Optical flow evaluation results Statistics:     Average   SD   R0.5   R1.0   R2.0   A90   A95   A99  
Error type:   endpoint   angle   interpolation   normalized interpolation  
Show images: below table   above table        
R0.5
normalized interpolation
error
avg. Mequon
(Hidden texture)
im0   GT   im1
Schefflera
(Hidden texture)
im0   GT   im1
Urban
(Synthetic)
im0   GT   im1
Teddy
(Stereo)
im0   GT   im1
Backyard
(High-speed camera)
im0   GT   im1
Basketball
(High-speed camera)
im0   GT   im1
Dumptruck
(High-speed camera)
im0   GT   im1
Evergreen
(High-speed camera)
im0   GT   im1
rank all disc untext all disc untext all disc untext all disc untext all disc untext all disc untext all disc untext all disc untext
DAIN [158]17.2 36.5 75 28.9 9 50.8 139 26.2 25 30.6 18 40.0 107 21.4 1 23.3 2 38.1 1 50.2 2 43.1 2 77.9 1 64.5 2 57.6 2 83.4 2 64.3 2 43.5 3 80.0 1 27.1 2 39.1 5 41.4 1 33.7 4 49.4 5 49.2 3
MDP-Flow2 [68]18.2 35.7 7 30.6 13 47.8 14 25.9 12 30.5 15 36.9 3 28.6 9 29.8 12 38.5 3 51.9 12 46.5 26 80.3 14 71.9 16 66.6 14 87.2 21 68.6 15 53.9 37 82.1 44 28.1 7 43.6 18 42.4 17 36.6 43 55.6 47 50.0 18
PMMST [114]19.0 35.8 9 30.8 14 47.9 24 26.5 36 31.0 26 37.3 37 28.6 9 29.9 13 38.4 2 51.7 8 46.0 14 80.2 9 72.0 21 66.7 20 87.3 29 68.5 11 53.3 17 82.0 18 28.1 7 43.7 23 42.4 17 36.5 33 55.5 42 50.0 18
PH-Flow [101]19.9 36.1 27 32.5 40 47.8 14 25.6 6 29.6 7 36.9 3 28.7 15 30.0 15 38.5 3 51.6 6 45.5 10 80.2 9 71.9 16 66.7 20 87.3 29 68.8 47 54.8 84 81.9 9 28.1 7 43.6 18 42.4 17 36.4 27 55.3 31 50.0 18
NNF-Local [87]20.3 35.7 7 31.4 17 47.6 8 25.5 4 29.6 7 36.9 3 28.6 9 29.9 13 38.5 3 52.4 40 48.0 78 80.3 14 72.0 21 66.6 14 87.4 50 68.7 29 54.4 62 82.0 18 28.1 7 43.5 16 42.4 17 36.2 13 55.0 20 50.0 18
NN-field [71]25.8 36.0 21 32.2 29 47.9 24 25.5 4 29.3 5 36.8 1 29.4 70 29.7 11 39.0 47 52.4 40 48.1 84 80.3 14 72.0 21 66.7 20 87.3 29 68.7 29 54.0 42 82.0 18 28.1 7 43.4 14 42.4 17 36.4 27 55.2 27 50.0 18
ProbFlowFields [128]30.1 35.9 14 32.4 34 48.0 37 25.8 10 30.5 15 37.2 32 28.6 9 30.3 19 38.5 3 52.1 25 46.4 22 80.7 69 72.3 82 67.1 82 87.5 101 68.6 15 53.8 28 82.1 44 28.0 6 42.8 9 42.3 4 36.1 10 54.6 15 50.1 38
MEMC-Net+ [155]30.1 39.1 139 30.2 12 53.7 148 27.9 88 31.5 36 41.9 129 23.3 2 25.1 3 39.7 104 51.1 3 43.2 5 77.9 1 67.3 7 61.1 7 83.9 5 65.3 4 47.3 6 80.0 1 27.5 3 39.5 6 41.6 2 33.6 3 50.2 7 49.0 2
IROF++ [58]30.2 36.2 34 33.0 58 47.8 14 26.1 18 30.9 23 36.9 3 29.1 37 31.0 39 38.9 30 51.6 6 45.6 11 80.4 25 72.0 21 66.8 30 87.2 21 68.6 15 53.4 18 82.2 67 28.3 25 44.6 56 42.4 17 36.5 33 55.3 31 50.4 94
Sparse-NonSparse [56]31.9 36.2 34 32.8 51 48.0 37 25.9 12 30.4 13 37.0 11 29.0 29 30.9 35 38.8 14 52.0 18 46.1 16 80.6 49 72.1 36 66.8 30 87.3 29 68.9 58 54.6 74 82.1 44 28.3 25 44.0 33 42.4 17 36.4 27 55.4 36 50.1 38
CombBMOF [113]33.3 35.9 14 31.0 15 47.8 14 25.8 10 30.5 15 36.8 1 29.2 45 30.8 31 39.5 90 52.4 40 47.4 54 80.3 14 72.1 36 66.8 30 87.4 50 68.9 58 54.5 67 82.1 44 28.5 70 44.6 56 42.3 4 36.0 9 54.6 15 50.0 18
AGIF+OF [85]35.1 36.2 34 32.8 51 47.9 24 26.1 18 30.8 20 37.1 18 29.0 29 30.7 27 38.9 30 51.8 10 46.2 19 80.1 7 72.3 82 67.2 97 87.3 29 68.9 58 55.2 102 81.9 9 28.3 25 43.6 18 42.4 17 36.6 43 56.0 66 49.9 9
nLayers [57]35.2 36.4 63 32.0 24 48.2 69 26.0 15 30.4 13 37.3 37 28.7 15 29.4 9 38.8 14 52.2 33 46.8 34 80.4 25 72.3 82 67.1 82 87.4 50 68.8 47 54.7 78 82.0 18 28.3 25 43.7 23 42.4 17 36.4 27 55.4 36 49.9 9
PyrWarp [159]36.4 25.7 1 22.7 1 35.3 1 23.6 1 26.2 1 39.9 106 24.0 3 22.5 1 47.3 150 49.5 1 40.9 1 79.5 3 61.7 1 54.0 1 82.5 1 65.7 5 43.7 4 82.5 114 33.2 156 36.9 3 63.3 157 34.1 5 45.7 2 57.1 155
2DHMM-SAS [92]37.4 36.4 63 33.9 97 47.9 24 27.1 63 32.6 55 37.0 11 28.5 8 30.4 22 38.9 30 51.8 10 45.6 11 80.4 25 72.1 36 66.9 42 87.4 50 68.8 47 54.5 67 82.0 18 28.3 25 44.2 41 42.3 4 36.7 56 56.1 75 50.0 18
NNF-EAC [103]37.9 36.3 50 32.4 34 48.0 37 26.6 40 31.7 38 37.1 18 29.3 55 30.2 16 39.0 47 52.4 40 46.9 37 81.1 114 72.0 21 66.7 20 87.4 50 68.7 29 53.7 25 82.1 44 28.2 16 43.9 27 42.4 17 36.7 56 55.9 61 50.0 18
Layers++ [37]38.8 36.3 50 32.4 34 48.2 69 25.7 7 29.2 4 37.3 37 28.9 25 30.6 24 38.9 30 52.0 18 46.4 22 80.4 25 72.2 51 67.0 58 87.5 101 68.9 58 55.2 102 82.0 18 28.3 25 44.0 33 42.4 17 36.6 43 55.5 42 50.1 38
LSM [39]40.2 36.3 50 33.7 86 48.0 37 26.1 18 31.0 26 37.0 11 29.1 37 31.8 57 38.9 30 52.2 33 46.9 37 80.6 49 72.1 36 66.9 42 87.3 29 69.0 73 54.9 88 82.1 44 28.3 25 44.1 38 42.4 17 36.5 33 55.7 51 50.0 18
ComponentFusion [96]40.7 36.0 21 32.2 29 48.0 37 26.1 18 31.1 31 36.9 3 29.1 37 32.3 67 38.8 14 52.0 18 47.0 41 80.3 14 72.2 51 67.1 82 87.3 29 68.7 29 53.9 37 82.1 44 28.5 70 46.1 114 42.4 17 36.7 56 55.8 58 50.2 60
FlowFields [110]41.2 36.0 21 32.7 48 47.9 24 26.4 33 32.0 41 37.3 37 29.0 29 32.6 74 38.7 9 52.5 46 47.9 75 80.7 69 72.3 82 67.0 58 87.5 101 68.6 15 54.4 62 82.0 18 28.2 16 44.0 33 42.4 17 36.3 16 55.2 27 50.1 38
S2F-IF [123]41.7 35.9 14 32.5 40 47.8 14 26.2 25 31.6 37 37.2 32 29.0 29 31.9 62 38.6 8 52.3 36 47.6 59 80.4 25 72.4 109 67.2 97 87.5 101 68.7 29 54.5 67 81.9 9 28.4 48 44.7 61 42.4 17 36.3 16 55.2 27 50.1 38
FlowFields+ [130]41.9 35.9 14 32.6 45 47.9 24 26.4 33 32.2 46 37.4 49 29.0 29 32.6 74 38.7 9 52.3 36 47.7 65 80.6 49 72.3 82 67.1 82 87.5 101 68.7 29 54.6 74 82.0 18 28.2 16 44.0 33 42.4 17 36.3 16 55.2 27 50.1 38
TV-L1-MCT [64]42.2 36.8 101 34.7 122 48.2 69 26.7 41 32.4 52 37.3 37 28.6 9 30.9 35 39.0 47 51.9 12 45.7 13 80.5 42 72.2 51 67.0 58 87.3 29 68.6 15 53.0 13 82.3 89 28.3 25 44.4 46 42.4 17 36.1 10 54.9 19 50.2 60
WLIF-Flow [93]42.7 36.1 27 32.5 40 47.8 14 26.3 31 31.2 32 37.1 18 29.1 37 30.7 27 39.1 56 52.0 18 46.4 22 80.6 49 72.1 36 66.8 30 87.4 50 69.0 73 54.9 88 82.3 89 28.3 25 43.9 27 42.5 73 36.8 63 55.9 61 50.1 38
MPRN [157]44.8 36.2 34 28.7 8 49.4 119 29.4 119 32.0 41 43.3 136 32.4 148 37.1 145 42.2 139 52.1 25 45.2 9 80.0 4 70.8 8 65.1 8 86.7 11 67.5 7 49.4 7 82.2 67 27.9 4 42.3 8 42.3 4 34.5 7 51.6 8 49.9 9
COFM [59]45.4 36.1 27 32.0 24 48.1 53 26.1 18 30.8 20 37.1 18 28.8 18 30.3 19 38.8 14 51.7 8 46.0 14 80.0 4 72.2 51 67.2 97 87.2 21 68.9 58 56.1 133 81.7 4 28.1 7 42.8 9 43.1 137 37.1 97 56.9 112 50.7 127
LME [70]45.8 35.8 9 31.0 15 47.8 14 26.9 52 32.2 46 38.4 93 29.2 45 32.6 74 38.8 14 51.9 12 46.7 31 80.4 25 72.6 138 67.4 126 87.7 147 68.8 47 54.9 88 82.0 18 28.1 7 43.5 16 42.4 17 36.3 16 55.3 31 50.0 18
FMOF [94]46.4 36.5 75 33.7 86 48.2 69 25.9 12 30.3 11 37.1 18 29.3 55 30.7 27 39.0 47 52.5 46 47.5 55 80.2 9 72.2 51 67.0 58 87.5 101 69.0 73 55.1 98 82.0 18 28.1 7 43.4 14 42.4 17 36.8 63 56.0 66 50.1 38
RNLOD-Flow [121]46.8 36.3 50 33.5 75 48.0 37 26.8 46 32.6 55 37.1 18 29.2 45 31.8 57 38.8 14 52.1 25 46.9 37 80.2 9 72.2 51 67.0 58 87.3 29 69.0 73 55.2 102 82.1 44 28.3 25 44.2 41 42.4 17 37.1 97 56.9 112 49.8 5
OFLAF [77]46.8 35.8 9 31.5 18 47.8 14 25.7 7 29.8 9 37.0 11 29.0 29 31.2 42 38.7 9 52.0 18 46.8 34 80.1 7 72.4 109 67.3 113 87.4 50 68.9 58 55.3 107 82.0 18 28.6 89 45.4 98 42.4 17 37.1 97 57.1 122 50.1 38
DeepFlow2 [108]47.0 36.2 34 32.4 34 48.2 69 27.1 63 32.9 61 37.8 78 29.2 45 32.9 82 39.0 47 52.5 46 47.5 55 80.5 42 72.2 51 66.9 42 87.5 101 68.5 11 52.9 12 82.1 44 28.3 25 44.4 46 42.4 17 36.4 27 55.4 36 50.2 60
CyclicGen [153]47.0 39.1 139 29.0 10 53.8 150 30.3 133 29.4 6 54.5 158 29.3 55 30.9 35 45.9 149 53.7 119 45.1 7 82.2 138 65.9 5 58.2 4 85.2 6 63.7 1 37.0 1 81.7 4 25.9 1 33.2 1 42.0 3 27.0 1 35.7 1 48.7 1
FF++_ROB [145]48.0 36.0 21 32.6 45 47.9 24 26.8 46 32.6 55 37.6 66 29.3 55 32.4 70 38.9 30 52.5 46 48.3 98 80.5 42 72.4 109 67.1 82 87.4 50 68.8 47 54.3 57 82.2 67 28.2 16 44.0 33 42.4 17 36.3 16 55.1 22 50.1 38
DeepFlow [86]48.4 36.1 27 31.8 22 48.1 53 27.3 70 32.9 61 38.4 93 29.3 55 33.3 96 39.1 56 52.6 61 47.0 41 80.7 69 72.2 51 66.8 30 87.5 101 68.7 29 52.8 11 82.5 114 28.1 7 43.6 18 42.3 4 36.2 13 55.0 20 50.2 60
EAI-Flow [151]48.6 36.3 50 32.5 40 48.2 69 27.2 67 33.2 69 38.4 93 29.4 70 33.1 90 39.0 47 52.2 33 47.3 47 80.3 14 72.2 51 67.1 82 87.3 29 68.7 29 53.5 22 82.1 44 28.4 48 44.8 70 42.4 17 36.2 13 54.5 13 50.2 60
Ramp [62]49.1 36.5 75 34.0 102 48.2 69 26.0 15 30.8 20 37.1 18 28.9 25 30.8 31 38.8 14 51.9 12 46.1 16 80.4 25 72.2 51 67.0 58 87.4 50 69.1 86 55.4 113 82.2 67 28.4 48 44.7 61 42.4 17 36.8 63 56.2 83 50.2 60
MDP-Flow [26]49.6 35.8 9 31.5 18 48.0 37 26.2 25 31.4 35 37.4 49 29.0 29 31.1 40 38.9 30 52.7 71 47.8 71 80.7 69 72.2 51 66.9 42 87.5 101 68.9 58 55.2 102 82.1 44 28.5 70 45.3 97 42.5 73 36.3 16 55.4 36 50.0 18
IROF-TV [53]50.8 36.3 50 33.6 82 48.2 69 26.2 25 31.0 26 37.0 11 29.3 55 33.6 102 39.1 56 51.9 12 46.5 26 80.8 83 72.3 82 67.0 58 87.6 137 68.5 11 53.9 37 81.9 9 28.3 25 44.9 74 42.3 4 36.6 43 55.6 47 50.4 94
CFRF [156]50.8 26.3 2 23.8 2 35.3 1 27.5 76 29.9 10 42.8 132 27.8 6 27.2 6 48.6 152 52.3 36 43.1 2 82.5 141 66.3 6 59.4 6 85.2 6 67.0 6 45.0 5 83.5 148 30.6 153 37.0 4 53.6 154 34.4 6 47.2 4 55.9 154
Classic+NL [31]51.0 36.5 75 34.0 102 48.2 69 26.2 25 30.9 23 37.1 18 28.8 18 30.6 24 38.8 14 52.1 25 46.5 26 80.6 49 72.2 51 67.0 58 87.4 50 69.2 97 55.3 107 82.2 67 28.4 48 44.6 56 42.4 17 36.8 63 56.2 83 50.2 60
PGM-C [120]51.3 36.2 34 33.3 67 48.1 53 26.5 36 32.2 46 37.5 58 29.2 45 32.9 82 38.8 14 52.5 46 48.3 98 80.7 69 72.3 82 67.0 58 87.5 101 68.6 15 54.0 42 82.0 18 28.3 25 44.6 56 42.4 17 36.5 33 55.5 42 50.4 94
DF-Auto [115]52.5 36.8 101 31.9 23 48.9 110 28.5 103 33.7 83 40.8 116 28.8 18 30.3 19 38.7 9 52.5 46 47.3 47 80.4 25 72.1 36 66.7 20 87.4 50 68.6 15 53.8 28 82.0 18 28.4 48 44.7 61 42.5 73 36.8 63 56.3 88 50.2 60
FC-2Layers-FF [74]53.8 36.4 63 33.8 92 48.1 53 25.7 7 29.1 3 37.4 49 28.9 25 30.9 35 38.8 14 52.1 25 46.8 34 80.6 49 72.3 82 67.2 97 87.4 50 69.1 86 55.5 116 82.1 44 28.4 48 44.7 61 42.5 73 36.9 78 56.3 88 50.0 18
HAST [109]55.3 36.1 27 31.7 20 48.1 53 26.1 18 31.0 26 37.0 11 29.3 55 31.7 52 39.2 70 51.9 12 46.5 26 80.3 14 72.3 82 67.3 113 87.2 21 69.3 109 56.4 141 82.0 18 28.4 48 45.0 82 42.5 73 37.3 113 57.3 126 50.0 18
SuperFlow [81]55.8 36.5 75 32.2 29 48.8 107 28.3 97 33.4 76 40.9 117 29.5 80 32.6 74 39.4 82 52.5 46 46.7 31 80.9 92 72.2 51 66.9 42 87.5 101 68.5 11 53.4 18 82.0 18 28.3 25 44.7 61 42.4 17 36.3 16 55.4 36 50.1 38
CPM-Flow [116]56.7 36.2 34 33.5 75 48.1 53 26.5 36 32.2 46 37.5 58 29.3 55 32.6 74 38.9 30 52.7 71 48.7 113 80.7 69 72.3 82 67.0 58 87.5 101 68.7 29 53.8 28 82.2 67 28.4 48 44.7 61 42.4 17 36.5 33 55.5 42 50.3 80
InterpCNN [160]57.1 42.4 153 29.5 11 59.4 158 29.8 127 31.2 32 48.2 152 29.6 85 28.0 7 50.5 157 54.0 129 46.1 16 80.3 14 65.1 4 58.2 4 83.4 2 64.8 3 43.4 2 81.0 3 27.9 4 36.4 2 44.1 151 33.4 2 46.6 3 53.3 150
Second-order prior [8]57.2 36.2 34 32.1 27 48.1 53 27.9 88 34.1 91 37.4 49 29.9 103 34.6 120 39.7 104 52.4 40 47.2 45 80.6 49 71.9 16 66.6 14 87.5 101 68.7 29 54.0 42 82.1 44 28.5 70 45.2 94 42.4 17 36.5 33 55.7 51 50.2 60
Classic+CPF [83]57.5 36.4 63 33.6 82 47.9 24 26.3 31 31.3 34 37.0 11 28.8 18 31.1 40 38.9 30 52.0 18 46.5 26 80.0 4 72.5 125 67.4 126 87.4 50 69.2 97 56.1 133 82.0 18 28.6 89 45.2 94 42.4 17 37.2 106 57.3 126 50.0 18
Aniso. Huber-L1 [22]58.3 36.7 95 33.5 75 48.6 100 28.5 103 34.3 95 38.2 89 29.3 55 31.8 57 38.9 30 52.5 46 47.5 55 80.6 49 72.0 21 66.7 20 87.4 50 68.6 15 54.3 57 81.9 9 28.5 70 45.0 82 42.4 17 36.8 63 56.0 66 50.3 80
EpicFlow [102]59.5 36.2 34 33.3 67 48.1 53 26.9 52 33.1 67 37.5 58 29.4 70 33.0 87 39.0 47 52.6 61 48.5 104 80.8 83 72.3 82 67.0 58 87.5 101 68.6 15 54.1 47 82.0 18 28.4 48 44.8 70 42.4 17 36.6 43 55.7 51 50.4 94
RFlow [90]59.5 36.2 34 33.0 58 48.2 69 27.6 79 33.7 83 37.1 18 29.3 55 32.5 73 39.2 70 52.6 61 47.8 71 80.6 49 72.0 21 66.8 30 87.3 29 68.6 15 53.8 28 81.9 9 28.5 70 45.5 103 42.6 105 37.2 106 56.9 112 50.3 80
Brox et al. [5]60.8 36.3 50 32.4 34 48.2 69 27.8 86 34.1 91 38.0 86 29.8 97 33.9 108 39.6 100 52.5 46 47.0 41 80.4 25 72.2 51 66.9 42 87.5 101 68.7 29 53.8 28 82.1 44 28.4 48 44.9 74 42.5 73 36.5 33 55.5 42 50.2 60
DMF_ROB [139]60.8 36.2 34 32.6 45 48.1 53 27.4 73 33.5 79 37.7 72 30.2 116 34.4 116 39.6 100 52.7 71 47.6 59 80.6 49 72.1 36 66.7 20 87.6 137 68.3 8 53.2 14 82.0 18 28.6 89 44.1 38 43.0 133 36.3 16 55.1 22 50.2 60
S2D-Matching [84]61.1 36.6 85 34.2 110 48.2 69 26.9 52 32.5 54 37.2 32 28.8 18 30.7 27 38.9 30 52.1 25 46.4 22 80.9 92 72.3 82 67.1 82 87.5 101 69.1 86 55.3 107 82.2 67 28.5 70 44.7 61 42.4 17 36.7 56 55.9 61 50.2 60
FESL [72]61.4 36.6 85 33.9 97 48.0 37 26.4 33 31.7 38 37.3 37 29.1 37 31.3 43 38.9 30 52.6 61 47.6 59 80.3 14 72.4 109 67.3 113 87.4 50 69.3 109 55.9 127 82.1 44 28.4 48 44.9 74 42.3 4 37.0 86 56.6 101 50.1 38
PWC-Net_ROB [147]61.4 36.4 63 34.4 115 48.0 37 27.3 70 33.9 89 37.7 72 29.7 91 34.8 124 39.1 56 52.5 46 48.9 120 80.4 25 72.4 109 67.3 113 87.4 50 69.0 73 54.1 47 82.2 67 28.2 16 43.9 27 42.4 17 36.3 16 55.1 22 49.9 9
p-harmonic [29]62.7 35.9 14 32.1 27 47.9 24 28.2 93 34.3 95 37.8 78 29.4 70 34.2 112 39.4 82 53.0 95 47.7 65 80.7 69 72.2 51 67.0 58 87.3 29 68.8 47 54.1 47 82.3 89 28.5 70 45.5 103 42.4 17 36.6 43 56.0 66 50.2 60
LiteFlowNet [142]63.4 36.4 63 34.2 110 48.0 37 27.1 63 33.4 76 37.5 58 29.6 85 35.1 132 39.1 56 53.3 107 50.7 144 80.6 49 72.1 36 66.9 42 87.3 29 69.1 86 55.1 98 82.0 18 28.6 89 45.5 103 42.3 4 36.1 10 54.8 18 49.9 9
ComplOF-FED-GPU [35]63.5 36.3 50 33.4 72 48.0 37 26.8 46 33.0 64 37.3 37 30.4 120 34.0 109 39.6 100 52.5 46 48.1 84 80.9 92 72.1 36 66.8 30 87.4 50 68.7 29 54.3 57 82.1 44 28.5 70 45.1 89 42.5 73 36.8 63 56.0 66 50.2 60
SepConv-v1 [127]63.6 27.1 4 27.5 6 36.4 4 24.9 3 31.0 26 40.3 109 27.6 5 28.4 8 48.9 153 54.0 129 47.3 47 83.0 148 72.0 21 66.7 20 87.1 13 69.1 86 52.6 10 83.6 149 32.2 154 43.9 27 55.7 155 37.0 86 53.8 11 55.4 153
ProFlow_ROB [146]63.9 36.2 34 32.8 51 48.2 69 26.9 52 33.3 73 37.7 72 29.1 37 32.2 65 38.8 14 52.6 61 48.7 113 80.7 69 72.4 109 67.2 97 87.4 50 68.7 29 53.8 28 82.2 67 28.5 70 45.2 94 42.4 17 37.0 86 56.5 96 50.3 80
TC-Flow [46]64.0 36.2 34 33.2 64 48.2 69 26.9 52 33.5 79 37.5 58 29.5 80 33.6 102 38.9 30 52.1 25 47.1 44 80.6 49 72.3 82 67.2 97 87.5 101 69.0 73 54.8 84 82.3 89 28.4 48 44.4 46 42.5 73 36.6 43 56.1 75 50.1 38
CtxSyn [136]64.2 26.8 3 25.3 3 36.2 3 23.8 2 27.2 2 39.5 102 26.3 4 26.1 5 48.0 151 51.4 5 44.1 6 82.1 136 71.6 12 66.0 12 87.1 13 70.2 144 54.2 54 84.5 156 39.3 159 45.6 107 77.2 159 38.0 136 52.3 10 59.7 157
EPPM w/o HM [88]65.3 35.8 9 32.3 33 47.6 8 26.7 41 33.0 64 36.9 3 30.0 107 35.5 136 39.4 82 52.6 61 48.9 120 80.4 25 72.2 51 67.1 82 87.4 50 69.3 109 55.9 127 82.3 89 28.4 48 44.9 74 42.5 73 36.8 63 56.1 75 50.1 38
DPOF [18]65.6 36.7 95 34.5 117 48.6 100 26.1 18 30.6 18 37.6 66 29.8 97 31.4 44 39.3 77 52.8 82 48.6 107 80.8 83 72.0 21 66.8 30 87.3 29 69.1 86 55.3 107 81.9 9 28.5 70 44.5 53 42.5 73 36.9 78 56.5 96 50.0 18
JOF [140]65.8 36.9 106 34.2 110 48.7 103 26.0 15 30.3 11 37.3 37 28.8 18 30.2 16 38.9 30 52.3 36 46.7 31 81.0 104 72.4 109 67.2 97 87.5 101 69.2 97 55.5 116 82.2 67 28.2 16 43.8 26 42.5 73 36.9 78 56.3 88 50.4 94
OFH [38]67.2 36.4 63 33.8 92 48.2 69 27.4 73 33.3 73 37.4 49 29.7 91 35.0 128 39.0 47 52.5 46 48.3 98 80.9 92 72.2 51 67.0 58 87.4 50 68.7 29 54.2 54 82.1 44 28.6 89 45.4 98 42.5 73 36.6 43 56.0 66 50.1 38
Efficient-NL [60]67.5 36.5 75 33.6 82 48.0 37 26.7 41 32.0 41 37.1 18 29.9 103 31.4 44 39.3 77 52.7 71 47.7 65 80.4 25 72.2 51 67.0 58 87.3 29 69.5 120 57.0 146 81.9 9 28.6 89 45.9 111 42.4 17 37.9 132 58.1 142 50.1 38
PMF [73]69.6 35.9 14 32.0 24 47.7 11 26.9 52 33.5 79 36.9 3 29.6 85 34.5 118 39.1 56 52.5 46 47.8 71 80.4 25 72.5 125 67.5 133 87.4 50 69.2 97 55.0 96 82.4 103 28.5 70 45.0 82 42.5 73 37.3 113 57.3 126 50.0 18
OAR-Flow [125]70.3 36.5 75 33.0 58 48.4 89 27.0 59 33.0 64 37.8 78 29.2 45 33.3 96 38.9 30 52.1 25 47.6 59 80.5 42 72.4 109 67.3 113 87.6 137 68.9 58 54.5 67 82.2 67 28.5 70 44.7 61 42.4 17 36.9 78 56.5 96 50.4 94
Local-TV-L1 [65]70.3 37.5 118 33.0 58 49.7 126 29.3 118 34.5 104 40.3 109 29.2 45 31.6 50 39.1 56 53.3 107 47.3 47 83.1 151 72.1 36 66.9 42 87.4 50 69.3 109 53.4 18 83.2 141 28.2 16 43.9 27 42.4 17 36.4 27 55.1 22 50.4 94
Sparse Occlusion [54]70.4 36.5 75 33.7 86 48.2 69 27.6 79 34.1 91 37.3 37 29.3 55 31.8 57 38.8 14 52.8 82 48.1 84 80.5 42 72.3 82 67.1 82 87.4 50 69.2 97 56.1 133 82.0 18 28.5 70 45.4 98 42.3 4 37.2 106 57.0 118 50.2 60
TC/T-Flow [76]70.5 36.6 85 33.7 86 47.9 24 26.8 46 32.9 61 37.1 18 29.1 37 31.9 62 38.8 14 52.7 71 48.5 104 80.4 25 72.5 125 67.4 126 87.5 101 69.1 86 55.2 102 82.1 44 28.6 89 45.4 98 42.5 73 37.0 86 56.9 112 50.0 18
LFNet_ROB [149]70.8 36.6 85 33.5 75 48.4 89 28.4 100 35.0 116 38.4 93 29.9 103 34.8 124 39.5 90 52.4 40 47.8 71 80.6 49 72.1 36 66.9 42 87.4 50 68.9 58 54.8 84 82.1 44 28.3 25 44.4 46 42.6 105 36.6 43 55.4 36 50.4 94
TF+OM [100]71.2 36.3 50 33.0 58 48.5 93 26.9 52 32.2 46 39.2 101 28.6 9 32.4 70 38.9 30 52.8 82 48.2 95 80.7 69 72.3 82 67.1 82 87.4 50 69.0 73 54.5 67 82.3 89 28.4 48 45.1 89 42.5 73 37.0 86 56.4 93 50.6 121
SRR-TVOF-NL [91]71.6 36.6 85 33.5 75 48.2 69 27.7 82 34.3 95 37.9 82 29.5 80 33.2 92 39.1 56 53.1 99 48.1 84 80.2 9 72.2 51 67.1 82 87.3 29 68.9 58 55.7 122 81.8 7 28.5 70 44.9 74 42.4 17 37.5 122 58.0 140 50.1 38
OFRI [161]72.0 38.3 133 27.4 5 53.7 148 31.4 137 31.8 40 56.5 160 27.8 6 25.4 4 54.3 160 51.1 3 43.1 2 81.2 119 64.8 3 58.1 3 83.8 4 68.4 9 51.8 9 83.4 146 41.6 160 42.9 11 81.5 161 38.1 138 50.1 6 63.8 160
ALD-Flow [66]72.9 36.7 95 33.9 97 48.6 100 27.0 59 33.2 69 37.9 82 29.3 55 33.4 99 38.9 30 52.5 46 48.0 78 80.9 92 72.4 109 67.2 97 87.6 137 68.9 58 54.4 62 82.2 67 28.2 16 43.6 18 42.4 17 37.0 86 56.6 101 50.3 80
CLG-TV [48]73.5 36.6 85 33.4 72 48.5 93 28.2 93 34.4 101 38.2 89 29.7 91 33.6 102 39.4 82 52.8 82 48.0 78 80.9 92 72.2 51 66.9 42 87.5 101 68.7 29 54.0 42 82.1 44 28.4 48 45.1 89 42.4 17 37.0 86 56.5 96 50.2 60
SimpleFlow [49]73.8 36.5 75 34.2 110 48.2 69 27.2 67 32.8 59 37.3 37 30.1 114 31.7 52 39.4 82 52.0 18 46.3 20 80.7 69 72.3 82 67.2 97 87.4 50 69.0 73 55.4 113 82.0 18 28.7 102 47.1 126 42.6 105 37.0 86 56.8 108 50.1 38
SIOF [67]73.9 36.7 95 34.1 106 48.2 69 29.1 114 35.4 125 39.7 104 29.4 70 32.9 82 39.1 56 52.7 71 47.7 65 80.9 92 71.9 16 66.6 14 87.4 50 69.1 86 54.3 57 82.4 103 28.3 25 44.6 56 42.4 17 37.3 113 56.8 108 50.3 80
Complementary OF [21]74.4 36.1 27 33.3 67 47.8 14 26.7 41 33.2 69 37.3 37 30.4 120 32.9 82 39.5 90 52.8 82 48.7 113 81.1 114 72.3 82 67.2 97 87.3 29 68.8 47 54.7 78 82.2 67 28.7 102 45.6 107 42.5 73 36.8 63 56.7 104 50.3 80
ContinualFlow_ROB [152]74.7 37.6 120 36.8 141 49.1 113 28.6 106 35.6 126 40.5 111 30.4 120 36.3 139 39.5 90 52.8 82 49.3 129 80.6 49 72.3 82 67.3 113 87.4 50 68.4 9 54.0 42 81.9 9 28.3 25 44.4 46 42.3 4 36.3 16 55.9 61 49.9 9
AggregFlow [97]74.9 37.1 111 34.8 125 48.5 93 27.3 70 33.2 69 38.1 87 28.7 15 30.2 16 38.5 3 52.9 91 48.6 107 80.3 14 72.4 109 67.2 97 87.6 137 69.3 109 54.5 67 82.6 123 28.3 25 44.2 41 42.5 73 36.7 56 56.0 66 50.4 94
LDOF [28]76.2 37.1 111 33.7 86 48.8 107 29.5 120 35.3 123 40.6 115 30.0 107 34.3 113 39.7 104 52.8 82 47.9 75 80.9 92 72.2 51 66.9 42 87.4 50 68.8 47 53.6 24 82.3 89 28.3 25 44.5 53 42.4 17 36.6 43 55.8 58 50.4 94
MLDP_OF [89]76.2 36.2 34 32.9 56 48.0 37 27.0 59 32.7 58 37.2 32 29.1 37 31.8 57 38.8 14 52.6 61 47.3 47 80.8 83 72.3 82 67.1 82 87.5 101 70.5 151 56.6 142 83.6 149 28.6 89 44.8 70 42.8 124 36.9 78 56.1 75 50.5 110
SuperSlomo [132]76.2 32.1 6 27.9 7 42.8 6 29.9 129 32.2 46 48.3 153 31.1 137 30.8 31 49.0 154 53.4 112 45.1 7 82.8 145 70.9 9 65.3 9 86.6 9 69.5 120 51.7 8 84.0 153 32.2 154 42.2 7 55.7 155 37.2 106 52.2 9 57.3 156
F-TV-L1 [15]76.3 37.4 116 34.6 119 49.2 115 28.8 111 34.9 114 38.3 91 29.7 91 34.1 111 39.5 90 52.7 71 47.6 59 81.0 104 71.7 13 66.5 13 87.4 50 68.8 47 53.5 22 82.4 103 28.3 25 44.3 44 42.4 17 37.1 97 56.3 88 50.6 121
Aniso-Texture [82]77.2 36.1 27 32.4 34 48.2 69 28.0 90 34.4 101 37.6 66 30.0 107 32.8 81 39.3 77 52.7 71 48.1 84 80.7 69 72.4 109 67.3 113 87.3 29 69.2 97 56.2 139 82.3 89 28.4 48 44.5 53 42.4 17 37.2 106 57.0 118 50.2 60
IAOF [50]77.8 38.0 131 34.2 110 49.8 127 31.7 140 37.9 141 41.1 120 28.9 25 32.6 74 39.4 82 53.7 119 48.1 84 80.8 83 72.0 21 66.7 20 87.5 101 68.9 58 54.1 47 82.2 67 28.3 25 45.1 89 42.3 4 36.8 63 56.1 75 50.2 60
Classic++ [32]77.8 36.4 63 33.5 75 48.4 89 27.4 73 33.7 83 37.6 66 29.6 85 33.6 102 39.2 70 52.7 71 47.3 47 80.9 92 72.2 51 67.0 58 87.5 101 69.1 86 54.5 67 82.5 114 28.5 70 44.9 74 42.6 105 36.8 63 56.2 83 50.3 80
Fusion [6]80.1 36.0 21 32.7 48 47.8 14 26.8 46 32.1 44 37.5 58 29.5 80 31.5 49 39.5 90 53.5 113 48.6 107 80.7 69 72.6 138 68.0 148 87.1 13 69.3 109 57.6 153 81.8 7 28.7 102 47.1 126 42.5 73 38.2 141 59.9 155 50.0 18
CostFilter [40]80.1 35.9 14 32.7 48 47.6 8 26.8 46 33.5 79 37.1 18 29.7 91 35.6 138 39.2 70 52.9 91 49.4 131 80.3 14 72.6 138 67.6 136 87.4 50 69.6 126 54.8 84 83.1 140 28.6 89 45.6 107 42.6 105 37.0 86 56.7 104 49.9 9
SVFilterOh [111]80.5 36.3 50 32.2 29 48.1 53 26.2 25 30.9 23 37.4 49 29.2 45 30.6 24 39.3 77 52.6 61 47.5 55 81.0 104 72.6 138 67.6 136 87.6 137 69.3 109 55.9 127 82.3 89 28.5 70 43.7 23 43.3 142 37.3 113 57.1 122 51.0 131
Shiralkar [42]80.6 36.5 75 34.6 119 48.1 53 28.3 97 34.3 95 37.2 32 29.8 97 36.9 144 40.0 116 53.9 125 49.0 123 80.5 42 71.8 14 66.6 14 87.2 21 69.2 97 55.1 98 82.4 103 29.2 127 48.0 137 42.5 73 36.6 43 55.7 51 50.1 38
FlowNetS+ft+v [112]80.7 36.8 101 33.0 58 48.7 103 29.5 120 35.6 126 40.5 111 29.8 97 34.3 113 39.5 90 52.8 82 48.2 95 80.8 83 72.2 51 67.0 58 87.4 50 68.7 29 53.9 37 82.1 44 28.6 89 45.9 111 42.5 73 36.7 56 56.0 66 50.4 94
Occlusion-TV-L1 [63]80.8 36.6 85 33.8 92 48.5 93 28.4 100 34.8 111 37.7 72 29.5 80 33.0 87 39.5 90 53.0 95 48.1 84 81.1 114 72.1 36 66.8 30 87.5 101 68.9 58 53.4 18 82.4 103 29.0 121 44.7 61 42.6 105 36.8 63 55.6 47 50.4 94
TriFlow [95]81.0 37.0 108 35.3 132 48.8 107 28.7 109 34.5 104 41.0 118 29.2 45 33.4 99 38.8 14 53.0 95 48.8 119 80.4 25 72.3 82 67.3 113 87.4 50 69.2 97 55.5 116 82.1 44 28.5 70 44.8 70 42.4 17 36.9 78 56.4 93 50.1 38
3DFlow [135]82.2 36.4 63 33.8 92 47.9 24 26.5 36 32.1 44 37.1 18 29.8 97 31.7 52 39.1 56 52.5 46 47.7 65 80.6 49 72.5 125 67.3 113 88.0 151 70.0 133 57.8 154 82.2 67 29.1 125 47.4 130 42.5 73 37.4 119 57.6 133 49.9 9
CNN-flow-warp+ref [117]82.5 36.3 50 31.7 20 48.7 103 28.5 103 34.7 107 39.5 102 30.4 120 35.0 128 39.8 109 54.0 129 48.1 84 81.2 119 72.3 82 67.0 58 87.4 50 68.6 15 53.2 14 82.4 103 28.8 110 47.1 126 42.5 73 36.6 43 55.7 51 50.3 80
CRTflow [80]82.5 36.7 95 33.8 92 48.5 93 27.7 82 33.8 86 37.4 49 30.7 129 35.3 133 40.9 134 52.9 91 48.1 84 81.8 133 72.2 51 66.9 42 87.4 50 68.9 58 54.1 47 82.3 89 28.4 48 44.9 74 42.5 73 36.8 63 56.1 75 50.5 110
TOF-M [154]82.6 31.5 5 27.3 4 42.1 5 28.6 106 32.8 59 46.1 148 31.1 137 31.9 62 50.6 158 53.6 116 46.3 20 82.8 145 71.2 11 65.6 11 87.1 13 70.3 149 53.7 25 85.0 157 38.0 157 43.3 13 76.6 158 39.1 152 54.5 13 61.2 159
FlowNet2 [122]84.9 39.4 142 38.2 147 50.4 134 29.2 117 34.8 111 41.9 129 30.0 107 34.6 120 39.4 82 53.3 107 51.0 147 80.6 49 72.5 125 67.4 126 87.4 50 68.8 47 54.3 57 82.0 18 28.4 48 45.0 82 42.3 4 36.5 33 55.7 51 49.8 5
TCOF [69]85.5 36.6 85 33.9 97 48.1 53 29.1 114 35.7 128 38.3 91 29.0 29 31.4 44 38.7 9 52.8 82 48.7 113 80.6 49 72.2 51 67.1 82 87.4 50 69.3 109 56.0 130 82.1 44 28.7 102 46.2 116 42.5 73 38.2 141 58.7 151 50.5 110
ResPWCR_ROB [144]85.6 36.4 63 34.0 102 48.1 53 28.2 93 34.9 114 38.9 98 30.7 129 35.0 128 39.7 104 53.7 119 50.6 141 81.5 127 71.8 14 66.6 14 87.0 12 70.7 153 56.0 130 84.3 154 28.4 48 44.9 74 42.5 73 36.5 33 55.9 61 50.0 18
Adaptive [20]85.8 36.8 101 34.4 115 48.5 93 28.8 111 35.2 122 37.7 72 29.4 70 33.2 92 39.2 70 52.6 61 47.6 59 80.6 49 72.3 82 67.0 58 87.5 101 69.1 86 54.7 78 82.3 89 28.7 102 46.0 113 42.4 17 37.3 113 56.9 112 50.4 94
Modified CLG [34]85.8 36.9 106 32.8 51 49.4 119 30.9 136 36.3 135 42.8 132 30.0 107 34.8 124 39.9 112 53.0 95 47.9 75 80.7 69 72.2 51 66.9 42 87.5 101 68.7 29 53.8 28 82.2 67 28.4 48 45.1 89 42.5 73 36.9 78 56.2 83 50.5 110
AugFNG_ROB [143]86.4 37.7 122 35.6 134 49.6 123 29.5 120 36.0 130 41.4 125 30.4 120 37.7 148 40.1 121 53.5 113 50.5 140 81.0 104 72.5 125 67.5 133 87.4 50 68.7 29 54.4 62 82.0 18 28.5 70 44.4 46 42.4 17 35.3 8 54.0 12 49.4 4
EPMNet [133]86.5 38.9 137 38.5 150 49.9 130 29.0 113 34.2 94 41.2 121 30.0 107 34.6 120 39.4 82 53.9 125 52.7 153 80.6 49 72.5 125 67.4 126 87.4 50 69.0 73 55.5 116 82.0 18 28.4 48 45.0 82 42.3 4 36.3 16 55.3 31 49.8 5
IIOF-NLDP [131]86.8 36.3 50 33.3 67 47.7 11 27.6 79 34.3 95 37.4 49 29.8 97 31.7 52 39.2 70 53.3 107 48.7 113 81.2 119 72.2 51 67.0 58 87.5 101 69.8 130 56.8 143 82.2 67 29.4 132 50.8 154 42.8 124 37.1 97 56.8 108 49.9 9
Steered-L1 [118]87.9 36.0 21 32.9 56 47.9 24 27.0 59 33.3 73 37.7 72 30.3 119 32.3 67 39.9 112 53.2 101 48.0 78 81.0 104 72.5 125 67.5 133 87.5 101 68.9 58 55.0 96 82.2 67 28.8 110 46.7 123 42.7 119 37.0 86 57.3 126 50.3 80
Nguyen [33]90.1 39.6 143 33.9 97 52.6 145 32.5 145 37.9 141 43.3 136 30.0 107 35.5 136 40.2 122 54.1 134 49.0 123 80.9 92 72.0 21 66.8 30 87.4 50 68.6 15 53.8 28 82.0 18 28.8 110 47.8 135 42.4 17 36.8 63 56.1 75 50.3 80
BriefMatch [124]90.1 36.3 50 33.3 67 48.0 37 27.2 67 33.4 76 38.5 97 30.6 127 32.6 74 40.6 127 54.0 129 48.6 107 82.8 145 72.4 109 67.3 113 87.3 29 70.2 144 55.6 121 83.9 152 28.3 25 44.3 44 42.7 119 36.6 43 55.7 51 50.5 110
StereoOF-V1MT [119]91.0 36.8 101 35.3 132 48.1 53 28.3 97 35.1 119 36.9 3 31.4 141 36.6 141 40.5 125 54.6 143 48.6 107 81.3 122 72.0 21 66.8 30 87.2 21 69.5 120 54.9 88 82.6 123 29.7 142 48.8 144 42.7 119 36.5 33 55.1 22 50.1 38
SPSA-learn [13]94.3 37.4 116 33.6 82 49.4 119 29.8 127 35.1 119 41.4 125 30.9 133 33.2 92 40.7 128 53.5 113 47.2 45 80.4 25 72.2 51 67.0 58 87.4 50 68.8 47 54.1 47 82.2 67 29.5 137 52.2 158 42.9 130 37.1 97 57.0 118 50.3 80
GraphCuts [14]94.6 38.0 131 35.1 129 49.5 122 28.4 100 33.9 89 41.3 123 31.3 140 30.8 31 40.7 128 53.7 119 48.3 98 81.0 104 72.1 36 67.1 82 87.1 13 68.6 15 54.9 88 81.7 4 28.8 110 46.3 117 42.8 124 37.7 126 58.5 147 50.4 94
Dynamic MRF [7]95.7 36.2 34 34.1 106 48.0 37 27.5 76 34.6 106 37.4 49 30.9 133 36.8 143 40.4 124 54.5 141 49.3 129 81.9 134 71.9 16 66.8 30 87.2 21 69.4 118 55.5 116 82.5 114 29.0 121 47.8 135 42.5 73 37.5 122 56.8 108 50.5 110
ROF-ND [107]95.8 37.0 108 32.8 51 48.1 53 27.7 82 34.7 107 37.6 66 29.7 91 32.3 67 39.1 56 54.2 136 51.4 148 80.4 25 72.4 109 67.3 113 87.4 50 69.5 120 56.9 144 82.0 18 29.7 142 49.0 147 43.2 139 37.8 130 57.7 137 50.2 60
HBpMotionGpu [43]95.8 38.8 136 35.9 136 50.9 140 32.1 142 38.2 143 44.4 141 29.2 45 31.7 52 39.3 77 53.9 125 49.6 133 81.5 127 72.1 36 67.0 58 87.1 13 69.5 120 54.9 88 82.4 103 28.3 25 44.4 46 42.5 73 37.3 113 56.5 96 51.1 132
2D-CLG [1]96.5 37.9 126 33.5 75 50.5 137 32.5 145 37.4 138 45.0 143 30.8 132 34.8 124 40.7 128 53.7 119 48.3 98 80.5 42 72.3 82 67.1 82 87.6 137 68.6 15 53.2 14 82.2 67 28.8 110 46.7 123 42.5 73 36.9 78 55.6 47 50.3 80
TV-L1-improved [17]96.8 36.6 85 34.1 106 48.4 89 28.6 106 35.1 119 37.8 78 30.5 125 33.2 92 40.0 116 52.7 71 48.0 78 80.9 92 72.3 82 67.2 97 87.4 50 69.1 86 54.9 88 82.3 89 28.8 110 47.3 129 42.6 105 37.2 106 56.7 104 50.6 121
Black & Anandan [4]97.0 37.9 126 34.1 106 49.6 123 30.7 134 36.0 130 41.2 121 31.0 135 34.7 123 40.3 123 53.9 125 48.6 107 80.7 69 72.3 82 67.0 58 87.4 50 69.0 73 53.8 28 82.5 114 28.8 110 46.5 118 42.4 17 37.0 86 56.1 75 50.4 94
CBF [12]97.8 36.4 63 32.5 40 48.9 110 27.5 76 33.8 86 37.9 82 29.3 55 31.6 50 39.1 56 53.2 101 48.1 84 82.6 142 72.4 109 67.2 97 87.7 147 69.2 97 55.3 107 82.3 89 28.7 102 46.1 114 42.9 130 37.9 132 57.7 137 51.7 141
Rannacher [23]100.2 36.7 95 34.5 117 48.7 103 28.7 109 35.3 123 38.1 87 30.5 125 34.0 109 39.9 112 52.7 71 48.0 78 80.8 83 72.4 109 67.2 97 87.5 101 69.0 73 54.6 74 82.3 89 28.8 110 47.0 125 42.6 105 37.1 97 56.4 93 50.6 121
Correlation Flow [75]100.6 36.2 34 33.4 72 47.7 11 27.7 82 34.3 95 37.3 37 29.4 70 31.4 44 38.8 14 53.1 99 48.5 104 81.3 122 72.8 145 67.6 136 88.6 159 70.1 137 57.1 147 82.6 123 29.4 132 48.8 144 43.0 133 37.7 126 57.9 139 50.5 110
UnFlow [129]101.6 39.2 141 37.9 144 50.6 138 32.3 143 38.9 149 41.3 123 31.6 146 38.5 151 40.8 133 53.2 101 48.7 113 80.9 92 72.0 21 66.7 20 87.4 50 69.5 120 54.6 74 82.4 103 28.2 16 43.2 12 42.4 17 39.3 155 58.3 144 51.2 133
HBM-GC [105]102.0 37.7 122 34.7 122 49.8 127 27.1 63 32.4 52 37.9 82 28.8 18 29.6 10 39.2 70 52.6 61 47.3 47 80.8 83 73.2 151 68.1 149 88.2 154 70.0 133 57.3 149 82.7 127 28.9 120 45.0 82 43.5 143 37.6 124 57.2 124 51.3 135
TriangleFlow [30]102.7 37.0 108 34.9 126 48.5 93 28.0 90 34.7 107 37.5 58 30.2 116 33.0 87 39.9 112 53.2 101 49.0 123 81.1 114 72.0 21 66.9 42 87.1 13 69.8 130 56.1 133 82.4 103 29.2 127 48.5 141 42.8 124 38.1 138 58.5 147 50.5 110
SegOF [10]103.4 37.6 120 33.2 64 50.0 131 29.1 114 34.7 107 41.0 118 31.4 141 35.3 133 40.7 128 53.6 116 50.7 144 80.6 49 72.3 82 67.2 97 87.5 101 69.0 73 55.3 107 82.2 67 29.0 121 48.6 142 42.7 119 36.7 56 55.8 58 50.4 94
TVL1_ROB [138]103.6 40.0 146 35.0 127 52.6 145 33.3 148 38.6 147 45.0 143 29.6 85 34.3 113 39.8 109 54.0 129 48.2 95 81.3 122 72.2 51 66.9 42 87.5 101 69.0 73 53.7 25 82.5 114 28.8 110 46.5 118 42.6 105 36.8 63 56.0 66 50.5 110
BlockOverlap [61]104.6 38.5 135 33.2 64 51.3 141 30.0 131 34.4 101 42.8 132 29.4 70 30.4 22 40.0 116 53.2 101 46.9 37 83.0 148 72.9 148 67.6 136 88.3 155 69.7 128 54.1 47 83.3 145 28.7 102 44.1 38 43.5 143 37.1 97 55.3 31 51.8 142
WRT [150]105.0 36.6 85 34.0 102 47.9 24 28.1 92 33.8 86 37.5 58 31.1 137 31.4 44 39.6 100 53.2 101 48.4 103 80.8 83 72.7 143 67.7 143 87.8 149 70.1 137 58.7 157 82.2 67 29.8 145 53.4 160 42.9 130 37.6 124 58.4 145 49.8 5
IAOF2 [51]106.4 37.9 126 35.9 136 49.1 113 29.6 124 36.1 133 40.0 107 29.3 55 33.4 99 40.0 116 54.1 134 50.2 139 81.0 104 72.4 109 67.4 126 87.4 50 69.2 97 54.9 88 82.4 103 28.6 89 45.5 103 42.4 17 37.9 132 57.6 133 50.6 121
Ad-TV-NDC [36]107.5 40.4 149 35.1 129 53.1 147 31.9 141 36.7 137 43.8 139 29.4 70 32.9 82 39.1 56 54.5 141 49.2 128 82.1 136 72.5 125 67.3 113 87.5 101 69.3 109 53.9 37 82.7 127 28.6 89 45.4 98 42.4 17 37.2 106 56.2 83 50.6 121
OFRF [134]112.3 38.9 137 36.1 139 50.4 134 29.5 120 35.0 116 40.5 111 29.6 85 34.4 116 39.0 47 53.3 107 48.9 120 81.1 114 72.6 138 67.7 143 87.3 29 70.1 137 57.3 149 82.5 114 29.1 125 47.6 133 42.6 105 37.4 119 58.0 140 50.0 18
LocallyOriented [52]115.5 37.5 118 35.9 136 49.2 115 29.6 124 36.2 134 39.1 100 30.1 114 33.8 106 39.5 90 53.7 119 50.0 137 81.3 122 72.3 82 67.2 97 87.5 101 70.2 144 56.2 139 82.9 136 28.8 110 45.6 107 42.5 73 37.7 126 57.6 133 50.5 110
AdaConv-v1 [126]116.3 37.2 115 36.6 140 47.5 7 34.3 152 39.1 152 51.1 156 36.1 156 39.4 153 52.9 159 58.2 155 53.1 154 83.8 153 70.9 9 65.4 10 86.6 9 69.7 128 54.7 78 84.4 155 38.6 158 46.5 118 77.4 160 38.2 141 54.6 15 60.3 158
ACK-Prior [27]116.5 36.4 63 33.7 86 48.1 53 26.7 41 33.1 67 37.1 18 30.7 129 33.3 96 39.7 104 53.6 116 50.0 137 81.0 104 73.5 155 68.6 152 88.3 155 70.8 154 59.8 158 82.7 127 29.7 142 48.7 143 43.6 146 39.5 156 62.1 158 51.3 135
WOLF_ROB [148]118.0 38.3 133 38.0 145 49.0 112 29.6 124 35.9 129 39.0 99 30.6 127 35.0 128 39.8 109 54.4 140 52.4 152 82.0 135 72.5 125 67.6 136 87.4 50 69.8 130 55.4 113 82.8 133 29.4 132 49.1 148 42.5 73 37.1 97 56.6 101 50.2 60
Horn & Schunck [3]118.0 37.9 126 35.1 129 49.6 123 31.4 137 37.7 140 41.8 128 31.7 147 37.4 147 41.5 136 55.8 147 50.6 141 81.3 122 72.2 51 67.0 58 87.4 50 69.2 97 54.2 54 82.7 127 29.5 137 48.9 146 42.6 105 37.8 130 57.2 124 50.9 130
StereoFlow [44]118.1 46.3 159 45.9 160 54.3 151 38.3 159 45.4 160 45.7 146 29.3 55 33.8 106 39.1 56 52.9 91 47.7 65 81.0 104 74.4 159 70.5 160 87.6 137 72.0 158 66.3 160 82.4 103 28.4 48 45.0 82 42.4 17 38.0 136 59.1 152 50.5 110
Filter Flow [19]118.8 37.8 125 34.6 119 49.8 127 30.8 135 36.0 130 44.3 140 29.4 70 32.4 70 39.5 90 54.2 136 48.1 84 82.2 138 72.7 143 67.7 143 87.6 137 69.2 97 55.1 98 82.5 114 28.7 102 46.5 118 42.6 105 38.3 147 58.4 145 51.4 138
TI-DOFE [24]121.2 42.0 152 37.5 143 54.8 154 35.2 154 41.1 157 46.8 150 31.4 141 37.7 148 41.6 138 56.1 149 50.6 141 81.6 129 72.0 21 66.9 42 87.2 21 69.4 118 54.4 62 82.6 123 29.2 127 47.6 133 42.6 105 38.2 141 57.5 130 50.8 129
SILK [79]122.8 39.6 143 38.1 146 51.5 144 32.4 144 38.5 146 43.6 138 32.4 148 37.2 146 41.5 136 55.4 145 49.7 134 83.0 148 72.2 51 67.0 58 87.4 50 70.0 133 54.7 78 83.4 146 29.0 121 46.5 118 42.8 124 37.4 119 56.7 104 50.7 127
Bartels [41]123.9 37.1 111 35.0 127 49.3 118 28.2 93 34.8 111 40.5 111 29.9 103 33.1 90 40.5 125 54.2 136 49.7 134 83.9 154 73.0 149 67.6 136 88.7 160 71.8 157 56.1 133 85.6 159 28.6 89 43.9 27 43.6 146 38.1 138 57.0 118 53.2 149
SLK [47]132.5 41.6 151 38.7 151 54.4 152 33.0 147 38.3 145 45.5 145 33.3 150 38.6 152 42.8 141 57.8 153 51.8 150 83.5 152 72.1 36 67.3 113 86.5 8 70.1 137 55.8 125 82.7 127 30.0 146 51.4 156 43.0 133 38.2 141 57.5 130 51.5 139
NL-TV-NCC [25]132.8 37.1 111 35.7 135 48.0 37 27.8 86 35.0 116 37.6 66 31.0 135 35.4 135 40.0 116 56.0 148 54.2 157 82.6 142 73.8 157 68.6 152 89.1 161 70.6 152 58.4 156 82.5 114 30.4 150 50.0 151 44.0 150 39.8 158 60.2 156 52.4 147
GroupFlow [9]135.0 40.3 148 40.1 153 51.3 141 31.5 139 38.9 149 42.6 131 33.5 152 39.5 154 43.8 143 54.7 144 52.3 151 81.0 104 73.2 151 68.6 152 87.6 137 70.4 150 57.3 149 83.0 138 29.3 130 48.1 138 42.5 73 37.9 132 58.2 143 50.1 38
FFV1MT [106]135.3 39.6 143 40.8 155 50.3 133 34.8 153 38.8 148 46.6 149 36.5 157 45.8 158 44.6 146 56.2 150 49.0 123 81.7 130 72.5 125 67.4 126 87.4 50 70.2 144 54.7 78 83.2 141 30.1 148 49.3 150 42.8 124 38.6 148 57.6 133 51.3 135
Learning Flow [11]135.4 37.7 122 37.0 142 49.2 115 29.9 129 37.5 139 39.7 104 31.5 145 36.3 139 40.7 128 55.4 145 51.6 149 82.6 142 72.8 145 67.8 147 87.8 149 69.6 126 55.7 122 82.8 133 29.3 130 48.4 139 42.7 119 39.2 153 59.8 154 51.2 133
Heeger++ [104]136.1 40.6 150 42.5 157 50.4 134 33.5 149 38.2 143 43.0 135 37.6 158 48.1 159 44.9 147 56.2 150 49.0 123 81.7 130 73.4 154 68.9 157 87.5 101 70.1 137 56.0 130 82.8 133 30.3 149 49.1 148 42.6 105 37.7 126 56.9 112 50.3 80
2bit-BM-tele [98]136.7 37.9 126 34.7 122 50.1 132 30.1 132 36.4 136 41.7 127 30.2 116 32.2 65 41.3 135 54.3 139 49.4 131 84.1 156 73.3 153 68.1 149 88.3 155 72.2 159 57.2 148 85.5 158 30.4 150 52.7 159 44.4 152 38.2 141 56.3 88 54.1 152
H+S_ROB [137]137.7 40.0 146 38.4 149 51.4 143 34.1 151 38.9 149 44.9 142 35.1 155 42.5 157 44.2 145 59.3 157 49.7 134 82.3 140 72.5 125 67.7 143 87.1 13 70.0 133 55.7 122 82.7 127 30.4 150 50.9 155 43.2 139 39.6 157 57.5 130 51.9 143
FOLKI [16]141.6 44.6 157 40.4 154 58.4 157 35.7 155 42.3 158 47.3 151 33.3 150 40.7 155 44.9 147 59.4 159 53.6 155 86.5 159 72.5 125 67.6 136 87.3 29 70.1 137 55.8 125 83.2 141 29.4 132 48.4 139 43.1 137 38.7 149 58.5 147 52.0 144
Pyramid LK [2]146.4 46.1 158 38.9 152 61.0 159 36.7 158 40.4 155 50.9 155 39.9 159 36.6 141 49.4 155 64.1 160 61.2 160 87.7 160 73.1 150 68.6 152 87.4 50 70.1 137 56.1 133 83.0 138 29.6 140 50.7 153 43.2 139 39.2 153 61.2 157 51.5 139
Adaptive flow [45]146.9 43.8 155 38.2 147 56.5 155 35.8 156 40.5 156 50.2 154 31.4 141 34.5 118 42.5 140 56.5 152 50.9 146 83.9 154 73.5 155 68.7 156 88.1 153 70.2 144 57.4 152 82.9 136 29.4 132 47.5 131 43.6 146 39.0 151 59.1 152 52.0 144
PGAM+LK [55]147.9 42.5 154 41.4 156 54.6 153 33.7 150 40.1 154 45.8 147 33.9 153 41.1 156 43.3 142 59.3 157 55.2 158 85.5 158 72.8 145 68.1 149 87.5 101 70.8 154 56.9 144 83.6 149 29.6 140 50.0 151 43.0 133 38.7 149 58.6 150 52.1 146
HCIC-L [99]151.8 49.1 160 42.6 158 63.0 160 35.8 156 39.4 153 52.5 157 34.6 154 37.7 148 43.9 144 58.0 154 53.9 156 81.7 130 74.0 158 69.3 158 88.5 158 71.7 156 60.5 159 83.2 141 29.5 137 47.5 131 43.9 149 40.5 159 62.9 159 52.8 148
Periodicity [78]156.6 44.4 156 43.3 159 56.9 156 42.8 160 43.4 159 56.2 159 40.9 160 49.1 160 49.5 156 58.9 156 58.6 159 84.9 157 74.4 159 70.2 159 88.0 151 73.1 160 57.9 155 86.2 160 30.0 146 51.5 157 43.5 143 41.8 160 63.4 160 53.7 151
AVG_FLOW_ROB [141]158.2 76.9 161 76.7 161 78.2 161 71.8 161 68.8 161 76.4 161 64.0 161 60.2 161 65.8 161 82.9 161 80.9 161 91.0 161 80.9 161 79.6 161 87.5 101 83.9 161 84.0 161 86.6 161 53.7 161 65.3 161 47.7 153 62.3 161 71.1 161 70.4 161
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] CtxSyn 0.07 2 color S. Niklaus and F. Liu. (Interpolation results only.) Context-aware synthesis for video frame interpolation. CVPR 2018.
[137] 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.
[138] 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.
[139] 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.
[140] 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.
[141] AVG_FLOW_ROB N/A 2 N/A Average flow field of ROB 2018 training set.
[142] 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.
[143] AugFNG_ROB 0.10 all color Anonymous. FusionNet and AugmentedFlowNet: Selective proxy ground truth for training on unlabeled images. ECCV 2018 submission 2834.
[144] ResPWCR_ROB 0.2 2 color Anonymous. Learning optical flow with residual connections. ROB 2018 submission.
[145] 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.
[146] ProFlow_ROB 76 3 color Anonymous. ProFlow: Learning to predict optical flow. BMVC 2018 submission 277.
[147] 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.
[148] WOLF_ROB 0.02 2 color Anonymous. Reversed deep neural network for optical flow. ROB 2018 submission.
[149] LFNet_ROB 0.068 2 color Anonymous. Learning a flow network. ROB 2018 submission.
[150] 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.
[151] 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.
[152] ContinualFlow_ROB 0.5 all color M Neoral, J. Sochman, and J. Matas. Continual occlusions and optical flow estimation. ACCV 2018.
[153] CyclicGen 0.088 2 color Anonymous. (Interpolation results only.) Deep video frame interpolation using cyclic frame generation. AAAI 2019 submission 323.
[154] TOF-M 0.393 2 color T. Xue, B. Chen, J. Wu, D. Wei, and W. Freeman. Video enhancement with task-oriented flow. arXiv 1711.09078, 2017.
[155] MEMC-Net+ 0.16 2 color W. Bao, W.-S. Lai, X. Zhang, Z. Gao, and M.-H. Yang. (Interpolation results only.) MEMC-Net: Motion estimation and motion compensation driven neural network for video interpolation and enhancement. Submitted to TPAMI 2018.
[156] CFRF 0.128 2 color Anonymous. (Interpolation results only.) Coarse-to-fine refinement framework for video frame interpolation. CVPR 2019 submission 1992.
[157] MPRN 0.32 4 color Anonymous. (Interpolation results only.) Multi-frame pyramid refinement network for video frame interpolation. CVPR 2019 submission 1361.
[158] DAIN 0.13 2 color Anonymous. (Interpolation results only.) DAIN: Depth-aware video frame interpolation. CVPR 2019 submission 1769.
[159] PyrWarp 0.14 2 color Anonymous. (Interpolation results only.) Feature pyramid warping for video frame interpolation. CVPR 2019 submission 868.
[160] InterpCNN 0.65 2 color Anonymous. (Interpolation results only.) Video frame interpolation with a stack of synthesis networks and intermediate optical flows. CVPR 2019 submission 6533.
[161] OFRI 0.31 2 color Anonymous. (Interpolation results only.) Efficient video frame interpolation via optical flow refinement. CVPR 2019 submission 6743.
* 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.