Optical flow evaluation results Statistics:     Average   SD   R2.5   R5.0   R10.0   A90   A95   A99  
Error type:   endpoint   angle   interpolation   normalized interpolation  
Show images: below table   above table        
A99
interpolation
error
avg. Mequon
(Hidden texture)
im0   GT   im1
Schefflera
(Hidden texture)
im0   GT   im1
Urban
(Synthetic)
im0   GT   im1
Teddy
(Stereo)
im0   GT   im1
Backyard
(High-speed camera)
im0   GT   im1
Basketball
(High-speed camera)
im0   GT   im1
Dumptruck
(High-speed camera)
im0   GT   im1
Evergreen
(High-speed camera)
im0   GT   im1
rank all disc untext all disc untext all disc untext all disc untext all disc untext all disc untext all disc untext all disc untext
CtxSyn [137]3.7 9.38 2 14.7 2 2.58 2 11.5 1 16.1 2 3.65 22 9.04 2 12.7 1 3.00 2 12.6 1 19.2 1 7.33 3 38.7 4 47.5 3 9.56 6 22.9 3 38.0 2 4.76 4 31.6 5 64.3 4 3.92 6 24.8 3 36.2 3 3.37 5
PMMST [114]13.8 11.2 14 21.1 9 2.71 3 13.8 10 19.7 9 3.65 22 10.3 4 19.2 11 2.71 1 16.8 6 30.8 26 7.53 35 41.1 9 51.1 11 10.0 21 24.6 6 43.0 10 4.93 19 34.2 14 70.9 16 4.04 12 28.8 18 45.4 30 3.42 14
MDP-Flow2 [68]14.1 11.0 7 20.7 8 2.71 3 13.9 12 19.9 11 3.46 2 10.3 4 20.3 18 3.00 2 16.7 4 30.0 18 7.35 8 41.0 8 50.7 8 10.1 33 27.1 42 44.9 24 4.97 26 33.6 7 70.1 11 3.92 6 29.2 22 47.0 41 3.42 14
SuperSlomo [132]20.9 9.66 3 16.1 3 3.37 122 15.4 38 20.4 17 6.06 134 8.43 1 14.4 2 3.00 2 17.1 8 23.8 3 8.58 122 38.5 3 47.8 4 8.76 4 23.1 4 40.8 5 4.80 6 30.2 3 62.5 3 3.87 3 25.2 4 37.4 4 3.32 3
CyclicGen [154]21.9 8.29 1 11.7 1 3.46 131 11.6 2 15.3 1 5.94 131 9.68 3 15.0 3 3.42 95 16.7 4 21.0 2 10.3 139 29.8 1 37.0 1 8.04 1 12.4 1 21.2 1 4.65 2 18.4 1 39.9 1 3.70 1 16.6 1 22.9 1 2.94 1
NNF-Local [87]24.2 11.4 20 21.6 13 2.71 3 12.8 3 18.4 4 3.56 4 10.4 6 20.0 16 3.00 2 19.8 65 37.3 99 7.35 8 41.5 15 51.4 13 10.0 21 28.2 71 47.3 44 5.07 55 34.5 17 71.9 29 4.04 12 29.1 21 46.1 36 3.37 5
SepConv-v1 [127]25.4 9.68 4 19.1 4 2.52 1 15.4 38 20.1 14 5.26 116 11.0 13 16.7 5 3.87 115 20.4 79 26.8 4 9.59 137 41.9 16 52.5 20 9.00 5 24.7 7 42.4 7 4.69 3 30.7 4 67.4 5 3.92 6 24.7 2 35.8 2 3.32 3
NN-field [71]27.5 11.5 28 22.9 27 2.71 3 13.0 5 18.6 5 3.42 1 12.3 86 19.7 12 3.00 2 21.1 90 39.8 116 7.44 21 41.4 13 51.4 13 10.0 21 27.5 51 46.4 35 4.97 26 33.8 10 71.0 18 4.04 12 29.3 24 46.2 37 3.37 5
FGIK [136]30.8 10.7 5 21.1 9 3.00 71 15.6 45 20.3 15 5.35 118 14.1 106 16.0 4 3.56 100 20.7 84 29.1 11 9.76 138 33.3 2 41.7 2 8.43 3 21.2 2 40.2 4 4.55 1 25.6 2 56.0 2 3.70 1 25.8 5 39.6 6 3.16 2
NNF-EAC [103]31.6 11.5 28 21.7 14 3.11 87 14.5 25 21.0 29 3.70 25 12.3 86 22.6 40 3.00 2 17.7 19 32.4 46 7.55 42 43.2 43 55.1 47 10.1 33 25.1 10 43.8 13 4.90 12 34.0 12 70.5 13 4.08 44 29.4 27 47.5 47 3.42 14
DeepFlow2 [108]34.4 11.4 20 23.5 31 3.00 71 16.7 69 23.0 73 4.04 62 11.0 13 20.3 18 3.00 2 19.0 52 29.8 16 7.53 35 42.7 29 54.0 29 10.3 58 25.0 8 43.0 10 4.93 19 35.2 33 73.8 41 4.04 12 28.9 19 44.9 26 3.56 79
DeepFlow [86]34.5 11.3 18 24.2 45 3.00 71 16.6 68 23.0 73 4.32 79 11.0 13 20.3 18 3.00 2 19.3 55 28.1 9 7.59 50 42.7 29 54.5 34 10.2 53 25.2 12 44.1 14 5.00 48 32.9 6 68.2 6 4.04 12 28.4 14 44.6 21 3.56 79
PH-Flow [101]35.5 11.9 61 25.7 75 2.83 23 13.3 7 19.7 9 3.56 4 10.7 8 22.7 41 3.00 2 16.5 3 30.2 19 7.33 3 42.3 22 52.1 17 10.1 33 28.7 88 50.9 103 5.20 82 35.6 40 77.0 67 4.04 12 29.6 33 47.0 41 3.51 59
CombBMOF [113]36.2 12.0 68 24.3 46 2.83 23 14.3 19 20.6 21 3.56 4 11.3 31 25.7 68 3.00 2 20.3 78 34.9 74 7.55 42 43.2 43 54.0 29 10.1 33 26.4 29 47.7 51 4.90 12 36.2 63 71.4 22 4.08 44 29.5 30 45.7 33 3.37 5
SuperFlow [81]36.3 11.0 7 22.1 21 3.11 87 17.1 76 22.7 63 4.69 95 11.7 51 18.7 6 3.37 77 18.7 44 27.4 6 7.70 72 41.3 11 51.2 12 9.98 17 26.3 28 46.9 38 4.80 6 34.7 21 76.0 55 4.08 44 28.1 12 41.5 9 3.42 14
DF-Auto [115]37.4 10.9 6 19.2 5 3.11 87 17.2 81 23.4 83 4.43 85 10.4 6 20.6 26 3.00 2 18.1 30 29.7 13 7.55 42 41.4 13 52.1 17 10.0 21 26.2 23 47.2 41 4.97 26 35.2 33 79.3 80 4.08 44 29.6 33 44.7 22 3.56 79
LME [70]39.7 11.4 20 22.0 19 2.71 3 15.1 33 21.8 41 3.87 53 11.3 31 36.0 136 3.00 2 17.4 11 32.0 43 7.48 25 44.5 68 57.0 66 11.4 142 27.6 53 47.2 41 4.97 26 33.6 7 69.7 8 4.04 12 30.0 40 48.6 59 3.42 14
CBF [12]39.9 11.0 7 19.8 6 3.00 71 17.1 76 22.9 69 4.24 76 12.0 66 19.0 8 3.00 2 17.8 24 28.0 8 7.85 89 40.6 7 49.9 6 9.97 13 26.2 23 44.6 19 4.97 26 36.3 65 76.3 60 4.12 96 27.9 8 41.2 8 3.70 124
Aniso. Huber-L1 [22]40.3 11.4 20 21.7 14 3.11 87 19.7 120 24.7 119 4.55 90 12.0 66 19.7 12 3.11 71 18.4 36 29.8 16 7.55 42 42.5 25 54.4 33 9.98 17 25.2 12 42.2 6 4.83 8 35.6 40 71.5 23 4.04 12 27.9 8 42.0 11 3.56 79
IROF++ [58]40.7 11.9 61 24.1 42 2.83 23 14.7 28 21.3 31 3.56 4 12.1 83 29.0 97 3.00 2 16.3 2 27.9 7 7.35 8 43.9 55 56.0 55 11.1 103 26.4 29 47.0 40 4.93 19 34.5 17 72.3 30 4.08 44 30.3 50 49.3 67 3.56 79
WLIF-Flow [93]40.8 11.5 28 22.1 21 2.83 23 15.2 34 21.6 37 3.79 43 11.3 31 26.4 77 3.00 2 17.4 11 30.3 21 7.59 50 42.5 25 53.5 25 10.4 67 29.0 94 51.1 106 5.29 100 34.8 24 69.7 8 4.04 12 30.0 40 48.4 55 3.46 44
FMOF [94]42.3 12.2 90 24.5 55 2.94 57 14.0 13 20.0 12 3.56 4 12.3 86 27.7 85 3.00 2 19.8 65 35.4 79 7.70 72 42.4 23 52.1 17 10.1 33 28.1 67 49.1 68 4.93 19 34.6 20 72.7 35 3.87 3 30.2 46 47.6 50 3.42 14
CLG-TV [48]42.6 11.1 12 21.8 17 3.11 87 18.8 102 24.0 99 4.43 85 11.3 31 20.0 16 3.70 101 18.6 42 28.9 10 7.72 78 42.8 32 55.0 46 10.0 21 25.0 8 42.9 9 4.93 19 36.0 55 71.6 24 4.04 12 29.0 20 44.0 17 3.56 79
IROF-TV [53]43.8 11.7 42 24.7 62 3.00 71 15.5 42 22.0 50 3.70 25 11.0 13 23.7 51 3.00 2 17.3 9 31.3 30 7.57 49 43.8 53 56.0 55 11.2 112 27.6 53 48.4 58 4.97 26 35.9 53 74.5 50 4.08 44 28.0 10 42.6 13 3.56 79
Brox et al. [5]44.2 11.4 20 24.9 67 2.94 57 15.9 52 22.2 53 4.04 62 11.3 31 21.0 27 3.37 77 18.4 36 27.0 5 7.59 50 42.2 20 53.3 23 10.0 21 28.2 71 51.5 109 5.00 48 36.8 69 88.0 111 4.04 12 28.4 14 42.3 12 3.42 14
ALD-Flow [66]45.4 12.0 68 28.4 106 3.11 87 16.3 60 22.8 65 3.83 48 11.0 13 21.7 33 3.00 2 17.9 27 33.6 59 7.39 15 43.4 48 54.6 39 10.8 89 25.8 17 44.8 23 5.00 48 34.1 13 70.4 12 4.04 12 31.9 83 50.3 78 3.46 44
nLayers [57]45.6 11.8 50 22.9 27 2.83 23 14.1 15 20.4 17 3.56 4 11.0 13 19.7 12 3.00 2 18.3 34 34.2 69 7.39 15 46.7 124 60.1 120 11.0 97 27.9 60 50.1 78 5.20 82 35.5 38 72.6 34 4.08 44 30.8 55 49.3 67 3.42 14
MDP-Flow [26]47.9 11.2 14 21.2 11 2.71 3 14.2 17 20.5 20 3.70 25 10.7 8 19.0 8 3.00 2 19.7 63 32.4 46 7.70 72 44.2 59 57.0 66 11.2 112 30.0 118 51.4 108 5.51 131 36.1 60 72.9 38 4.08 44 30.8 55 48.4 55 3.42 14
JOF [141]48.0 12.0 68 23.6 33 3.11 87 14.0 13 20.0 12 3.70 25 11.0 13 23.8 54 3.00 2 18.1 30 31.5 35 7.35 8 44.7 71 57.6 75 11.3 131 29.5 110 50.1 78 5.07 55 34.5 17 71.6 24 4.04 12 31.2 62 50.3 78 3.51 59
Layers++ [37]48.0 11.4 20 21.7 14 2.94 57 12.8 3 18.2 3 3.46 2 11.0 13 26.7 80 3.00 2 17.7 19 32.9 51 7.53 35 46.6 122 60.9 132 10.6 81 30.9 132 60.2 141 5.00 48 34.9 30 72.7 35 3.87 3 29.9 39 47.5 47 3.46 44
LDOF [28]48.9 11.4 20 22.5 24 3.56 134 16.1 54 21.4 36 6.35 140 12.0 66 20.3 18 3.70 101 19.0 52 29.7 13 7.94 94 41.2 10 50.9 9 10.1 33 26.8 35 50.2 81 4.90 12 34.8 24 80.2 84 4.08 44 29.4 27 44.5 19 3.46 44
p-harmonic [29]48.9 11.4 20 23.5 31 2.83 23 19.1 108 24.3 108 4.80 101 11.3 31 22.0 35 3.70 101 20.9 88 31.7 37 7.62 56 42.6 28 54.2 31 10.1 33 25.7 16 43.5 12 5.07 55 36.1 60 71.8 26 4.08 44 29.6 33 46.5 38 3.51 59
Second-order prior [8]49.7 11.3 18 22.0 19 3.11 87 19.0 107 24.2 106 4.32 79 13.3 99 27.7 85 3.70 101 18.8 47 31.6 36 7.51 29 42.9 34 54.7 41 10.0 21 26.2 23 45.0 25 4.97 26 35.6 40 71.2 19 4.04 12 29.5 30 45.4 30 3.56 79
COFM [59]50.2 11.8 50 24.3 46 2.94 57 14.5 25 20.9 26 3.65 22 11.0 13 26.4 77 3.00 2 17.4 11 32.3 44 7.35 8 44.2 59 55.1 47 10.1 33 30.0 118 54.4 129 5.20 82 35.8 48 79.3 80 4.08 44 31.2 62 48.8 62 3.51 59
SIOF [67]50.3 11.7 42 23.1 29 3.11 87 19.4 115 24.8 122 4.76 98 11.3 31 25.7 68 3.11 71 18.4 36 31.4 32 8.04 101 40.3 6 50.3 7 9.95 11 25.8 17 45.3 27 4.97 26 33.9 11 71.2 19 4.08 44 30.0 40 47.4 44 3.70 124
Local-TV-L1 [65]51.2 11.2 14 21.5 12 3.56 134 19.6 118 24.4 111 5.57 125 11.0 13 19.1 10 3.00 2 18.3 34 30.4 24 7.87 92 42.8 32 54.5 34 10.2 53 26.2 23 44.7 20 5.45 119 34.2 14 76.1 58 4.08 44 28.0 10 42.8 15 3.65 119
ProbFlowFields [128]51.8 11.6 34 25.4 71 2.83 23 14.4 21 21.1 30 3.56 4 10.7 8 23.7 51 3.00 2 18.4 36 33.4 56 7.59 50 46.2 112 59.2 106 11.2 112 28.5 84 50.7 97 5.32 106 34.7 21 76.9 65 4.08 44 29.4 27 46.5 38 3.46 44
FlowFields [110]51.9 11.8 50 25.6 74 2.83 23 14.4 21 20.9 26 3.56 4 11.3 31 24.3 59 3.00 2 20.0 71 38.1 104 7.51 29 43.6 50 54.5 34 11.0 97 28.2 71 50.7 97 5.16 77 34.8 24 75.1 54 4.04 12 32.0 89 52.0 103 3.46 44
EAI-Flow [152]52.8 12.5 111 26.8 89 2.83 23 15.8 50 21.8 41 4.20 74 12.3 86 30.4 115 3.00 2 19.3 55 34.0 66 7.39 15 44.9 78 57.1 70 11.1 103 26.1 22 46.0 30 5.00 48 36.0 55 72.4 32 4.08 44 29.3 24 45.2 28 3.37 5
TV-L1-MCT [64]54.8 12.4 106 24.7 62 2.83 23 16.4 61 23.1 75 3.83 48 11.9 65 32.7 125 3.00 2 17.6 16 31.7 37 7.53 35 47.0 133 61.2 133 11.0 97 25.5 14 44.7 20 4.97 26 36.0 55 80.7 89 4.04 12 28.4 14 44.8 24 3.46 44
Sparse-NonSparse [56]56.0 12.0 68 24.3 46 2.83 23 15.0 31 21.3 31 3.56 4 11.7 51 29.0 97 3.00 2 17.6 16 29.7 13 7.39 15 45.7 96 59.3 107 11.0 97 28.8 89 48.7 64 5.07 55 38.6 95 90.1 122 4.04 12 32.4 98 51.8 99 3.42 14
HAST [109]56.6 11.7 42 23.6 33 2.94 57 13.8 10 19.6 8 3.56 4 12.0 66 31.7 121 3.00 2 17.8 24 31.7 37 7.14 1 45.3 86 57.0 66 9.97 13 33.7 145 62.8 148 5.10 72 38.4 90 88.4 113 4.04 12 33.0 108 51.0 86 3.42 14
BlockOverlap [61]57.1 11.1 12 20.1 7 3.56 134 19.3 112 23.7 92 6.16 136 11.3 31 20.4 24 3.70 101 18.4 36 29.6 12 8.72 126 43.1 40 54.5 34 10.2 53 27.4 48 48.6 61 5.35 113 34.8 24 72.8 37 4.08 44 27.2 7 40.9 7 3.56 79
OAR-Flow [125]57.4 12.0 68 24.9 67 3.00 71 16.4 61 22.4 56 4.08 67 11.0 13 20.5 25 3.00 2 17.4 11 33.6 59 7.33 3 46.2 112 60.0 119 11.3 131 27.0 38 47.6 48 5.23 90 37.6 79 74.0 44 4.08 44 31.0 60 49.2 65 3.46 44
CPM-Flow [116]57.6 11.8 50 27.3 93 2.83 23 14.4 21 20.4 17 3.70 25 11.7 51 24.0 55 3.00 2 21.4 98 40.1 119 7.77 82 45.5 92 58.1 86 11.2 112 26.6 34 48.0 54 5.07 55 36.0 55 72.3 30 4.04 12 30.9 58 50.4 80 3.56 79
Modified CLG [34]58.7 11.0 7 21.9 18 3.11 87 19.6 118 23.9 96 5.94 131 12.4 91 26.3 75 3.87 115 19.8 65 30.8 26 8.12 106 42.1 18 52.9 21 10.1 33 27.0 38 48.1 56 5.23 90 34.7 21 70.8 15 4.08 44 29.5 30 45.3 29 3.56 79
FlowFields+ [130]58.7 11.8 50 26.1 83 2.71 3 14.1 15 20.6 21 3.70 25 11.2 30 24.8 64 3.00 2 20.1 73 40.2 121 7.53 35 45.5 92 58.0 83 11.2 112 28.6 87 50.6 93 5.20 82 35.6 40 77.5 72 4.04 12 32.2 93 52.5 107 3.42 14
AGIF+OF [85]58.8 12.2 90 24.3 46 2.71 3 15.2 34 21.8 41 3.70 25 11.7 51 27.7 85 3.00 2 18.0 28 33.0 53 7.55 42 45.8 100 58.8 103 11.2 112 30.0 118 53.4 123 5.07 55 35.4 35 74.8 52 3.92 6 32.2 93 52.6 110 3.37 5
2DHMM-SAS [92]59.0 12.2 90 24.5 55 2.83 23 17.9 90 24.1 103 3.87 53 12.0 66 28.7 94 3.00 2 17.3 9 31.4 32 7.51 29 45.1 82 58.2 92 11.2 112 27.9 60 49.0 66 4.83 8 37.0 71 76.1 58 4.08 44 31.9 83 50.5 81 3.42 14
ComponentFusion [96]59.6 12.0 68 29.6 115 2.71 3 14.5 25 21.3 31 3.56 4 11.0 13 22.0 35 3.00 2 18.8 47 36.2 92 7.33 3 45.5 92 58.2 92 10.7 86 27.2 43 46.3 33 4.97 26 40.5 118 93.3 132 4.12 96 34.4 125 58.3 135 3.42 14
F-TV-L1 [15]60.1 12.0 68 26.5 87 3.56 134 19.2 110 24.7 119 4.83 104 11.7 51 21.5 31 4.00 117 19.3 55 32.7 49 7.68 65 43.1 40 55.3 50 9.83 7 25.1 10 42.8 8 5.07 55 34.8 24 74.0 44 4.16 104 28.5 17 42.7 14 3.56 79
AdaConv-v1 [126]60.7 15.0 143 28.2 104 3.70 138 17.6 87 20.7 24 7.68 148 17.4 124 22.0 35 7.00 141 27.5 134 33.7 64 17.0 149 39.9 5 49.8 5 8.19 2 23.8 5 39.5 3 4.76 4 34.2 14 68.5 7 4.12 96 26.9 6 39.5 5 3.42 14
TC/T-Flow [76]61.0 12.4 106 26.4 85 2.83 23 16.5 66 23.1 75 3.83 48 11.0 13 22.4 39 3.00 2 18.9 49 34.5 70 7.33 3 45.5 92 58.1 86 11.4 142 27.3 47 47.6 48 4.93 19 41.1 120 80.4 87 4.20 111 30.9 58 49.7 71 3.37 5
DPOF [18]61.7 12.3 99 29.4 113 3.11 87 13.3 7 19.1 7 3.56 4 15.7 114 25.2 66 3.70 101 19.4 58 37.5 101 7.59 50 43.1 40 54.6 39 10.0 21 29.1 99 49.7 72 4.90 12 36.6 67 77.0 67 4.08 44 31.5 73 50.5 81 3.51 59
PGM-C [120]63.2 11.8 50 27.3 93 2.83 23 14.4 21 20.7 24 3.70 25 12.3 86 23.0 45 3.00 2 20.6 83 42.3 127 7.62 56 45.8 100 59.5 113 11.2 112 27.2 43 47.4 45 4.97 26 37.1 73 79.2 77 4.04 12 32.4 98 55.0 123 3.51 59
PMF [73]63.5 12.2 90 25.9 78 2.71 3 15.4 38 21.8 41 3.56 4 12.7 93 35.7 134 3.00 2 20.2 76 35.9 85 7.51 29 44.4 66 54.9 45 10.1 33 28.4 77 50.5 91 5.32 106 37.9 84 81.1 92 4.04 12 34.2 122 54.1 117 3.37 5
Ramp [62]63.6 12.0 68 24.6 58 2.94 57 14.8 29 21.3 31 3.70 25 11.7 51 29.4 103 3.00 2 16.9 7 30.3 21 7.39 15 45.4 89 58.5 95 11.0 97 30.2 125 50.9 103 5.23 90 39.8 110 89.6 119 4.04 12 32.4 98 52.5 107 3.42 14
TF+OM [100]63.8 11.6 34 30.1 120 3.11 87 15.0 31 21.6 37 4.04 62 11.7 51 24.0 55 3.00 2 21.3 94 39.0 114 7.68 65 44.3 62 56.7 64 10.3 58 28.8 89 50.4 88 5.07 55 37.7 80 83.5 103 4.08 44 29.2 22 46.0 34 3.56 79
Ad-TV-NDC [36]64.2 12.2 90 22.5 24 4.32 147 20.6 140 24.8 122 5.80 126 11.7 51 21.6 32 3.37 77 21.6 99 31.8 40 8.04 101 42.5 25 53.4 24 9.97 13 26.4 29 47.6 48 5.16 77 36.8 69 70.9 16 4.08 44 28.3 13 41.8 10 3.70 124
ProFlow_ROB [147]64.8 11.8 50 27.5 97 2.83 23 15.8 50 22.6 59 3.79 43 11.4 47 20.3 18 3.00 2 18.9 49 37.0 97 7.35 8 47.3 135 61.8 141 11.2 112 25.5 14 44.2 15 4.83 8 40.0 115 81.4 94 4.08 44 34.3 124 56.5 130 3.56 79
ComplOF-FED-GPU [35]64.9 12.0 68 27.9 101 2.94 57 15.7 48 22.2 53 3.79 43 16.0 115 21.4 29 3.70 101 18.4 36 33.6 59 7.48 25 44.9 78 57.7 76 10.7 86 27.4 48 45.9 29 5.00 48 36.6 67 78.7 76 4.08 44 32.6 106 52.3 105 3.51 59
AggregFlow [97]64.9 13.7 131 37.1 138 3.11 87 16.2 58 22.6 59 4.04 62 11.0 13 23.3 50 3.00 2 21.8 101 40.7 123 7.66 63 43.2 43 53.5 25 10.3 58 27.0 38 46.0 30 5.00 48 38.0 85 82.4 100 4.08 44 31.9 83 51.9 102 3.42 14
OFLAF [77]65.3 11.7 42 24.5 55 2.71 3 13.6 9 20.3 15 3.56 4 11.0 13 23.0 45 3.00 2 17.6 16 31.3 30 7.39 15 47.3 135 61.7 138 11.2 112 29.6 111 51.9 116 5.32 106 41.8 127 95.6 137 4.16 104 33.6 114 52.1 104 3.42 14
S2F-IF [123]65.5 12.1 82 29.8 117 2.71 3 14.2 17 20.6 21 3.56 4 11.3 31 26.3 75 3.00 2 20.2 76 40.1 119 7.53 35 45.9 105 58.7 102 11.3 131 28.4 77 50.7 97 5.20 82 35.7 45 76.0 55 4.08 44 32.3 96 53.1 111 3.46 44
Classic+NL [31]65.9 12.1 82 24.3 46 3.00 71 15.3 37 21.8 41 3.70 25 11.7 51 29.4 103 3.00 2 17.4 11 31.4 32 7.53 35 45.7 96 59.4 109 10.8 89 29.0 94 49.8 75 5.10 72 39.6 107 90.4 124 4.08 44 32.2 93 51.8 99 3.46 44
Classic++ [32]66.6 11.6 34 23.7 35 3.11 87 17.8 89 24.4 111 4.08 67 11.7 51 20.3 18 3.37 77 20.1 73 33.8 65 7.62 56 44.7 71 57.8 79 10.0 21 28.0 63 49.7 72 5.35 113 37.4 77 81.4 94 4.08 44 30.7 54 49.5 69 3.56 79
DMF_ROB [140]66.7 11.9 61 25.4 71 3.00 71 17.1 76 22.8 65 4.08 67 19.4 132 29.7 106 3.70 101 20.4 79 34.5 70 7.68 65 45.3 86 58.1 86 11.1 103 26.4 29 45.6 28 4.97 26 35.7 45 73.8 41 4.08 44 31.2 62 50.2 73 3.42 14
LSM [39]67.8 12.3 99 24.7 62 2.83 23 15.4 38 21.9 47 3.56 4 12.0 66 30.3 113 3.00 2 18.7 44 33.2 55 7.44 21 46.1 110 59.4 109 11.1 103 29.3 103 51.9 116 5.07 55 39.2 101 91.0 127 4.04 12 32.3 96 52.5 107 3.42 14
FC-2Layers-FF [74]67.9 12.1 82 26.0 82 2.83 23 13.0 5 18.7 6 3.56 4 11.4 47 25.7 68 3.00 2 17.8 24 33.5 58 7.48 25 46.5 118 60.3 125 11.2 112 30.4 128 52.3 121 5.32 106 39.8 110 90.0 121 4.08 44 31.8 79 51.6 95 3.46 44
FlowNetS+ft+v [112]68.4 11.5 28 23.7 35 3.46 131 19.9 125 24.6 117 7.87 150 12.0 66 21.1 28 3.37 77 19.5 60 30.6 25 8.91 129 43.7 51 56.6 63 11.2 112 26.0 19 44.5 18 4.97 26 38.6 95 87.8 109 4.08 44 30.0 40 46.0 34 3.51 59
MLDP_OF [89]68.8 11.9 61 24.7 62 2.83 23 17.4 84 23.8 94 3.87 53 10.7 8 24.6 62 3.00 2 20.5 82 33.6 59 8.35 115 44.1 57 56.5 61 10.1 33 29.3 103 50.5 91 5.57 132 35.8 48 73.4 40 4.20 111 31.2 62 50.6 84 3.70 124
RNLOD-Flow [121]70.0 11.8 50 24.6 58 2.89 53 17.3 83 24.0 99 3.74 39 12.7 93 36.0 136 3.11 71 18.1 30 31.2 29 7.48 25 45.8 100 59.6 114 11.1 103 29.3 103 50.6 93 5.16 77 35.4 35 74.1 47 4.08 44 32.0 89 51.6 95 3.42 14
TCOF [69]70.2 12.0 68 24.7 62 2.83 23 20.3 135 26.4 150 5.07 109 11.1 29 29.0 97 3.00 2 17.7 19 32.4 46 7.68 65 43.2 43 55.5 51 9.97 13 28.8 89 46.3 33 5.07 55 41.2 123 94.9 135 4.08 44 31.8 79 51.3 90 3.70 124
Fusion [6]70.3 11.6 34 24.3 46 2.89 53 15.6 45 21.9 47 3.83 48 11.0 13 23.7 51 3.37 77 21.0 89 33.4 56 7.62 56 44.1 57 56.3 59 10.1 33 30.3 127 54.1 127 5.45 119 38.0 85 83.7 104 4.08 44 34.0 120 54.7 119 3.56 79
CRTflow [80]70.5 11.7 42 24.4 53 3.32 118 19.5 117 24.9 126 4.51 87 12.0 66 22.7 41 4.00 117 18.1 30 30.3 21 7.68 65 45.0 80 58.1 86 11.3 131 26.0 19 45.1 26 4.97 26 37.7 80 87.9 110 4.08 44 30.8 55 50.2 73 3.56 79
RFlow [90]70.7 11.6 34 24.3 46 3.00 71 19.3 112 24.8 122 4.36 81 11.6 50 29.7 106 3.37 77 20.0 71 36.1 88 7.72 78 43.0 35 55.2 49 10.1 33 27.9 60 51.8 115 4.97 26 37.1 73 82.8 102 4.08 44 31.6 75 49.5 69 3.56 79
Sparse Occlusion [54]71.2 11.7 42 25.9 78 3.00 71 18.1 94 24.6 117 3.83 48 11.3 31 22.7 41 3.11 71 18.7 44 34.1 68 7.70 72 45.0 80 58.0 83 11.1 103 28.5 84 44.2 15 5.26 94 39.3 104 83.7 104 3.92 6 31.9 83 51.7 97 3.56 79
S2D-Matching [84]71.6 12.3 99 25.7 75 2.94 57 17.2 81 23.7 92 4.00 59 11.7 51 28.7 94 3.00 2 17.7 19 31.9 42 7.55 42 46.8 128 60.1 120 10.4 67 30.0 118 51.5 109 5.29 100 37.0 71 77.7 73 4.04 12 31.8 79 50.9 85 3.46 44
TC-Flow [46]72.3 12.0 68 30.3 122 2.89 53 16.8 71 23.4 83 3.92 58 11.7 51 21.4 29 3.00 2 19.5 60 36.1 88 8.12 106 46.5 118 59.8 117 11.3 131 27.0 38 48.4 58 5.26 94 35.5 38 74.6 51 4.04 12 33.3 111 54.5 118 3.51 59
SVFilterOh [111]72.4 11.9 61 26.1 83 2.94 57 14.3 19 20.9 26 3.70 25 12.0 66 26.7 80 3.00 2 19.9 69 36.1 88 7.62 56 46.7 124 59.8 117 11.4 142 30.7 131 55.1 130 5.07 55 36.0 55 77.2 69 4.04 12 32.4 98 53.2 113 3.51 59
HBM-GC [105]73.5 11.8 50 23.8 38 3.11 87 16.8 71 24.2 106 3.87 53 10.7 8 18.7 6 3.00 2 18.9 49 32.9 51 7.68 65 46.8 128 60.8 129 11.5 149 34.5 149 61.7 143 5.48 127 37.7 80 81.9 99 4.04 12 30.5 53 47.8 51 3.51 59
3DFlow [135]73.8 12.4 106 27.1 92 2.83 23 15.5 42 22.1 51 3.87 53 13.7 101 24.0 55 3.00 2 19.2 54 38.7 111 7.68 65 44.0 56 56.0 55 10.1 33 31.2 135 53.8 126 5.60 134 39.5 105 79.2 77 4.16 104 31.1 61 48.0 52 3.56 79
Black & Anandan [4]74.2 12.3 99 24.0 40 3.46 131 21.2 143 25.4 134 5.35 118 18.1 127 25.0 65 5.35 132 24.4 123 34.9 74 7.77 82 42.2 20 53.5 25 10.1 33 26.9 37 46.5 36 4.97 26 39.5 105 77.2 69 4.08 44 29.3 24 42.8 15 3.56 79
Classic+CPF [83]74.2 12.2 90 24.6 58 2.83 23 15.6 45 22.1 51 3.74 39 12.0 66 30.7 116 3.00 2 17.7 19 30.9 28 7.44 21 47.2 134 61.3 134 11.2 112 31.2 135 55.9 131 5.26 94 39.9 112 88.8 116 4.04 12 33.6 114 54.0 116 3.42 14
FESL [72]75.0 12.2 90 25.1 70 2.83 23 14.9 30 21.6 37 3.70 25 12.1 83 33.7 130 3.00 2 19.7 63 35.0 76 7.72 78 46.2 112 60.2 124 11.3 131 29.3 103 50.4 88 5.32 106 39.6 107 88.6 115 3.92 6 32.4 98 51.2 88 3.42 14
CostFilter [40]75.5 13.1 125 33.1 129 2.71 3 15.2 34 21.3 31 3.56 4 14.0 104 42.7 146 3.00 2 22.0 103 44.4 134 7.26 2 45.8 100 57.2 72 10.4 67 27.2 43 48.1 56 5.45 119 39.9 112 89.4 118 4.08 44 35.6 130 56.1 128 3.37 5
Efficient-NL [60]76.3 11.8 50 23.8 38 2.83 23 16.7 69 23.3 80 3.70 25 18.4 129 29.0 97 3.70 101 19.4 58 34.0 66 7.51 29 45.1 82 58.5 95 11.1 103 30.0 118 51.5 109 5.07 55 40.1 116 88.9 117 4.08 44 33.0 108 52.4 106 3.42 14
EpicFlow [102]76.5 11.9 61 27.6 98 2.83 23 16.0 53 22.2 53 3.79 43 11.8 64 21.7 33 3.00 2 21.3 94 42.9 129 7.85 89 46.3 115 59.4 109 11.2 112 27.4 48 47.5 47 5.16 77 38.2 87 76.6 62 4.12 96 35.2 129 58.0 133 3.56 79
SRR-TVOF-NL [91]77.2 12.9 119 28.7 107 3.00 71 16.9 74 23.1 75 4.69 95 11.5 49 27.0 83 3.00 2 22.2 105 37.3 99 7.59 50 44.8 76 57.9 81 11.0 97 29.1 99 51.9 116 4.90 12 35.7 45 77.7 73 4.08 44 33.0 108 51.5 94 3.56 79
Filter Flow [19]77.8 11.8 50 23.1 29 3.37 122 20.0 127 25.1 128 5.23 115 12.2 85 26.0 72 3.70 101 22.1 104 32.7 49 7.94 94 42.1 18 51.9 15 10.4 67 28.1 67 49.0 66 5.07 55 38.4 90 81.6 96 4.16 104 30.0 40 45.5 32 3.74 140
2D-CLG [1]78.0 11.6 34 24.1 42 3.11 87 19.4 115 23.3 80 6.24 138 18.7 130 24.3 59 4.69 126 22.4 107 31.8 40 8.66 125 43.3 47 56.1 58 10.4 67 26.0 19 44.2 15 5.35 113 40.2 117 91.5 130 4.20 111 29.6 33 44.5 19 3.51 59
Bartels [41]78.2 12.2 90 29.9 118 3.37 122 17.4 84 24.3 108 4.83 104 11.3 31 24.7 63 3.70 101 21.2 91 35.4 79 9.15 135 41.3 11 51.0 10 9.87 8 29.7 113 50.2 81 6.32 150 33.7 9 70.7 14 4.20 111 30.2 46 48.4 55 3.79 143
Steered-L1 [118]78.9 11.2 14 22.6 26 2.89 53 16.2 58 22.6 59 4.55 90 21.7 134 32.4 124 5.00 129 23.4 117 38.3 106 10.7 141 44.7 71 57.4 73 9.88 9 28.0 63 48.5 60 5.32 106 37.1 73 79.2 77 4.12 96 31.4 69 51.1 87 3.51 59
Occlusion-TV-L1 [63]81.8 11.6 34 25.0 69 3.11 87 19.8 123 26.0 144 4.83 104 11.3 31 23.0 45 3.46 96 22.5 110 43.0 131 7.94 94 43.0 35 54.8 44 9.88 9 28.0 63 50.7 97 5.32 106 39.6 107 76.6 62 4.62 138 31.5 73 50.5 81 3.56 79
OFH [38]82.0 12.0 68 27.3 93 3.00 71 18.1 94 23.4 83 4.20 74 12.4 91 32.7 125 3.00 2 18.6 42 35.4 79 7.35 8 46.5 118 60.1 120 10.8 89 27.5 51 47.2 41 5.26 94 41.1 120 81.1 92 4.20 111 35.7 131 56.1 128 3.46 44
EPPM w/o HM [88]82.2 12.7 116 30.9 124 2.71 3 16.1 54 23.1 75 3.70 25 17.7 125 42.4 145 3.70 101 21.3 94 42.5 128 7.70 72 43.0 35 53.1 22 10.3 58 30.2 125 57.1 134 4.97 26 38.5 93 89.6 119 4.12 96 32.4 98 51.3 90 3.42 14
Adaptive [20]82.8 11.6 34 26.7 88 3.11 87 20.2 131 25.9 141 5.07 109 12.0 66 23.0 45 3.37 77 20.4 79 36.6 95 7.77 82 44.3 62 58.5 95 9.98 17 28.3 75 49.1 68 5.16 77 42.5 132 90.6 126 4.08 44 31.6 75 48.8 62 3.65 119
LFNet_ROB [150]83.7 13.4 128 37.5 140 2.71 3 16.1 54 21.8 41 4.08 67 12.0 66 36.3 139 3.37 77 20.7 84 36.0 86 7.94 94 45.4 89 57.8 79 11.6 152 30.6 129 57.4 135 5.20 82 34.9 30 71.8 26 4.08 44 31.3 67 49.9 72 3.70 124
CNN-flow-warp+ref [117]84.0 11.0 7 22.4 23 3.11 87 17.6 87 22.9 69 5.92 130 16.1 116 28.3 93 4.00 117 23.5 118 30.2 19 10.7 141 44.8 76 58.5 95 11.3 131 26.5 33 46.5 36 5.29 100 41.5 125 91.5 130 4.32 126 30.4 51 47.5 47 3.51 59
FF++_ROB [146]84.0 12.1 82 28.2 104 2.71 3 15.5 42 21.6 37 3.74 39 12.0 66 29.4 103 3.00 2 23.3 116 49.1 140 7.83 88 48.1 140 61.6 137 11.3 131 29.1 99 49.5 71 5.94 145 36.5 66 76.4 61 4.08 44 32.1 91 51.3 90 3.65 119
Horn & Schunck [3]85.5 12.1 82 23.7 35 3.32 118 21.4 145 25.6 137 5.89 129 17.0 121 28.2 92 5.35 132 27.3 133 37.9 102 8.04 101 42.4 23 54.3 32 10.3 58 26.2 23 44.7 20 5.07 55 40.9 119 81.7 97 4.20 111 30.2 46 44.3 18 3.70 124
IAOF [50]85.9 13.0 122 29.2 111 3.37 122 23.7 151 27.4 153 6.45 142 16.4 118 28.7 94 3.46 96 22.7 111 33.1 54 8.37 117 43.4 48 55.6 53 10.0 21 27.6 53 50.1 78 4.97 26 38.3 88 82.7 101 4.08 44 30.0 40 46.8 40 3.56 79
TriFlow [95]86.0 12.5 111 36.7 136 3.00 71 18.7 100 24.5 114 4.76 98 11.7 51 28.1 91 3.00 2 21.7 100 41.4 125 7.62 56 46.8 128 60.4 126 11.2 112 29.9 115 51.7 114 4.97 26 37.8 83 76.9 65 4.08 44 31.7 78 48.6 59 3.51 59
PWC-Net_ROB [148]86.5 13.5 130 35.8 134 2.71 3 16.4 61 23.3 80 3.74 39 12.0 66 30.3 113 3.00 2 21.3 94 45.6 135 7.51 29 48.8 147 63.0 143 11.2 112 29.7 113 53.0 122 5.29 100 35.8 48 74.9 53 4.08 44 34.2 122 56.0 127 3.51 59
TV-L1-improved [17]86.8 11.5 28 25.4 71 3.11 87 20.1 130 26.0 144 5.26 116 16.8 119 19.7 12 4.04 122 19.5 60 32.3 44 7.79 85 43.8 53 56.5 61 10.0 21 28.9 93 51.1 106 5.07 55 43.2 135 98.9 141 4.43 133 31.4 69 50.2 73 3.70 124
TVL1_ROB [139]87.1 11.9 61 26.4 85 3.70 138 21.7 146 25.8 138 6.06 134 12.0 66 25.7 68 3.46 96 23.1 113 36.3 94 8.35 115 43.0 35 54.5 34 10.1 33 28.2 71 50.0 77 5.10 72 41.3 124 91.4 129 4.24 122 29.7 37 44.8 24 3.56 79
HBpMotionGpu [43]87.3 12.3 99 32.0 127 3.79 142 20.6 140 25.4 134 6.00 133 11.3 31 26.1 74 3.00 2 23.2 114 44.0 133 7.85 89 44.3 62 56.9 65 10.8 89 29.0 94 53.5 125 5.26 94 34.9 30 69.8 10 4.04 12 31.8 79 51.4 93 3.70 124
Nguyen [33]88.4 12.0 68 25.9 78 3.37 122 21.2 143 24.5 114 6.27 139 12.7 93 28.0 89 3.70 101 23.8 119 34.7 72 8.58 122 43.0 35 54.7 41 10.1 33 27.7 58 50.7 97 4.97 26 43.4 137 93.7 133 4.43 133 30.2 46 47.4 44 3.56 79
BriefMatch [124]89.0 12.1 82 29.2 111 3.11 87 16.5 66 22.5 57 6.61 144 18.0 126 22.7 41 5.69 135 26.2 128 35.5 83 18.2 151 43.7 51 54.7 41 10.4 67 29.6 111 50.2 81 5.94 145 35.8 48 72.5 33 4.16 104 32.1 91 50.2 73 3.56 79
GraphCuts [14]89.0 13.9 136 30.2 121 3.32 118 16.4 61 22.5 57 4.36 81 33.4 149 24.1 58 5.35 132 22.3 106 34.7 72 7.87 92 44.5 68 57.0 66 9.98 17 28.3 75 50.3 86 4.90 12 38.5 93 88.2 112 4.20 111 33.9 119 53.6 115 3.56 79
FlowNet2 [122]90.6 19.1 147 47.5 148 3.11 87 17.1 76 24.1 103 4.55 90 14.2 108 29.8 109 3.37 77 23.8 119 42.9 129 8.33 113 45.9 105 58.1 86 10.6 81 27.6 53 49.4 70 4.93 19 39.2 101 81.0 90 4.08 44 31.6 75 49.2 65 3.56 79
TI-DOFE [24]91.3 12.7 116 27.6 98 3.87 146 22.2 149 25.3 130 6.66 145 14.1 106 25.3 67 4.36 124 27.7 135 38.7 111 9.06 133 42.7 29 53.6 28 10.1 33 26.8 35 48.8 65 4.97 26 38.3 88 76.0 55 4.24 122 31.9 83 44.7 22 3.87 145
AugFNG_ROB [144]92.5 13.7 131 36.6 135 3.00 71 17.5 86 22.9 69 4.80 101 14.3 110 36.0 136 3.37 77 27.8 136 64.0 148 7.96 100 48.6 146 63.0 143 11.4 142 28.0 63 51.6 113 4.83 8 36.1 60 76.6 62 4.08 44 31.9 83 47.2 43 3.42 14
ROF-ND [107]93.5 12.4 106 24.4 53 2.83 23 17.9 90 23.9 96 4.08 67 12.0 66 26.6 79 3.00 2 29.5 141 48.9 139 8.72 126 45.4 89 58.6 99 11.1 103 31.1 134 53.4 123 5.26 94 38.9 99 74.2 48 4.20 111 38.0 137 60.3 139 3.56 79
NL-TV-NCC [25]95.5 13.7 131 27.3 93 2.94 57 18.5 96 24.7 119 4.04 62 15.0 112 29.0 97 3.70 101 25.6 126 46.4 137 7.94 94 42.0 17 51.9 15 10.4 67 30.6 129 51.9 116 5.29 100 41.9 128 81.7 97 4.40 128 31.3 67 48.6 59 3.79 143
TriangleFlow [30]95.5 12.5 111 25.9 78 3.11 87 18.8 102 24.3 108 4.24 76 13.2 98 29.7 106 3.46 96 21.2 91 35.4 79 7.94 94 44.4 66 57.7 76 9.95 11 29.4 108 48.6 61 5.07 55 43.9 139 99.9 142 4.43 133 42.1 147 69.7 150 3.56 79
Correlation Flow [75]95.8 12.6 115 28.0 102 2.71 3 20.0 127 25.8 138 4.36 81 11.3 31 22.3 38 3.00 2 20.7 84 38.6 110 7.72 78 45.7 96 59.0 105 10.3 58 33.4 143 60.4 142 5.45 119 45.6 143 99.9 142 4.40 128 33.4 112 54.9 122 3.56 79
Complementary OF [21]95.9 12.4 106 34.5 132 2.83 23 16.4 61 23.5 87 3.79 43 30.7 141 32.2 123 7.05 144 19.9 69 43.9 132 7.44 21 46.9 131 60.4 126 10.7 86 28.1 67 47.7 51 5.23 90 41.1 120 80.3 86 4.12 96 42.0 145 62.0 143 3.56 79
LocallyOriented [52]96.7 12.2 90 28.1 103 3.27 116 20.5 138 25.9 141 5.07 109 14.3 110 30.0 111 3.37 77 24.2 122 41.7 126 7.66 63 44.7 71 57.1 70 10.1 33 28.8 89 47.4 45 5.48 127 42.4 130 80.6 88 4.12 96 32.4 98 51.2 88 3.56 79
IAOF2 [51]97.2 12.7 116 28.7 107 3.32 118 20.4 136 25.9 141 4.76 98 12.7 93 31.7 121 3.11 71 22.4 107 35.8 84 8.06 105 45.9 105 59.6 114 10.8 89 29.9 115 51.5 109 5.10 72 39.0 100 79.7 82 4.08 44 31.2 62 49.0 64 3.56 79
ContinualFlow_ROB [153]98.4 14.3 139 37.0 137 2.94 57 17.0 75 23.6 89 4.51 87 13.8 103 33.5 129 3.37 77 23.2 114 53.7 144 7.70 72 49.9 148 66.1 149 11.3 131 27.2 43 49.8 75 4.90 12 38.7 97 86.6 107 4.04 12 42.0 145 61.1 141 3.56 79
EPMNet [133]99.3 19.3 148 47.9 149 3.11 87 16.8 71 23.2 79 4.55 90 14.2 108 29.8 109 3.37 77 33.0 146 78.1 153 8.29 111 45.9 105 58.1 86 10.6 81 30.0 118 51.9 116 4.97 26 39.2 101 81.0 90 4.08 44 33.8 118 53.1 111 3.51 59
Aniso-Texture [82]99.8 11.5 28 24.1 42 2.83 23 20.2 131 26.0 144 4.97 107 20.0 133 24.4 61 3.37 77 26.9 131 50.7 141 9.11 134 46.1 110 60.5 128 11.4 142 32.7 142 62.2 147 5.94 145 37.3 76 80.2 84 4.04 12 34.1 121 55.0 123 3.42 14
ACK-Prior [27]101.2 12.5 111 29.7 116 2.83 23 16.1 54 22.7 63 4.00 59 25.6 137 27.7 85 5.72 137 22.4 107 36.0 86 10.7 141 45.7 96 59.3 107 11.4 142 31.8 139 50.6 93 5.35 113 38.8 98 79.9 83 4.16 104 33.5 113 51.7 97 3.70 124
Rannacher [23]101.8 11.7 42 28.7 107 3.16 115 20.4 136 26.3 148 5.07 109 19.0 131 26.0 72 4.80 128 19.8 65 38.1 104 7.79 85 44.5 68 57.4 73 10.1 33 29.0 94 50.3 86 5.20 82 42.6 133 97.0 138 4.40 128 33.7 116 55.9 126 3.70 124
IIOF-NLDP [131]102.2 12.9 119 29.0 110 2.71 3 18.6 99 24.8 122 4.08 67 13.4 100 26.7 80 3.00 2 21.9 102 39.8 116 8.16 108 45.8 100 59.4 109 10.4 67 31.6 138 59.9 139 6.06 149 54.7 152 99.9 142 6.03 151 35.7 131 57.2 132 3.42 14
LiteFlowNet [143]103.4 14.1 138 39.6 144 2.71 3 15.7 48 21.9 47 4.00 59 14.0 104 43.0 147 3.00 2 36.3 151 70.9 151 9.02 130 48.3 142 63.2 146 11.5 149 30.1 124 57.7 136 5.10 72 42.1 129 87.4 108 4.24 122 32.4 98 50.2 73 3.51 59
Learning Flow [11]103.6 12.1 82 24.6 58 3.27 116 19.7 120 25.2 129 5.00 108 39.7 151 47.7 152 7.68 146 24.6 124 35.0 76 8.19 110 45.2 85 58.6 99 10.5 80 28.4 77 48.0 54 5.45 119 38.4 90 77.8 75 4.40 128 32.6 106 48.4 55 3.92 147
2bit-BM-tele [98]104.4 11.7 42 27.0 91 3.79 142 20.2 131 26.3 148 5.07 109 12.0 66 23.2 49 4.00 117 21.2 91 36.1 88 8.16 108 45.3 86 58.0 83 10.3 58 34.0 147 61.8 144 5.92 142 54.1 151 99.9 142 5.72 149 29.8 38 47.4 44 3.74 140
FOLKI [16]105.6 13.0 122 30.9 124 4.97 151 22.2 149 24.9 126 9.00 151 17.3 123 33.0 127 7.00 141 33.4 147 38.7 111 17.0 149 44.3 62 55.8 54 10.4 67 27.6 53 49.7 72 5.48 127 36.2 63 74.2 48 4.80 141 30.4 51 44.9 26 4.08 149
SimpleFlow [49]106.1 12.0 68 24.0 40 2.94 57 18.5 96 24.4 111 4.24 76 32.7 144 39.0 140 5.69 135 18.0 28 36.2 92 7.55 42 46.9 131 60.8 129 11.1 103 31.4 137 58.1 137 5.35 113 49.4 147 99.9 142 5.16 147 40.0 141 63.0 146 3.46 44
SILK [79]107.0 13.3 127 30.7 123 3.83 145 22.0 148 25.3 130 7.16 146 34.7 150 40.0 142 7.77 148 26.6 129 36.6 95 8.60 124 45.1 82 57.9 81 10.0 21 28.4 77 50.9 103 6.03 148 34.8 24 71.8 26 4.51 137 31.4 69 48.0 52 3.74 140
ResPWCR_ROB [145]107.2 12.9 119 34.8 133 2.94 57 17.1 76 24.0 99 4.36 81 16.8 119 31.4 120 3.37 77 25.7 127 57.3 146 8.29 111 46.7 124 60.8 129 11.2 112 29.9 115 58.2 138 5.92 142 35.6 40 74.0 44 4.20 111 36.6 134 60.5 140 3.56 79
H+S_ROB [138]108.6 13.7 131 27.7 100 3.11 87 18.8 102 22.6 59 5.80 126 33.0 146 43.0 147 8.00 149 26.7 130 33.6 59 9.04 132 44.2 59 56.3 59 10.4 67 27.8 59 48.6 61 5.35 113 43.7 138 99.9 142 5.10 146 37.1 135 58.3 135 3.70 124
StereoFlow [44]108.7 22.8 153 48.3 150 3.74 141 20.5 138 26.8 151 5.07 109 11.3 31 29.3 102 3.37 77 20.1 73 37.0 97 7.62 56 59.3 151 75.2 151 10.8 89 39.3 153 71.4 152 5.45 119 35.8 48 73.9 43 4.08 44 35.7 131 55.1 125 3.70 124
StereoOF-V1MT [119]110.6 13.7 131 32.7 128 3.00 71 18.7 100 23.6 89 4.80 101 21.8 136 28.0 89 5.07 131 31.6 142 40.6 122 9.57 136 46.5 118 58.9 104 11.5 149 29.2 102 50.2 81 6.45 152 42.4 130 94.7 134 4.80 141 31.4 69 48.3 54 3.46 44
OFRF [134]110.9 14.4 140 38.4 142 3.70 138 19.9 125 25.3 130 5.48 122 13.0 97 33.0 127 3.11 71 20.7 84 38.4 107 7.79 85 47.8 139 61.4 135 10.9 95 31.9 140 56.8 133 5.29 100 41.5 125 90.5 125 4.08 44 34.4 125 54.7 119 3.42 14
Shiralkar [42]111.0 13.2 126 31.6 126 3.00 71 19.7 120 24.5 114 4.65 94 17.0 121 30.7 116 4.08 123 32.1 144 53.1 143 8.04 101 46.3 115 59.7 116 10.3 58 28.4 77 50.2 81 5.45 119 45.5 142 95.2 136 4.24 122 39.2 140 62.6 144 3.42 14
Dynamic MRF [7]113.0 12.1 82 26.8 89 2.94 57 18.0 93 23.9 96 4.16 73 18.3 128 30.7 116 5.00 129 28.9 138 39.8 116 10.5 140 45.9 105 58.6 99 11.2 112 30.9 132 56.0 132 5.80 141 43.0 134 90.3 123 4.65 139 33.7 116 51.8 99 3.70 124
Adaptive flow [45]113.6 13.4 128 25.8 77 4.51 148 21.8 147 25.4 134 7.26 147 13.7 101 27.5 84 4.69 126 24.1 121 35.2 78 8.76 128 47.3 135 61.5 136 10.2 53 33.8 146 61.9 145 5.45 119 35.9 53 73.2 39 4.20 111 34.7 128 54.7 119 3.70 124
UnFlow [129]116.3 14.9 142 40.2 145 3.11 87 18.5 96 23.4 83 5.48 122 15.3 113 31.3 119 4.36 124 22.8 112 38.0 103 8.45 119 48.3 142 63.0 143 10.9 95 32.4 141 62.0 146 5.72 136 35.4 35 71.2 19 4.32 126 45.5 150 66.0 149 3.87 145
SPSA-learn [13]119.0 12.3 99 33.7 131 3.37 122 19.2 110 23.6 89 5.45 121 30.0 140 39.7 141 7.00 141 26.9 131 41.3 124 8.41 118 46.7 124 60.1 120 10.2 53 29.4 108 50.6 93 5.20 82 53.7 150 99.9 142 8.43 153 51.4 152 72.0 152 3.51 59
SegOF [10]121.5 12.3 99 33.1 129 3.11 87 17.9 90 23.8 94 4.51 87 29.0 139 34.3 132 6.16 138 32.8 145 78.9 154 8.33 113 48.1 140 63.6 147 11.2 112 28.5 84 54.3 128 5.72 136 44.6 140 99.9 142 4.97 144 37.9 136 61.4 142 3.51 59
WRT [151]121.6 13.0 122 29.4 113 2.83 23 18.8 102 23.5 87 4.69 95 32.7 144 30.0 111 6.73 139 24.7 125 39.3 115 9.02 130 47.4 138 61.7 138 10.4 67 34.4 148 63.1 149 5.92 142 57.2 153 99.9 142 7.68 152 49.8 151 72.2 153 3.56 79
HCIC-L [99]121.8 21.0 152 41.8 146 5.07 152 20.2 131 26.1 147 5.80 126 16.3 117 42.3 144 4.00 117 31.7 143 51.0 142 8.50 120 44.7 71 55.5 51 10.4 67 35.2 150 69.8 151 5.07 55 39.9 112 91.2 128 4.16 104 40.4 144 58.0 133 3.65 119
FFV1MT [106]123.5 17.0 145 37.6 141 3.37 122 19.3 112 22.9 69 6.40 141 28.2 138 46.7 151 6.95 140 29.3 139 38.4 107 11.4 145 46.3 115 58.2 92 10.4 67 29.0 94 50.4 88 5.72 136 46.7 144 88.5 114 4.93 143 39.0 139 56.9 131 4.43 151
PGAM+LK [55]123.9 15.5 144 39.4 143 4.55 149 19.8 123 24.0 99 7.68 148 33.1 148 43.4 149 8.00 149 34.5 149 45.7 136 11.2 144 46.6 122 57.7 76 10.6 81 29.3 103 50.8 102 5.74 140 37.4 77 77.2 69 4.43 133 34.4 125 53.3 114 4.24 150
Heeger++ [104]125.1 19.8 149 44.7 147 3.11 87 18.9 106 22.8 65 6.45 142 33.0 146 35.2 133 7.16 145 29.3 139 38.4 107 11.4 145 51.5 149 65.2 148 11.3 131 28.4 77 46.9 38 6.78 153 47.9 146 84.5 106 4.69 140 40.1 142 58.8 137 3.70 124
SLK [47]126.1 13.9 136 29.9 118 3.79 142 20.0 127 22.8 65 6.22 137 32.0 143 33.7 130 7.72 147 33.4 147 46.4 137 16.1 148 48.5 145 61.7 138 10.3 58 28.4 77 47.9 53 5.72 136 43.2 135 97.9 139 4.97 144 38.7 138 59.8 138 4.04 148
WOLF_ROB [149]135.1 19.8 149 50.0 152 3.37 122 21.0 142 25.8 138 5.42 120 21.7 134 43.4 149 3.37 77 28.0 137 54.1 145 8.54 121 48.3 142 62.7 142 11.3 131 33.4 143 60.0 140 5.57 132 49.5 148 99.9 142 4.40 128 40.3 143 64.3 147 3.65 119
Pyramid LK [2]135.8 14.4 140 37.3 139 4.93 150 23.7 151 25.3 130 9.98 153 42.2 152 35.7 134 12.3 152 56.2 153 64.2 149 35.8 153 65.6 152 83.9 152 10.6 81 28.1 67 46.1 32 5.48 127 45.2 141 99.9 142 5.89 150 53.6 153 75.1 154 5.42 152
GroupFlow [9]137.0 19.9 151 49.6 151 3.42 130 19.1 108 24.1 103 5.48 122 31.4 142 40.0 142 8.19 151 36.2 150 61.9 147 12.1 147 55.6 150 71.3 150 11.4 142 36.3 151 67.0 150 5.60 134 46.7 144 98.5 140 4.20 111 43.6 148 62.8 145 3.56 79
Periodicity [78]151.2 17.6 146 55.7 153 5.45 153 26.8 153 27.0 152 9.75 152 49.4 154 51.5 154 17.7 153 51.3 152 70.3 150 27.9 152 66.6 153 86.3 153 11.7 153 38.7 152 82.5 153 6.38 151 51.8 149 99.9 142 5.48 148 44.3 149 65.5 148 5.80 153
AVG_FLOW_ROB [142]153.2 64.1 154 67.2 154 12.2 154 44.0 154 47.3 154 16.5 154 48.3 153 50.5 153 29.0 154 68.4 154 77.2 152 51.4 154 78.9 154 90.3 154 20.4 154 80.7 154 99.9 154 17.7 154 73.1 154 99.9 142 14.0 154 64.2 154 71.2 151 16.7 154
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.