 25.1 64 16.9 36 40.8 37 20.8 59 14.0 67 43.7 71 10.1 69 31.2 46 39.5 42 28.6 62 26.8 72 58.7 78 25.6 70 18.8 7 48.4 18 8.08 8 5.28 23 13.6 23 7.77 36
AggregFlow [97]47.8 36.3 55 52.7 35 35.7 63 26.6 88 52.6 78 26.7 73 23.3 70 48.2 61 28.9 89 12.1 44 34.9 35 8.63 57 30.2 39 40.3 49 21.4 18 15.9 5 38.3 2 16.8 8 26.1 27 47.3 14 16.8 18 12.6 79 20.3 50 20.0 91
Occlusion-TV-L1 [63]48.5 34.3 43 58.5 47 25.1 22 20.0 31 46.5 36 20.8 35 22.3 67 49.9 62 20.6 58 12.6 56 41.7 62 8.41 54 35.2 72 44.9 80 32.3 77 17.7 12 52.6 44 21.1 40 28.2 33 52.5 41 13.0 11 9.66 60 23.7 64 10.6 57
3DFlow [135]50.8 35.7 51 56.6 44 28.3 40 19.1 27 46.2 34 17.4 22 18.8 49 41.7 40 14.5 17 13.9 65 33.3 22 9.85 67 29.0 33 36.6 30 23.0 30 32.9 92 65.1 101 34.7 101 49.9 108 65.5 96 77.7 143 3.87 2 10.3 3 3.02 3
Classic++ [32]52.5 27.7 29 51.0 25 28.7 44 21.5 44 45.9 30 24.3 56 18.1 46 44.3 48 19.9 55 10.3 30 37.7 45 7.14 28 33.4 54 44.1 70 27.9 57 24.0 57 57.9 73 21.4 44 46.3 97 55.6 58 49.7 112 8.45 52 20.7 51 9.69 54
Aniso-Texture [82]52.8 28.5 34 50.9 24 32.7 58 21.2 39 41.7 11 25.3 67 17.6 40 41.2 38 21.1 62 5.72 1 33.2 21 2.54 1 35.1 71 43.7 67 31.3 75 23.6 52 53.6 52 22.6 53 62.2 132 75.8 130 53.5 120 6.89 37 17.4 43 8.20 39
IIOF-NLDP [131]53.4 33.3 39 59.3 49 24.9 21 24.4 70 53.8 87 22.9 43 18.8 49 46.2 54 15.0 18 13.5 61 39.7 54 9.86 68 32.3 51 41.2 54 23.3 32 29.9 85 60.2 87 29.6 84 28.1 32 60.8 78 27.0 56 7.51 42 17.2 42 7.29 26
MDP-Flow [26]53.5 35.5 50 65.0 60 32.4 55 20.6 37 43.8 18 24.4 58 18.0 45 43.4 46 19.9 55 14.9 76 41.8 64 11.5 78 30.8 44 39.6 43 25.3 45 23.8 55 57.4 71 22.2 49 31.0 43 59.3 72 16.8 18 10.8 72 26.5 72 10.9 59
OFH [38]54.6 41.9 75 61.6 53 48.5 88 14.7 5 42.7 14 14.1 6 17.4 39 47.6 57 12.6 8 10.6 35 38.5 48 8.39 53 34.8 68 43.7 67 34.5 84 27.2 73 61.9 93 29.5 83 21.3 13 57.6 68 21.4 37 12.2 78 29.8 87 16.2 79
DeepFlow2 [108]54.6 39.1 64 66.4 64 44.4 81 20.1 33 47.9 43 20.9 36 23.8 73 52.8 69 26.3 80 12.2 46 43.2 70 7.64 36 31.4 47 41.7 58 22.9 28 18.2 18 52.4 41 17.9 11 29.6 38 44.7 6 39.0 78 16.2 97 31.2 95 22.7 99
CPM-Flow [116]56.2 35.2 45 67.0 66 25.3 23 25.8 80 53.6 84 28.2 85 25.5 76 58.5 90 27.5 82 12.4 50 48.0 85 8.11 46 33.9 59 44.2 71 27.1 51 17.6 10 51.8 36 19.1 16 21.3 13 51.7 36 22.6 43 9.92 65 27.2 77 11.3 61
S2F-IF [123]56.8 35.7 51 69.0 76 26.6 32 24.1 67 54.0 92 26.0 70 25.6 78 60.9 97 25.9 76 12.4 50 46.6 82 8.38 52 34.3 64 44.2 71 27.7 56 18.0 17 53.2 50 19.2 18 21.2 12 51.7 36 22.9 44 8.97 55 24.9 68 9.11 50
PGM-C [120]57.0 35.2 45 67.1 68 25.4 25 25.8 80 53.6 84 28.2 85 25.8 79 59.3 95 27.5 82 12.4 50 48.1 86 8.16 47 33.9 59 44.3 74 27.1 51 17.7 12 52.8 48 19.1 16 20.8 9 50.3 28 22.4 41 9.82 63 27.2 77 11.5 65
BriefMatch [124]57.2 34.1 42 59.4 50 30.2 47 17.1 17 43.6 17 16.1 15 14.8 13 36.3 19 13.4 13 9.41 19 34.3 32 6.26 16 33.4 54 41.2 54 31.8 76 40.3 122 63.5 96 42.7 125 47.2 102 61.1 79 59.1 126 12.7 81 23.4 61 21.6 97
EpicFlow [102]59.0 35.2 45 67.2 69 25.3 23 25.8 80 53.8 87 28.2 85 26.1 81 60.1 96 27.5 82 12.4 50 48.1 86 8.10 45 34.2 63 44.5 76 28.0 58 17.8 14 52.7 46 19.4 23 21.0 10 52.1 39 22.5 42 10.4 69 27.3 79 12.7 68
FlowFields+ [130]59.5 35.8 54 69.2 78 26.0 30 25.6 78 55.3 100 27.7 82 26.7 82 62.9 102 27.7 86 12.7 57 45.9 80 8.84 60 34.0 62 44.2 71 26.6 49 17.6 10 54.9 58 19.0 15 20.6 8 53.9 49 22.3 40 9.31 59 26.5 72 8.79 47
CostFilter [40]59.7 43.6 83 65.7 62 40.8 74 20.8 38 45.9 30 21.0 37 18.8 49 44.9 51 18.3 34 17.3 87 39.5 53 13.9 89 26.7 16 32.8 11 22.7 26 30.9 89 60.0 85 31.6 89 59.7 126 81.5 142 59.3 128 4.31 10 11.9 16 5.93 8
RFlow [90]60.2 44.4 85 75.2 100 50.5 92 16.5 14 42.1 13 17.2 21 21.4 61 52.2 65 16.0 20 11.3 40 36.2 41 7.25 29 35.2 72 44.3 74 32.3 77 24.3 60 55.6 62 22.6 53 38.6 68 55.3 55 41.6 86 13.4 84 29.9 88 16.9 84
FlowFields [110]61.5 35.7 51 68.7 74 25.8 29 25.6 78 55.1 99 27.7 82 26.8 85 62.8 101 27.6 85 13.0 59 46.8 83 9.13 64 34.6 67 44.9 80 28.1 59 17.8 14 54.9 58 19.4 23 21.3 13 54.8 53 24.2 49 9.25 57 26.4 71 8.47 42
TV-L1-improved [17]62.5 27.7 29 57.9 45 20.5 13 18.2 22 44.9 24 19.1 27 19.4 53 47.7 58 17.0 28 10.1 25 38.5 48 6.75 22 35.9 79 46.0 88 27.3 53 43.7 129 70.2 123 47.5 130 51.4 111 60.5 77 50.3 114 10.3 68 26.8 76 10.8 58
DMF_ROB [140]62.7 43.0 80 71.7 88 45.5 84 22.4 57 48.3 45 24.0 52 28.0 93 62.4 100 26.9 81 12.8 58 45.8 79 7.80 40 33.9 59 43.1 61 30.0 71 20.9 25 56.4 65 21.2 42 22.3 19 42.0 3 27.0 56 12.7 81 27.6 80 17.1 85
Steered-L1 [118]64.9 38.7 63 67.9 70 43.1 79 11.6 1 34.7 1 12.3 2 16.3 32 41.3 39 13.9 14 12.1 44 38.6 50 8.30 50 34.5 66 43.4 64 32.5 79 29.4 83 61.1 92 25.9 71 60.6 129 67.0 103 70.1 139 15.9 94 30.6 91 24.0 101
Sparse Occlusion [54]65.2 38.6 62 61.8 55 32.5 57 26.0 84 48.8 47 29.5 93 20.0 55 45.2 52 19.2 40 14.3 71 38.4 46 9.67 66 34.4 65 42.6 60 26.7 50 25.4 67 52.4 41 22.4 52 67.3 137 75.9 131 48.3 108 8.07 48 19.9 49 7.36 28
MLDP_OF [89]66.4 48.8 94 77.3 103 52.2 94 20.1 33 49.6 57 19.3 31 23.5 72 54.6 73 18.9 37 12.3 49 38.6 50 7.65 37 33.5 57 40.9 51 29.1 69 28.4 79 55.9 63 31.6 89 49.1 106 62.2 86 60.5 130 7.91 47 16.8 38 8.95 48
DeepFlow [86]66.7 47.3 91 71.9 89 64.0 111 21.4 42 48.2 44 22.7 42 27.9 90 58.2 87 31.6 95 15.1 77 42.7 66 10.6 74 31.5 49 42.1 59 22.6 24 19.6 20 56.7 67 19.4 23 27.6 31 46.2 12 39.7 80 20.6 106 35.3 112 28.0 109
PWC-Net_ROB [148]67.1 49.3 96 75.1 99 39.6 71 28.7 95 54.0 92 29.4 92 27.2 86 58.7 91 31.5 94 15.6 79 38.4 46 8.76 59 36.5 85 45.6 87 28.3 60 26.2 70 60.3 88 26.4 74 13.0 2 49.3 22 4.98 3 7.79 44 19.4 47 7.36 28
WRT [151]67.3 38.4 61 61.6 53 26.8 34 31.9 100 58.0 108 32.7 96 28.0 93 56.1 78 21.7 65 14.6 73 41.9 65 9.03 63 30.7 42 37.0 34 24.9 42 35.7 103 62.5 94 31.8 93 34.3 53 63.2 89 37.1 73 7.23 39 15.2 31 7.62 34
TF+OM [100]68.2 39.8 67 55.2 41 30.4 49 20.4 36 41.0 8 23.5 46 19.5 54 39.4 31 28.0 87 18.4 90 37.0 44 18.0 96 35.0 69 41.1 52 39.6 99 29.6 84 52.6 44 29.1 82 49.0 105 66.8 102 43.6 93 14.4 86 29.3 84 18.0 87
CombBMOF [113]69.8 42.4 76 71.5 86 31.3 53 25.2 74 52.9 81 25.1 64 17.9 43 45.8 53 16.1 21 14.2 69 41.7 62 11.4 77 33.5 57 40.2 47 29.6 70 34.5 98 59.9 84 37.3 107 55.2 121 69.9 112 46.5 105 6.78 35 16.8 38 8.62 43
EPPM w/o HM [88]70.6 43.1 81 72.4 91 38.1 69 18.7 25 52.5 75 16.9 19 21.3 60 56.2 79 17.5 29 16.2 83 45.3 77 12.7 85 33.4 54 39.6 43 30.0 71 33.5 94 65.4 103 33.8 96 45.5 92 66.1 100 65.5 136 6.88 36 18.1 44 9.16 52
FF++_ROB [146]70.8 38.0 60 69.4 79 32.4 55 25.8 80 54.8 98 27.6 80 28.3 98 63.7 103 30.3 90 15.7 80 49.6 90 12.9 86 35.0 69 44.9 80 28.9 66 23.0 47 55.4 61 23.0 62 22.1 18 53.1 45 24.0 48 10.1 66 25.5 69 12.8 70
Complementary OF [21]72.5 51.9 103 74.9 98 59.3 104 14.2 3 41.6 10 13.7 3 20.4 56 46.6 55 19.7 49 22.2 97 40.8 58 21.0 101 36.0 81 43.4 64 38.5 97 33.9 97 63.8 97 31.6 89 31.1 44 51.9 38 36.2 71 18.9 103 34.4 110 29.4 111
Aniso. Huber-L1 [22]73.0 33.9 40 65.1 61 32.8 59 34.0 101 54.0 92 40.0 104 27.9 90 55.0 75 38.4 101 15.2 78 49.9 91 12.0 80 35.3 74 44.6 77 28.5 61 23.9 56 55.9 63 20.7 35 50.6 109 62.1 85 39.7 80 8.15 49 20.7 51 8.39 40
TCOF [69]74.9 45.0 86 70.0 80 51.5 93 25.5 77 53.7 86 26.7 73 26.7 82 56.2 79 32.0 96 21.9 96 43.1 69 22.2 103 37.8 92 48.9 104 25.7 47 18.8 19 44.7 7 20.0 30 52.1 115 67.5 105 25.8 50 10.7 71 26.7 74 11.4 63
Rannacher [23]75.1 43.1 81 71.0 84 45.2 83 24.1 67 49.4 52 26.4 72 26.0 80 56.9 83 26.0 78 14.2 69 42.7 66 10.5 73 37.1 89 47.9 99 30.7 74 32.3 91 65.2 102 27.0 76 44.0 89 56.0 62 39.7 80 7.83 45 21.2 55 9.19 53
ComplOF-FED-GPU [35]75.7 49.5 97 75.7 102 55.3 97 15.3 7 47.0 38 13.8 4 21.1 59 52.7 68 16.1 21 17.1 86 40.6 57 14.2 90 35.7 78 45.1 84 32.5 79 35.3 101 67.5 114 34.4 99 46.5 99 59.0 71 50.8 115 12.8 83 29.6 85 16.6 82
ACK-Prior [27]76.2 55.9 105 72.7 93 59.3 104 17.5 19 43.3 15 16.0 14 17.8 41 42.3 41 16.3 25 17.6 88 41.3 60 12.0 80 35.3 74 41.1 52 35.9 87 37.4 115 59.6 82 34.3 98 59.8 128 61.1 79 74.7 140 17.7 101 29.2 83 27.4 104
F-TV-L1 [15]77.2 66.8 115 84.2 113 77.3 122 27.1 91 52.0 71 29.1 91 27.2 86 57.3 85 24.2 74 24.1 102 52.0 96 19.5 99 39.3 99 47.7 98 39.3 98 24.2 59 56.5 66 24.9 68 33.2 51 53.7 48 20.1 34 6.69 34 18.6 45 5.94 9
LDOF [28]77.8 41.7 74 70.7 81 47.2 87 24.0 66 53.9 91 24.6 60 26.7 82 58.4 89 25.5 75 15.8 81 57.4 108 10.2 71 36.0 81 45.3 85 34.8 85 22.1 36 58.1 75 21.3 43 30.6 41 56.8 64 23.4 45 22.5 115 38.6 121 30.2 112
ROF-ND [107]78.0 49.2 95 71.5 86 49.4 90 22.5 59 46.2 34 20.6 33 21.4 61 50.0 63 18.1 32 23.1 99 53.7 100 16.0 92 35.9 79 46.1 90 28.6 62 33.5 94 58.5 76 30.3 86 60.7 130 70.5 113 60.4 129 9.67 61 20.9 53 10.2 56
SIOF [67]79.5 50.3 100 66.0 63 47.1 86 19.8 30 48.4 46 20.7 34 29.8 100 55.1 76 32.6 98 25.9 105 48.3 88 25.0 106 37.7 90 46.4 92 36.8 90 32.9 92 58.6 77 35.6 104 37.2 64 53.1 45 18.6 26 16.8 99 33.0 100 21.3 96
LocallyOriented [52]79.8 39.4 65 60.5 51 35.7 63 27.9 94 57.8 107 28.5 89 28.1 95 58.3 88 30.6 93 13.9 65 41.3 60 10.1 69 37.8 92 47.3 96 33.0 83 24.8 66 52.5 43 28.1 79 39.4 75 62.4 87 37.5 75 15.7 92 33.0 100 18.2 89
Second-order prior [8]80.3 37.6 58 70.9 83 37.2 67 22.4 57 51.2 64 23.7 48 24.5 75 59.1 94 22.6 68 11.2 39 41.2 59 8.48 56 37.9 94 49.6 112 28.8 65 29.0 81 68.8 118 26.0 72 55.5 122 64.7 94 52.7 119 14.7 89 34.1 108 17.3 86
Brox et al. [5]80.5 43.7 84 74.4 96 56.4 101 27.0 90 52.5 75 30.5 95 23.4 71 54.5 72 23.3 71 13.4 60 50.0 92 8.66 58 39.8 102 46.5 93 47.6 121 21.6 31 59.8 83 22.8 60 30.9 42 59.3 72 7.61 5 23.1 119 37.0 117 33.4 121
FlowNetS+ft+v [112]80.9 36.8 57 67.0 66 39.4 70 25.3 76 52.1 73 27.6 80 27.8 89 57.2 84 35.5 99 13.5 61 50.9 94 9.44 65 40.1 103 49.5 111 36.8 90 20.9 25 57.0 68 20.8 37 46.3 97 66.3 101 41.0 84 17.4 100 34.3 109 24.1 102
NL-TV-NCC [25]81.2 46.1 88 68.3 72 43.4 80 25.2 74 55.3 100 23.5 46 21.8 66 47.2 56 16.9 27 16.9 85 44.1 72 11.9 79 38.5 96 49.1 108 27.3 53 36.5 109 65.7 105 34.9 102 46.2 94 75.7 129 45.7 99 12.6 79 28.8 81 9.10 49
SRR-TVOF-NL [91]81.9 47.6 92 69.1 77 41.9 77 23.3 61 52.7 79 23.3 44 25.5 76 56.8 82 25.9 76 13.5 61 47.9 84 8.09 44 36.9 88 43.8 69 32.9 82 25.8 68 57.7 72 22.6 53 62.8 134 75.2 126 49.1 111 21.0 108 29.7 86 31.6 115
CRTflow [80]81.9 40.9 69 72.5 92 36.2 65 21.3 41 49.4 52 21.8 38 22.9 68 57.5 86 19.0 38 14.0 67 44.7 74 10.3 72 35.4 77 45.0 83 30.4 73 46.6 134 73.4 128 53.8 135 38.8 70 65.5 96 38.2 76 19.5 105 38.5 120 27.6 107
DF-Auto [115]82.0 41.6 73 64.2 58 32.8 59 42.0 111 58.8 111 49.6 113 34.8 107 62.2 99 47.2 112 20.7 94 53.4 99 14.5 91 36.2 84 44.7 78 37.2 94 15.1 3 42.9 6 17.4 9 45.3 91 69.0 109 13.0 11 24.2 123 37.7 118 31.9 116
DPOF [18]82.5 45.3 87 68.9 75 37.6 68 26.7 89 56.9 103 26.9 76 24.3 74 54.6 73 26.2 79 18.6 91 54.6 103 13.6 88 33.3 53 43.1 61 28.9 66 27.6 77 60.8 89 26.3 73 47.2 102 55.3 55 76.0 142 14.3 85 31.0 94 15.3 76
TriangleFlow [30]82.8 41.5 72 63.2 56 42.6 78 21.4 42 52.4 74 20.2 32 21.7 65 53.6 71 16.2 23 14.8 74 44.4 73 10.9 75 43.2 117 52.9 125 43.4 110 36.8 113 65.9 107 38.8 111 42.2 84 65.4 95 41.8 88 15.8 93 35.2 111 22.5 98
Bartels [41]84.2 48.6 93 63.2 56 61.4 109 23.4 63 44.0 19 27.1 78 21.4 61 44.6 49 23.8 73 26.2 106 43.0 68 25.4 107 36.8 86 44.7 78 41.7 106 33.7 96 60.0 85 41.7 122 52.6 116 67.2 104 61.1 131 11.2 74 23.4 61 16.3 80
SuperFlow [81]84.8 35.2 45 61.1 52 34.4 61 35.2 102 53.5 83 42.5 105 27.7 88 52.5 67 43.5 109 27.5 107 60.5 114 27.6 110 36.0 81 43.3 63 42.2 107 22.6 43 59.2 81 22.2 49 46.2 94 62.0 84 23.4 45 21.8 113 36.0 113 32.5 118
Dynamic MRF [7]85.0 49.5 97 78.0 104 55.8 100 17.2 18 47.4 40 16.5 18 21.6 64 56.3 81 16.2 23 14.8 74 46.4 81 12.5 83 41.2 109 49.0 106 45.1 115 35.3 101 70.7 124 38.5 110 37.1 63 57.7 69 55.1 121 21.3 110 36.7 116 31.2 114
CBF [12]85.2 41.4 70 74.0 95 48.5 88 40.2 108 51.5 68 51.7 115 22.9 68 50.8 64 28.5 88 14.3 71 44.7 74 11.2 76 38.3 95 46.1 90 36.1 88 26.3 71 55.1 60 24.9 68 61.5 131 71.0 114 52.0 118 11.6 77 26.7 74 14.8 75
CLG-TV [48]85.6 41.4 70 68.0 71 40.8 74 37.0 106 53.1 82 45.3 107 30.9 101 58.7 91 40.2 103 22.8 98 62.0 116 19.3 98 39.0 98 47.6 97 38.2 96 27.2 73 61.0 91 27.1 77 46.2 94 57.8 70 29.3 62 9.74 62 24.6 67 9.11 50
CNN-flow-warp+ref [117]86.0 42.6 78 71.2 85 49.8 91 31.6 99 53.8 87 37.3 101 32.7 103 63.8 104 42.7 107 16.0 82 55.1 104 12.1 82 38.6 97 46.0 88 43.8 113 23.3 49 59.1 80 23.6 64 23.5 22 50.9 33 21.9 39 24.2 123 36.2 115 32.9 119
Local-TV-L1 [65]86.2 56.8 107 79.1 105 74.5 116 39.5 107 53.8 87 46.1 109 38.1 108 66.2 106 43.1 108 23.9 101 52.9 98 21.1 102 32.3 51 41.4 57 27.5 55 23.2 48 54.8 57 22.7 58 25.8 25 47.5 15 33.4 68 26.9 126 40.5 125 40.5 130
HBM-GC [105]88.7 73.5 127 79.7 106 79.4 125 30.0 96 49.7 59 33.6 97 29.3 99 47.8 59 30.4 91 35.4 115 45.2 76 33.3 115 30.5 41 35.1 21 32.6 81 34.6 99 52.0 37 35.4 103 70.9 140 80.1 137 62.6 132 8.83 54 19.1 46 13.0 72
TriFlow [95]91.7 47.1 90 66.5 65 41.1 76 31.2 98 49.6 57 37.3 101 27.9 90 52.3 66 39.4 102 24.7 103 49.3 89 22.3 104 37.7 90 43.5 66 43.1 109 28.4 79 52.7 46 29.0 81 76.7 144 73.7 119 99.5 147 16.2 97 30.5 90 20.1 92
OFRF [134]92.9 50.4 101 64.5 59 55.4 99 41.2 110 57.5 105 46.2 111 34.5 106 58.9 93 40.6 104 29.6 109 45.6 78 28.8 111 30.7 42 40.2 47 22.3 22 31.1 90 57.9 73 30.4 88 43.5 87 62.5 88 51.9 117 29.8 130 39.3 122 49.1 138
p-harmonic [29]93.9 50.4 101 86.6 123 56.5 103 27.1 91 54.5 95 28.9 90 32.9 104 69.3 112 30.4 91 19.8 92 65.1 119 16.2 93 39.6 101 47.2 95 40.3 101 30.4 87 66.5 109 32.2 94 45.6 93 64.4 92 28.9 61 10.2 67 24.1 66 13.2 73
Learning Flow [11]94.6 42.9 79 70.7 81 44.8 82 30.2 97 54.7 96 34.7 98 28.2 96 55.9 77 32.3 97 17.6 88 57.1 107 12.6 84 44.0 120 52.6 123 47.8 124 30.5 88 64.6 99 30.3 86 46.9 100 63.3 90 42.8 92 14.7 89 31.6 98 16.3 80
Fusion [6]96.9 40.5 68 75.6 101 45.9 85 20.0 31 50.0 61 22.5 40 20.8 58 52.8 69 22.8 69 16.2 83 52.8 97 13.5 87 43.1 115 49.0 106 47.5 118 39.6 121 67.8 116 43.8 127 63.9 135 75.0 125 46.3 104 35.5 136 42.6 131 53.3 142
Shiralkar [42]98.9 46.1 88 85.6 119 54.3 96 19.7 29 57.7 106 18.2 23 28.2 96 70.8 113 19.3 41 20.5 93 59.0 110 18.4 97 39.5 100 49.7 114 36.3 89 40.4 123 76.1 130 41.2 121 51.9 113 65.8 98 64.2 134 21.0 108 42.4 130 25.3 103
StereoFlow [44]99.4 95.9 147 96.0 146 97.4 147 88.3 147 96.2 147 86.2 143 82.6 144 94.8 145 73.7 141 91.4 146 96.3 146 90.3 145 53.0 136 61.6 140 52.8 129 11.2 1 39.3 3 11.7 1 10.5 1 42.5 4 1.70 1 11.5 75 23.6 63 18.0 87
LiteFlowNet [143]100.6 62.6 109 88.2 130 55.3 97 35.7 103 65.1 122 36.9 100 39.6 109 74.5 116 38.1 100 23.4 100 50.3 93 17.4 94 43.1 115 51.4 120 42.2 107 35.7 103 69.3 120 33.9 97 41.2 80 75.2 126 17.8 23 14.4 86 28.9 82 16.7 83
ContinualFlow_ROB [153]100.7 67.9 118 87.6 127 62.0 110 54.1 124 68.0 126 60.3 123 52.2 126 80.5 125 54.5 122 32.5 111 61.0 115 26.1 109 47.3 129 57.0 128 40.4 102 43.1 128 69.7 122 48.7 131 18.7 6 49.0 20 6.92 4 11.5 75 23.8 65 12.8 70
SegOF [10]103.2 56.3 106 71.9 89 37.1 66 57.3 127 62.9 119 68.3 129 46.0 117 69.0 111 57.2 127 41.0 118 59.5 112 37.2 117 43.5 118 48.3 102 56.4 133 38.2 119 69.6 121 39.1 112 17.9 5 64.5 93 3.40 2 22.7 118 33.0 100 32.0 117
StereoOF-V1MT [119]104.0 49.7 99 86.0 120 56.4 101 21.2 39 68.8 127 16.3 17 32.2 102 80.7 126 20.8 59 21.3 95 66.1 122 17.4 94 47.0 128 57.0 128 47.5 118 41.7 126 81.2 135 40.9 118 38.6 68 68.1 107 48.6 109 23.2 120 42.2 129 27.7 108
Ad-TV-NDC [36]104.8 73.7 128 85.4 117 89.5 141 56.9 126 60.4 115 67.5 128 51.0 123 75.9 118 57.6 129 45.7 121 65.7 120 47.9 125 35.3 74 45.3 85 28.9 66 27.3 76 57.2 70 28.2 80 34.6 55 55.0 54 27.3 58 34.0 133 48.7 137 47.5 135
EAI-Flow [152]105.5 71.2 122 88.2 130 74.6 117 35.9 104 59.6 113 38.0 103 40.3 110 74.4 115 44.4 110 33.3 112 55.4 105 31.8 113 40.7 108 49.7 114 36.9 92 37.2 114 65.8 106 39.3 114 57.5 124 72.5 117 39.3 79 11.1 73 25.5 69 12.7 68
WOLF_ROB [149]106.7 56.9 108 87.9 128 59.5 106 35.9 104 66.0 124 35.7 99 42.5 112 78.3 120 44.7 111 24.7 103 59.5 112 22.9 105 41.2 109 49.1 108 43.4 110 37.7 117 72.0 127 35.9 105 34.3 53 61.3 82 27.4 59 22.5 115 40.1 124 33.0 120
Modified CLG [34]106.9 68.7 119 80.5 108 76.1 120 52.0 122 61.0 116 63.6 125 51.9 124 79.4 121 55.6 124 47.4 124 72.1 128 46.7 123 41.2 109 49.7 114 46.0 116 26.0 69 64.7 100 26.7 75 31.4 47 55.6 58 19.9 33 29.0 129 43.8 133 39.9 129
IAOF2 [51]108.2 54.9 104 73.7 94 53.9 95 42.6 112 58.3 109 50.7 114 33.9 105 61.9 98 42.1 106 64.4 134 75.7 131 74.3 136 41.5 112 49.9 118 37.1 93 36.4 108 64.0 98 34.4 99 59.7 126 69.6 111 41.3 85 19.4 104 33.4 105 23.0 100
AugFNG_ROB [144]110.9 72.6 126 79.9 107 59.6 107 59.5 128 72.0 131 68.5 130 55.6 130 83.3 129 56.1 125 34.1 114 59.3 111 30.3 112 49.8 132 58.3 132 47.4 117 37.9 118 71.3 125 39.6 115 35.5 58 73.9 120 8.00 7 15.9 94 33.2 103 18.4 90
Filter Flow [19]111.3 62.9 110 74.4 96 60.8 108 42.8 113 60.1 114 49.4 112 42.5 112 66.0 105 51.2 116 52.1 130 69.5 125 50.2 126 44.8 123 49.7 114 54.4 131 41.9 127 66.7 110 43.6 126 74.3 143 88.9 145 42.6 91 10.6 70 21.6 58 12.6 67
GroupFlow [9]113.0 66.4 113 85.2 116 80.8 128 61.6 131 75.4 135 69.0 131 51.9 124 83.6 130 57.0 126 33.5 113 63.9 117 32.5 114 49.6 131 61.0 135 39.9 100 51.3 139 81.7 136 59.4 139 22.8 21 51.1 34 16.5 16 28.0 128 41.9 128 37.9 127
SPSA-learn [13]113.1 65.1 111 87.4 126 72.7 115 45.7 118 59.3 112 53.4 118 45.2 116 74.7 117 52.2 118 41.6 119 69.9 127 42.5 119 42.7 114 48.9 104 48.7 126 38.8 120 69.0 119 42.5 124 39.1 73 61.9 83 19.3 28 36.0 137 45.6 134 48.3 136
2D-CLG [1]113.5 77.2 131 82.3 111 75.4 119 61.5 130 65.9 123 73.7 135 63.2 136 89.6 138 60.8 134 82.8 143 88.3 139 86.8 143 43.5 118 49.3 110 54.8 132 35.1 100 67.5 114 36.1 106 21.3 13 50.8 32 15.5 14 34.4 135 46.3 136 46.0 132
IAOF [50]113.5 66.3 112 81.3 110 77.8 123 50.1 121 58.4 110 59.7 122 45.0 115 74.1 114 49.6 115 50.8 127 68.2 123 58.0 132 40.2 104 48.7 103 37.8 95 36.7 110 66.9 111 33.5 95 54.9 119 63.5 91 40.8 83 30.1 132 41.3 127 43.5 131
TVL1_ROB [139]113.7 88.5 142 92.7 137 96.3 145 69.8 139 66.8 125 83.7 141 67.9 137 90.2 140 70.7 139 77.3 139 89.0 141 80.6 139 40.3 106 49.6 112 40.9 103 24.0 57 65.5 104 27.6 78 21.8 17 57.3 67 10.2 9 36.2 138 51.1 139 47.2 134
BlockOverlap [61]113.9 77.2 131 86.0 120 82.8 131 48.2 119 55.8 102 57.6 121 46.9 119 66.9 107 51.9 117 49.1 126 54.1 101 51.2 127 36.8 86 41.3 56 47.5 118 40.5 124 59.0 79 39.6 115 68.9 139 80.2 138 65.1 135 20.8 107 30.8 92 34.9 123
HBpMotionGpu [43]114.0 67.0 116 80.7 109 72.3 114 55.3 125 57.3 104 66.7 127 44.7 114 67.2 109 54.1 121 39.5 117 57.8 109 38.4 118 42.0 113 48.2 101 48.3 125 35.9 105 60.9 90 39.2 113 65.1 136 72.0 116 50.1 113 22.6 117 32.5 99 36.9 124
LFNet_ROB [150]114.7 72.3 123 93.8 141 65.3 112 45.6 117 78.1 140 45.9 108 48.8 122 87.7 135 41.8 105 31.5 110 66.0 121 25.6 108 47.8 130 57.1 130 47.6 121 37.6 116 71.5 126 38.0 108 49.3 107 74.5 123 26.4 55 15.9 94 33.5 106 20.9 94
GraphCuts [14]116.1 66.6 114 87.0 124 80.0 127 43.1 114 63.0 120 46.1 109 41.8 111 67.0 108 53.4 120 28.5 108 64.0 118 20.8 100 40.2 104 48.1 100 43.5 112 46.5 132 63.4 95 40.5 117 62.7 133 75.4 128 69.5 138 23.8 122 33.3 104 38.5 128
Black & Anandan [4]116.5 70.3 121 88.0 129 84.1 132 45.5 116 61.4 117 52.0 116 47.4 120 77.3 119 52.9 119 42.3 120 77.5 132 42.8 120 44.0 120 51.8 121 45.0 114 35.9 105 75.9 129 38.3 109 50.8 110 71.3 115 17.8 23 29.8 130 42.7 132 37.6 126
EPMNet [133]118.5 78.2 133 91.5 135 78.0 124 61.6 131 75.7 136 69.6 132 55.3 129 79.5 122 57.3 128 48.1 125 56.6 106 47.7 124 45.3 124 54.0 126 41.1 104 36.7 110 67.3 112 41.0 119 56.4 123 81.4 141 26.3 53 18.6 102 36.0 113 20.7 93
FlowNet2 [122]119.0 78.8 134 86.2 122 79.4 125 63.5 136 70.2 129 72.8 134 57.2 133 82.0 127 59.7 133 46.2 122 51.4 95 45.6 122 45.3 124 54.0 126 41.1 104 36.7 110 67.3 112 41.0 119 67.6 138 80.8 139 55.1 121 14.5 88 30.0 89 13.9 74
ResPWCR_ROB [145]119.5 78.9 135 92.8 139 75.1 118 40.4 109 63.5 121 43.5 106 48.1 121 80.2 123 48.8 114 39.2 116 69.2 124 35.6 116 44.0 120 50.5 119 49.2 127 44.6 131 76.7 131 47.0 129 54.6 118 78.7 134 31.4 65 23.4 121 38.0 119 30.8 113
2bit-BM-tele [98]121.0 82.4 138 87.2 125 91.8 142 44.4 115 54.7 96 52.9 117 46.0 117 68.3 110 47.4 113 51.5 129 54.4 102 53.4 130 40.5 107 47.1 94 47.7 123 47.9 135 66.4 108 53.1 133 71.7 141 83.6 143 75.1 141 21.9 114 39.3 122 29.2 110
Nguyen [33]121.5 75.6 130 85.4 117 85.8 134 67.0 137 61.6 118 83.0 140 57.1 132 80.2 123 64.1 137 70.8 135 80.2 134 77.4 138 45.9 127 52.2 122 56.4 133 36.3 107 68.1 117 42.0 123 41.4 81 66.0 99 19.7 31 34.1 134 45.6 134 46.1 133
SILK [79]122.5 72.5 124 85.1 115 88.3 138 61.9 133 71.8 130 73.7 135 54.9 128 85.3 131 58.0 131 53.6 131 69.8 126 54.6 131 52.8 134 57.9 131 61.8 137 46.5 132 77.6 132 48.9 132 31.3 45 54.3 52 38.9 77 37.8 139 50.6 138 49.7 139
UnFlow [129]124.9 89.6 144 94.1 142 86.4 135 72.2 140 83.9 143 77.9 138 71.4 141 93.2 142 69.8 138 54.8 132 74.4 129 51.4 128 62.0 142 66.9 143 69.1 144 50.0 137 80.9 133 57.1 136 53.2 117 69.0 109 7.68 6 15.6 91 30.8 92 21.0 95
Periodicity [78]125.8 68.9 120 83.5 112 65.5 113 52.2 123 69.8 128 57.0 120 78.4 143 82.5 128 87.2 145 47.2 123 74.7 130 45.5 121 69.7 147 81.9 147 65.6 141 59.5 142 84.9 143 60.7 140 36.3 62 79.8 136 19.2 27 40.7 141 66.5 146 53.1 141
Horn & Schunck [3]126.3 74.1 129 93.2 140 86.9 136 49.1 120 73.8 132 53.9 119 53.1 127 89.0 137 54.6 123 50.9 128 81.4 135 52.4 129 51.3 133 58.8 133 54.3 130 41.2 125 82.3 137 44.6 128 55.0 120 74.3 122 19.6 30 40.7 141 56.8 141 48.8 137
Heeger++ [104]128.9 86.1 140 91.0 134 77.1 121 67.6 138 91.6 146 65.1 126 85.6 146 94.6 144 83.6 144 71.0 136 88.2 138 68.8 134 62.9 143 69.8 145 67.3 142 69.4 146 91.0 146 70.8 144 40.2 78 80.9 140 26.3 53 21.7 111 31.2 95 27.4 104
SLK [47]130.0 67.5 117 90.3 133 82.1 129 72.2 140 84.7 144 84.8 142 58.4 134 94.0 143 58.1 132 78.1 140 82.5 136 84.6 142 55.4 139 61.5 138 68.1 143 49.4 136 83.7 141 57.7 137 36.2 61 68.1 107 26.2 52 50.3 144 60.0 143 65.2 146
FFV1MT [106]131.7 85.0 139 92.0 136 84.4 133 60.6 129 83.8 142 60.8 124 85.4 145 92.2 141 88.2 146 71.2 137 89.9 143 69.1 135 64.0 145 69.3 144 78.0 146 69.2 145 91.5 147 73.2 146 51.7 112 73.9 120 45.3 96 21.7 111 31.2 95 27.4 104
FOLKI [16]132.4 72.5 124 84.5 114 87.5 137 62.3 134 74.9 134 74.0 137 56.9 131 87.2 133 57.7 130 59.8 133 78.1 133 65.4 133 53.3 137 61.2 137 62.7 138 50.9 138 81.0 134 62.7 141 47.3 104 73.3 118 56.8 125 49.0 143 62.9 144 64.3 145
TI-DOFE [24]132.8 90.7 145 94.6 144 97.1 146 76.9 145 79.5 141 89.4 146 73.1 142 96.1 146 74.4 142 84.6 144 93.6 145 88.3 144 52.9 135 59.9 134 63.9 139 44.5 130 83.6 140 53.5 134 42.9 85 68.0 106 17.0 21 50.5 145 65.5 145 62.4 143
H+S_ROB [138]133.4 79.9 137 90.1 132 82.1 129 76.2 144 90.8 145 81.2 139 71.3 140 96.3 147 61.1 135 90.5 145 89.3 142 92.5 146 58.3 141 63.2 141 71.1 145 59.0 141 85.7 145 64.1 142 32.5 50 78.0 133 25.8 50 55.9 146 57.0 142 62.6 144
PGAM+LK [55]138.6 79.4 136 92.7 137 88.7 139 62.9 135 76.8 139 71.9 133 59.9 135 88.3 136 62.8 136 74.9 138 90.6 144 75.6 137 54.4 138 61.0 135 64.6 140 58.1 140 82.7 138 58.6 138 77.6 145 85.6 144 78.1 144 38.8 140 52.7 140 50.7 140
Adaptive flow [45]139.3 91.3 146 95.7 145 96.2 143 75.8 143 75.9 137 86.2 143 68.9 138 85.6 132 72.4 140 80.6 142 85.2 137 83.9 141 57.3 140 61.5 138 60.2 136 66.7 143 84.1 142 70.4 143 90.2 146 92.7 146 95.0 145 27.7 127 40.5 125 37.0 125
HCIC-L [99]140.2 88.6 143 96.7 147 89.3 140 80.8 146 74.4 133 93.1 147 86.6 147 87.2 133 95.2 147 95.8 147 97.4 147 96.9 147 63.4 144 66.7 142 57.8 135 68.9 144 82.7 138 73.1 145 96.0 147 95.9 147 98.6 146 25.3 125 33.9 107 34.7 122
Pyramid LK [2]142.7 86.7 141 94.5 143 96.2 143 73.1 142 76.5 138 86.4 145 70.8 139 89.6 138 78.5 143 78.1 140 88.6 140 82.7 140 68.8 146 76.4 146 80.4 147 75.7 147 85.1 144 78.1 147 73.1 142 79.6 135 69.3 137 60.8 147 74.5 147 79.9 147
AdaConv-v1 [126]148.3 100.0 148 100.0 149 100.0 148 100.0 148 100.0 148 100.0 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 149 99.9 149 99.9 149 99.9 149 100.0 149 99.7 149 99.9 149 99.9 148 99.9 148 99.9 148
SepConv-v1 [127]148.3 100.0 148 100.0 149 100.0 148 100.0 148 100.0 148 100.0 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 149 99.9 149 99.9 149 99.9 149 100.0 149 99.7 149 99.9 149 99.9 148 99.9 148 99.9 148
SuperSlomo [132]148.3 100.0 148 100.0 149 100.0 148 100.0 148 100.0 148 100.0 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 149 99.9 149 99.9 149 99.9 149 100.0 149 99.7 149 99.9 149 99.9 148 99.9 148 99.9 148
FGIK [136]148.3 100.0 148 100.0 149 100.0 148 100.0 148 100.0 148 100.0 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 149 99.9 149 99.9 149 99.9 149 100.0 149 99.7 149 99.9 149 99.9 148 99.9 148 99.9 148
CtxSyn [137]148.3 100.0 148 100.0 149 100.0 148 100.0 148 100.0 148 100.0 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 149 99.9 149 99.9 149 99.9 149 100.0 149 99.7 149 99.9 149 99.9 148 99.9 148 99.9 148
CyclicGen [154]148.3 100.0 148 100.0 149 100.0 148 100.0 148 100.0 148 100.0 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 148 99.9 149 99.9 149 99.9 149 99.9 149 100.0 149 99.7 149 99.9 149 99.9 148 99.9 148 99.9 148
AVG_FLOW_ROB [142]148.8 100.0 148 99.9 148 100.0 148 100.0 148 100.0 148 100.0 148 99.9 148 99.9 148 99.9 148 100.0 154 99.9 148 100.0 154 99.9 148 100.0 154 99.8 148 99.2 148 99.5 148 98.3 148 99.6 148 97.0 148 99.7 148 99.9 148 99.9 148 99.9 148
Move the mouse over the numbers in the table to see the corresponding images. Click to compare with the ground truth.

References

Methodtime*framescolor 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. IEEE TIP 26(8):4055-4067, 2017.
[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] IIOF-NLDP 150 2 color D.-H. Trinh, W. Blondel, and C. Daul. A general form of illumination-invariant descriptors in variational optical flow estimation. ICIP 2017.
[132] SuperSlomo 0.5 2 color Anonymous. (Interpolation results only.) Super SloMo: High quality estimation of multiple intermediate frames for video interpolation. CVPR 2018 submission 325.
[133] EPMNet 0.061 2 color Anonymous. EPM-convolution multilayer-network for optical flow estimation. ICME 2018 submission 1119.
[134] OFRF 90 2 color T. Mai, M. Gouiffes, and S. Bouchafa. Optical Flow refinement using iterative propagation under color, proximity and flow reliability constraints. Submitted to Signal, Image and Video Processing 2017.
[135] 3DFlow 328 2 color J. Chen, Z. Cai, J. Lai, and X. Xie. A filtering based framework for optical flow estimation. IEEE TCSVT 2018.
[136] FGIK 0.18 2 color Anonymous. (Interpolation results only.) Learning flow-guided interpolation kernels for video frame synthesis. ECCV 2018 submission 433.
[137] CtxSyn 0.07 2 color S. Niklaus and F. Liu. (Interpolation results only.) Context-aware synthesis for video frame interpolation. CVPR 2018.
[138] H+S_ROB 5 2 color ROB 2018 baseline submission, based on: E. Meinhardt-Llopis, J. Sanchez, and D. Kondermann. Horn-Schunck optical flow with a multi-scale strategy. Image Processing On Line 3:151–172, 2013.
[139] TVL1_ROB 1 2 color ROB 2018 baseline submission, based on: J. Sanchez, E. Meinhardt-Llopis, and G. Facciolo. TV-L1 optical flow estimation. Image Processing On Line 3:137-150, 2013.
[140] DMF_ROB 10 2 color ROB 2018 baseline submission, based on: P. Weinzaepfel, J. Revaud, Z. Harchaoui, and C. Schmid. DeepFlow: large displacement optical flow with deep matching. ICCV 2013.
[141] JOF 657 2 gray C. Zhang, L. Ge, Z. Chen, M. Li, W. Liu, and H. Chen. Refined TV-L1 optical flow estimation using joint filtering. Submitted to IEEE TMM, 2018.
[142] AVG_FLOW_ROB N/A 2 N/A Average flow field of ROB 2018 training set.
[143] LiteFlowNet 0.06 2 color T.-W. Hui, X. Tang, and C. C. Loy. LiteFlowNet: A lightweight convolutional neural network for optical flow estimation. CVPR 2018.
[144] AugFNG_ROB 0.10 all color Anonymous. FusionNet and AugmentedFlowNet: Selective proxy ground truth for training on unlabeled images. ECCV 2018 submission 2834.
[145] ResPWCR_ROB 0.2 2 color Anonymous. Learning optical flow with residual connections. ROB 2018 submission.
[146] FF++_ROB 17.43 2 color R. Schuster, C. Bailer, O. Wasenmueller, D. Stricker. FlowFields++: Accurate optical flow correspondences meet robust interpolation. ICIP 2018. Submitted to ROB 2018.
[147] ProFlow_ROB 76 3 color Anonymous. ProFlow: Learning to predict optical flow. BMVC 2018 submission 277.
[148] PWC-Net_ROB 0.069 2 color D. Sun, X. Yang, M.-Y. Liu, and J. Kautz. PWC-Net: CNNs for optical flow using pyramid, warping, and cost volume. CVPR 2018.
[149] WOLF_ROB 0.02 2 color Anonymous. Reversed deep neural network for optical flow. ROB 2018 submission.
[150] LFNet_ROB 0.068 2 color Anonymous. Learning a flow network. ROB 2018 submission.
[151] WRT 9 2 color L. Mei, J. Lai, X. Xie, J. Zhu, and J. Chen. Illumination-invariance optical flow estimation using weighted regularization transform. Submitted to IEEE TCSVT 2018.
[152] EAI-Flow 2.1 2 color Anonymous. Hierarchical coherency sensitive hashing and interpolation with RANSAC for large displacement optical flow. CVIU 2018 submission 17-678.
[153] ContinualFlow_ROB 0.5 all color Anonymous. Continual Flow. ACCV 2018 submission 273.
[154] CyclicGen 0.088 2 color Anonymous. (Interpolation results only.) Deep video frame interpolation using cyclic frame generation. AAAI 2019 submission 323.
* The "time" column lists the reported runtime in seconds on the "Urban" sequence. Note that these runtimes are not normalized by processor speed or type.