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
MDP-Flow2 [68]13.8 35.7 4 30.6 5 47.8 11 25.9 11 30.5 13 36.9 3 28.6 5 29.8 7 38.5 2 51.9 8 46.5 18 80.3 10 71.9 9 66.6 7 87.2 14 68.6 9 53.9 30 82.1 42 28.1 4 43.6 11 42.4 15 36.6 38 55.6 40 50.0 16
PMMST [114]14.7 35.8 6 30.8 6 47.9 21 26.5 34 31.0 23 37.3 37 28.6 5 29.9 8 38.4 1 51.7 4 46.0 8 80.2 5 72.0 14 66.7 13 87.3 22 68.5 5 53.3 11 82.0 16 28.1 4 43.7 16 42.4 15 36.5 28 55.5 35 50.0 16
PH-Flow [101]15.7 36.1 24 32.5 32 47.8 11 25.6 5 29.6 6 36.9 3 28.7 11 30.0 10 38.5 2 51.6 2 45.5 4 80.2 5 71.9 9 66.7 13 87.3 22 68.8 41 54.8 77 81.9 7 28.1 4 43.6 11 42.4 15 36.4 22 55.3 24 50.0 16
NNF-Local [87]15.9 35.7 4 31.4 9 47.6 5 25.5 3 29.6 6 36.9 3 28.6 5 29.9 8 38.5 2 52.4 34 48.0 70 80.3 10 72.0 14 66.6 7 87.4 43 68.7 23 54.4 55 82.0 16 28.1 4 43.5 9 42.4 15 36.2 8 55.0 13 50.0 16
NN-field [71]21.4 36.0 18 32.2 21 47.9 21 25.5 3 29.3 4 36.8 1 29.4 66 29.7 6 39.0 46 52.4 34 48.1 77 80.3 10 72.0 14 66.7 13 87.3 22 68.7 23 54.0 35 82.0 16 28.1 4 43.4 7 42.4 15 36.4 22 55.2 20 50.0 16
ProbFlowFields [128]25.9 35.9 11 32.4 26 48.0 34 25.8 9 30.5 13 37.2 32 28.6 5 30.3 14 38.5 2 52.1 21 46.4 14 80.7 65 72.3 75 67.1 75 87.5 94 68.6 9 53.8 21 82.1 42 28.0 3 42.8 4 42.3 3 36.1 5 54.6 8 50.1 36
IROF++ [58]26.0 36.2 31 33.0 51 47.8 11 26.1 17 30.9 20 36.9 3 29.1 33 31.0 34 38.9 29 51.6 2 45.6 5 80.4 20 72.0 14 66.8 23 87.2 14 68.6 9 53.4 12 82.2 65 28.3 22 44.6 49 42.4 15 36.5 28 55.3 24 50.4 92
Sparse-NonSparse [56]27.7 36.2 31 32.8 44 48.0 34 25.9 11 30.4 11 37.0 11 29.0 25 30.9 30 38.8 13 52.0 14 46.1 10 80.6 45 72.1 29 66.8 23 87.3 22 68.9 52 54.6 67 82.1 42 28.3 22 44.0 26 42.4 15 36.4 22 55.4 29 50.1 36
CombBMOF [113]28.9 35.9 11 31.0 7 47.8 11 25.8 9 30.5 13 36.8 1 29.2 41 30.8 26 39.5 89 52.4 34 47.4 46 80.3 10 72.1 29 66.8 23 87.4 43 68.9 52 54.5 60 82.1 42 28.5 67 44.6 49 42.3 3 36.0 4 54.6 8 50.0 16
nLayers [57]30.7 36.4 59 32.0 16 48.2 66 26.0 14 30.4 11 37.3 37 28.7 11 29.4 3 38.8 13 52.2 28 46.8 26 80.4 20 72.3 75 67.1 75 87.4 43 68.8 41 54.7 71 82.0 16 28.3 22 43.7 16 42.4 15 36.4 22 55.4 29 49.9 8
AGIF+OF [85]30.8 36.2 31 32.8 44 47.9 21 26.1 17 30.8 17 37.1 18 29.0 25 30.7 22 38.9 29 51.8 6 46.2 12 80.1 3 72.3 75 67.2 90 87.3 22 68.9 52 55.2 95 81.9 7 28.3 22 43.6 11 42.4 15 36.6 38 56.0 59 49.9 8
2DHMM-SAS [92]32.9 36.4 59 33.9 90 47.9 21 27.1 61 32.6 48 37.0 11 28.5 4 30.4 17 38.9 29 51.8 6 45.6 5 80.4 20 72.1 29 66.9 35 87.4 43 68.8 41 54.5 60 82.0 16 28.3 22 44.2 34 42.3 3 36.7 51 56.1 68 50.0 16
NNF-EAC [103]33.3 36.3 46 32.4 26 48.0 34 26.6 38 31.7 33 37.1 18 29.3 51 30.2 11 39.0 46 52.4 34 46.9 29 81.1 110 72.0 14 66.7 13 87.4 43 68.7 23 53.7 19 82.1 42 28.2 13 43.9 20 42.4 15 36.7 51 55.9 54 50.0 16
Layers++ [37]34.4 36.3 46 32.4 26 48.2 66 25.7 6 29.2 3 37.3 37 28.9 21 30.6 19 38.9 29 52.0 14 46.4 14 80.4 20 72.2 44 67.0 51 87.5 94 68.9 52 55.2 95 82.0 16 28.3 22 44.0 26 42.4 15 36.6 38 55.5 35 50.1 36
LSM [39]35.8 36.3 46 33.7 79 48.0 34 26.1 17 31.0 23 37.0 11 29.1 33 31.8 52 38.9 29 52.2 28 46.9 29 80.6 45 72.1 29 66.9 35 87.3 22 69.0 67 54.9 81 82.1 42 28.3 22 44.1 31 42.4 15 36.5 28 55.7 44 50.0 16
ComponentFusion [96]36.2 36.0 18 32.2 21 48.0 34 26.1 17 31.1 28 36.9 3 29.1 33 32.3 61 38.8 13 52.0 14 47.0 33 80.3 10 72.2 44 67.1 75 87.3 22 68.7 23 53.9 30 82.1 42 28.5 67 46.1 107 42.4 15 36.7 51 55.8 51 50.2 58
FlowFields [110]36.5 36.0 18 32.7 40 47.9 21 26.4 31 32.0 35 37.3 37 29.0 25 32.6 68 38.7 8 52.5 40 47.9 67 80.7 65 72.3 75 67.0 51 87.5 94 68.6 9 54.4 55 82.0 16 28.2 13 44.0 26 42.4 15 36.3 11 55.2 20 50.1 36
S2F-IF [123]37.1 35.9 11 32.5 32 47.8 11 26.2 24 31.6 32 37.2 32 29.0 25 31.9 57 38.6 7 52.3 31 47.6 51 80.4 20 72.4 102 67.2 90 87.5 94 68.7 23 54.5 60 81.9 7 28.4 45 44.7 54 42.4 15 36.3 11 55.2 20 50.1 36
FlowFields+ [130]37.2 35.9 11 32.6 37 47.9 21 26.4 31 32.2 39 37.4 49 29.0 25 32.6 68 38.7 8 52.3 31 47.7 57 80.6 45 72.3 75 67.1 75 87.5 94 68.7 23 54.6 67 82.0 16 28.2 13 44.0 26 42.4 15 36.3 11 55.2 20 50.1 36
TV-L1-MCT [64]37.6 36.8 96 34.7 115 48.2 66 26.7 39 32.4 45 37.3 37 28.6 5 30.9 30 39.0 46 51.9 8 45.7 7 80.5 38 72.2 44 67.0 51 87.3 22 68.6 9 53.0 7 82.3 86 28.3 22 44.4 39 42.4 15 36.1 5 54.9 12 50.2 58
WLIF-Flow [93]38.2 36.1 24 32.5 32 47.8 11 26.3 29 31.2 29 37.1 18 29.1 33 30.7 22 39.1 55 52.0 14 46.4 14 80.6 45 72.1 29 66.8 23 87.4 43 69.0 67 54.9 81 82.3 86 28.3 22 43.9 20 42.5 71 36.8 58 55.9 54 50.1 36
LME [70]41.1 35.8 6 31.0 7 47.8 11 26.9 50 32.2 39 38.4 93 29.2 41 32.6 68 38.8 13 51.9 8 46.7 23 80.4 20 72.6 131 67.4 119 87.7 140 68.8 41 54.9 81 82.0 16 28.1 4 43.5 9 42.4 15 36.3 11 55.3 24 50.0 16
COFM [59]41.2 36.1 24 32.0 16 48.1 50 26.1 17 30.8 17 37.1 18 28.8 14 30.3 14 38.8 13 51.7 4 46.0 8 80.0 1 72.2 44 67.2 90 87.2 14 68.9 52 56.1 126 81.7 2 28.1 4 42.8 4 43.1 135 37.1 92 56.9 105 50.7 125
FMOF [94]42.0 36.5 71 33.7 79 48.2 66 25.9 11 30.3 9 37.1 18 29.3 51 30.7 22 39.0 46 52.5 40 47.5 47 80.2 5 72.2 44 67.0 51 87.5 94 69.0 67 55.1 91 82.0 16 28.1 4 43.4 7 42.4 15 36.8 58 56.0 59 50.1 36
RNLOD-Flow [121]42.1 36.3 46 33.5 68 48.0 34 26.8 44 32.6 48 37.1 18 29.2 41 31.8 52 38.8 13 52.1 21 46.9 29 80.2 5 72.2 44 67.0 51 87.3 22 69.0 67 55.2 95 82.1 42 28.3 22 44.2 34 42.4 15 37.1 92 56.9 105 49.8 3
DeepFlow2 [108]42.2 36.2 31 32.4 26 48.2 66 27.1 61 32.9 53 37.8 78 29.2 41 32.9 76 39.0 46 52.5 40 47.5 47 80.5 38 72.2 44 66.9 35 87.5 94 68.5 5 52.9 6 82.1 42 28.3 22 44.4 39 42.4 15 36.4 22 55.4 29 50.2 58
OFLAF [77]42.5 35.8 6 31.5 10 47.8 11 25.7 6 29.8 8 37.0 11 29.0 25 31.2 37 38.7 8 52.0 14 46.8 26 80.1 3 72.4 102 67.3 106 87.4 43 68.9 52 55.3 100 82.0 16 28.6 86 45.4 91 42.4 15 37.1 92 57.1 115 50.1 36
FF++_ROB [146]43.3 36.0 18 32.6 37 47.9 21 26.8 44 32.6 48 37.6 66 29.3 51 32.4 64 38.9 29 52.5 40 48.3 91 80.5 38 72.4 102 67.1 75 87.4 43 68.8 41 54.3 50 82.2 65 28.2 13 44.0 26 42.4 15 36.3 11 55.1 15 50.1 36
DeepFlow [86]43.7 36.1 24 31.8 14 48.1 50 27.3 68 32.9 53 38.4 93 29.3 51 33.3 90 39.1 55 52.6 55 47.0 33 80.7 65 72.2 44 66.8 23 87.5 94 68.7 23 52.8 5 82.5 111 28.1 4 43.6 11 42.3 3 36.2 8 55.0 13 50.2 58
CyclicGen [154]43.9 39.1 134 29.0 4 53.8 144 30.3 126 29.4 5 54.5 152 29.3 51 30.9 30 45.9 147 53.7 112 45.1 2 82.2 133 65.9 1 58.2 1 85.2 1 63.7 1 37.0 1 81.7 2 25.9 1 33.2 1 42.0 1 27.0 1 35.7 1 48.7 1
EAI-Flow [152]43.9 36.3 46 32.5 32 48.2 66 27.2 65 33.2 61 38.4 93 29.4 66 33.1 84 39.0 46 52.2 28 47.3 39 80.3 10 72.2 44 67.1 75 87.3 22 68.7 23 53.5 16 82.1 42 28.4 45 44.8 63 42.4 15 36.2 8 54.5 7 50.2 58
FGIK [136]44.5 38.3 128 32.7 40 51.4 138 30.5 127 35.5 118 45.9 143 25.0 1 29.5 4 40.5 123 54.4 132 48.0 70 80.4 20 66.3 2 59.5 2 85.6 2 65.3 2 44.9 2 81.1 1 26.8 2 37.4 2 42.0 1 34.2 2 51.0 2 49.8 3
Ramp [62]44.8 36.5 71 34.0 95 48.2 66 26.0 14 30.8 17 37.1 18 28.9 21 30.8 26 38.8 13 51.9 8 46.1 10 80.4 20 72.2 44 67.0 51 87.4 43 69.1 80 55.4 106 82.2 65 28.4 45 44.7 54 42.4 15 36.8 58 56.2 76 50.2 58
MDP-Flow [26]45.1 35.8 6 31.5 10 48.0 34 26.2 24 31.4 31 37.4 49 29.0 25 31.1 35 38.9 29 52.7 65 47.8 63 80.7 65 72.2 44 66.9 35 87.5 94 68.9 52 55.2 95 82.1 42 28.5 67 45.3 90 42.5 71 36.3 11 55.4 29 50.0 16
IROF-TV [53]46.3 36.3 46 33.6 75 48.2 66 26.2 24 31.0 23 37.0 11 29.3 51 33.6 96 39.1 55 51.9 8 46.5 18 80.8 79 72.3 75 67.0 51 87.6 130 68.5 5 53.9 30 81.9 7 28.3 22 44.9 67 42.3 3 36.6 38 55.6 40 50.4 92
Classic+NL [31]46.6 36.5 71 34.0 95 48.2 66 26.2 24 30.9 20 37.1 18 28.8 14 30.6 19 38.8 13 52.1 21 46.5 18 80.6 45 72.2 44 67.0 51 87.4 43 69.2 91 55.3 100 82.2 65 28.4 45 44.6 49 42.4 15 36.8 58 56.2 76 50.2 58
PGM-C [120]46.6 36.2 31 33.3 60 48.1 50 26.5 34 32.2 39 37.5 58 29.2 41 32.9 76 38.8 13 52.5 40 48.3 91 80.7 65 72.3 75 67.0 51 87.5 94 68.6 9 54.0 35 82.0 16 28.3 22 44.6 49 42.4 15 36.5 28 55.5 35 50.4 92
DF-Auto [115]47.5 36.8 96 31.9 15 48.9 107 28.5 99 33.7 75 40.8 114 28.8 14 30.3 14 38.7 8 52.5 40 47.3 39 80.4 20 72.1 29 66.7 13 87.4 43 68.6 9 53.8 21 82.0 16 28.4 45 44.7 54 42.5 71 36.8 58 56.3 81 50.2 58
FC-2Layers-FF [74]49.4 36.4 59 33.8 85 48.1 50 25.7 6 29.1 2 37.4 49 28.9 21 30.9 30 38.8 13 52.1 21 46.8 26 80.6 45 72.3 75 67.2 90 87.4 43 69.1 80 55.5 109 82.1 42 28.4 45 44.7 54 42.5 71 36.9 73 56.3 81 50.0 16
SuperFlow [81]50.9 36.5 71 32.2 21 48.8 104 28.3 93 33.4 68 40.9 115 29.5 76 32.6 68 39.4 81 52.5 40 46.7 23 80.9 88 72.2 44 66.9 35 87.5 94 68.5 5 53.4 12 82.0 16 28.3 22 44.7 54 42.4 15 36.3 11 55.4 29 50.1 36
HAST [109]50.9 36.1 24 31.7 12 48.1 50 26.1 17 31.0 23 37.0 11 29.3 51 31.7 47 39.2 69 51.9 8 46.5 18 80.3 10 72.3 75 67.3 106 87.2 14 69.3 103 56.4 134 82.0 16 28.4 45 45.0 75 42.5 71 37.3 108 57.3 119 50.0 16
CPM-Flow [116]52.0 36.2 31 33.5 68 48.1 50 26.5 34 32.2 39 37.5 58 29.3 51 32.6 68 38.9 29 52.7 65 48.7 106 80.7 65 72.3 75 67.0 51 87.5 94 68.7 23 53.8 21 82.2 65 28.4 45 44.7 54 42.4 15 36.5 28 55.5 35 50.3 78
Second-order prior [8]52.4 36.2 31 32.1 19 48.1 50 27.9 85 34.1 83 37.4 49 29.9 98 34.6 114 39.7 103 52.4 34 47.2 37 80.6 45 71.9 9 66.6 7 87.5 94 68.7 23 54.0 35 82.1 42 28.5 67 45.2 87 42.4 15 36.5 28 55.7 44 50.2 58
Classic+CPF [83]53.0 36.4 59 33.6 75 47.9 21 26.3 29 31.3 30 37.0 11 28.8 14 31.1 35 38.9 29 52.0 14 46.5 18 80.0 1 72.5 118 67.4 119 87.4 43 69.2 91 56.1 126 82.0 16 28.6 86 45.2 87 42.4 15 37.2 101 57.3 119 50.0 16
Aniso. Huber-L1 [22]53.4 36.7 90 33.5 68 48.6 97 28.5 99 34.3 87 38.2 89 29.3 51 31.8 52 38.9 29 52.5 40 47.5 47 80.6 45 72.0 14 66.7 13 87.4 43 68.6 9 54.3 50 81.9 7 28.5 67 45.0 75 42.4 15 36.8 58 56.0 59 50.3 78
RFlow [90]54.7 36.2 31 33.0 51 48.2 66 27.6 76 33.7 75 37.1 18 29.3 51 32.5 67 39.2 69 52.6 55 47.8 63 80.6 45 72.0 14 66.8 23 87.3 22 68.6 9 53.8 21 81.9 7 28.5 67 45.5 96 42.6 103 37.2 101 56.9 105 50.3 78
EpicFlow [102]54.8 36.2 31 33.3 60 48.1 50 26.9 50 33.1 59 37.5 58 29.4 66 33.0 81 39.0 46 52.6 55 48.5 97 80.8 79 72.3 75 67.0 51 87.5 94 68.6 9 54.1 40 82.0 16 28.4 45 44.8 63 42.4 15 36.6 38 55.7 44 50.4 92
Brox et al. [5]55.8 36.3 46 32.4 26 48.2 66 27.8 83 34.1 83 38.0 86 29.8 92 33.9 102 39.6 99 52.5 40 47.0 33 80.4 20 72.2 44 66.9 35 87.5 94 68.7 23 53.8 21 82.1 42 28.4 45 44.9 67 42.5 71 36.5 28 55.5 35 50.2 58
DMF_ROB [140]56.0 36.2 31 32.6 37 48.1 50 27.4 71 33.5 71 37.7 72 30.2 111 34.4 110 39.6 99 52.7 65 47.6 51 80.6 45 72.1 29 66.7 13 87.6 130 68.3 3 53.2 8 82.0 16 28.6 86 44.1 31 43.0 131 36.3 11 55.1 15 50.2 58
S2D-Matching [84]56.4 36.6 80 34.2 103 48.2 66 26.9 50 32.5 47 37.2 32 28.8 14 30.7 22 38.9 29 52.1 21 46.4 14 80.9 88 72.3 75 67.1 75 87.5 94 69.1 80 55.3 100 82.2 65 28.5 67 44.7 54 42.4 15 36.7 51 55.9 54 50.2 58
PWC-Net_ROB [148]56.6 36.4 59 34.4 108 48.0 34 27.3 68 33.9 81 37.7 72 29.7 86 34.8 118 39.1 55 52.5 40 48.9 113 80.4 20 72.4 102 67.3 106 87.4 43 69.0 67 54.1 40 82.2 65 28.2 13 43.9 20 42.4 15 36.3 11 55.1 15 49.9 8
FESL [72]56.8 36.6 80 33.9 90 48.0 34 26.4 31 31.7 33 37.3 37 29.1 33 31.3 38 38.9 29 52.6 55 47.6 51 80.3 10 72.4 102 67.3 106 87.4 43 69.3 103 55.9 120 82.1 42 28.4 45 44.9 67 42.3 3 37.0 81 56.6 94 50.1 36
p-harmonic [29]57.8 35.9 11 32.1 19 47.9 21 28.2 89 34.3 87 37.8 78 29.4 66 34.2 106 39.4 81 53.0 89 47.7 57 80.7 65 72.2 44 67.0 51 87.3 22 68.8 41 54.1 40 82.3 86 28.5 67 45.5 96 42.4 15 36.6 38 56.0 59 50.2 58
ComplOF-FED-GPU [35]58.7 36.3 46 33.4 65 48.0 34 26.8 44 33.0 56 37.3 37 30.4 115 34.0 103 39.6 99 52.5 40 48.1 77 80.9 88 72.1 29 66.8 23 87.4 43 68.7 23 54.3 50 82.1 42 28.5 67 45.1 82 42.5 71 36.8 58 56.0 59 50.2 58
LiteFlowNet [143]58.8 36.4 59 34.2 103 48.0 34 27.1 61 33.4 68 37.5 58 29.6 81 35.1 126 39.1 55 53.3 101 50.7 137 80.6 45 72.1 29 66.9 35 87.3 22 69.1 80 55.1 91 82.0 16 28.6 86 45.5 96 42.3 3 36.1 5 54.8 11 49.9 8
SepConv-v1 [127]58.8 27.1 2 27.5 2 36.4 2 24.9 2 31.0 23 40.3 107 27.6 3 28.4 2 48.9 149 54.0 122 47.3 39 83.0 141 72.0 14 66.7 13 87.1 7 69.1 80 52.6 4 83.6 143 32.2 150 43.9 20 55.7 151 37.0 81 53.8 5 55.4 150
ProFlow_ROB [147]59.2 36.2 31 32.8 44 48.2 66 26.9 50 33.3 65 37.7 72 29.1 33 32.2 59 38.8 13 52.6 55 48.7 106 80.7 65 72.4 102 67.2 90 87.4 43 68.7 23 53.8 21 82.2 65 28.5 67 45.2 87 42.4 15 37.0 81 56.5 89 50.3 78
TC-Flow [46]59.2 36.2 31 33.2 57 48.2 66 26.9 50 33.5 71 37.5 58 29.5 76 33.6 96 38.9 29 52.1 21 47.1 36 80.6 45 72.3 75 67.2 90 87.5 94 69.0 67 54.8 77 82.3 86 28.4 45 44.4 39 42.5 71 36.6 38 56.1 68 50.1 36
CtxSyn [137]59.8 26.8 1 25.3 1 36.2 1 23.8 1 27.2 1 39.5 102 26.3 2 26.1 1 48.0 148 51.4 1 44.1 1 82.1 131 71.6 5 66.0 5 87.1 7 70.2 138 54.2 47 84.5 150 39.3 153 45.6 100 77.2 153 38.0 131 52.3 4 59.7 152
EPPM w/o HM [88]60.5 35.8 6 32.3 25 47.6 5 26.7 39 33.0 56 36.9 3 30.0 102 35.5 130 39.4 81 52.6 55 48.9 113 80.4 20 72.2 44 67.1 75 87.4 43 69.3 103 55.9 120 82.3 86 28.4 45 44.9 67 42.5 71 36.8 58 56.1 68 50.1 36
DPOF [18]61.1 36.7 90 34.5 110 48.6 97 26.1 17 30.6 16 37.6 66 29.8 92 31.4 39 39.3 76 52.8 76 48.6 100 80.8 79 72.0 14 66.8 23 87.3 22 69.1 80 55.3 100 81.9 7 28.5 67 44.5 46 42.5 71 36.9 73 56.5 89 50.0 16
JOF [141]61.3 36.9 101 34.2 103 48.7 100 26.0 14 30.3 9 37.3 37 28.8 14 30.2 11 38.9 29 52.3 31 46.7 23 81.0 100 72.4 102 67.2 90 87.5 94 69.2 91 55.5 109 82.2 65 28.2 13 43.8 19 42.5 71 36.9 73 56.3 81 50.4 92
OFH [38]62.5 36.4 59 33.8 85 48.2 66 27.4 71 33.3 65 37.4 49 29.7 86 35.0 122 39.0 46 52.5 40 48.3 91 80.9 88 72.2 44 67.0 51 87.4 43 68.7 23 54.2 47 82.1 42 28.6 86 45.4 91 42.5 71 36.6 38 56.0 59 50.1 36
Efficient-NL [60]62.8 36.5 71 33.6 75 48.0 34 26.7 39 32.0 35 37.1 18 29.9 98 31.4 39 39.3 76 52.7 65 47.7 57 80.4 20 72.2 44 67.0 51 87.3 22 69.5 114 57.0 139 81.9 7 28.6 86 45.9 104 42.4 15 37.9 127 58.1 135 50.1 36
PMF [73]64.7 35.9 11 32.0 16 47.7 8 26.9 50 33.5 71 36.9 3 29.6 81 34.5 112 39.1 55 52.5 40 47.8 63 80.4 20 72.5 118 67.5 126 87.4 43 69.2 91 55.0 89 82.4 100 28.5 67 45.0 75 42.5 71 37.3 108 57.3 119 50.0 16
Local-TV-L1 [65]65.1 37.5 113 33.0 51 49.7 122 29.3 113 34.5 96 40.3 107 29.2 41 31.6 45 39.1 55 53.3 101 47.3 39 83.1 144 72.1 29 66.9 35 87.4 43 69.3 103 53.4 12 83.2 137 28.2 13 43.9 20 42.4 15 36.4 22 55.1 15 50.4 92
OAR-Flow [125]65.6 36.5 71 33.0 51 48.4 86 27.0 57 33.0 56 37.8 78 29.2 41 33.3 90 38.9 29 52.1 21 47.6 51 80.5 38 72.4 102 67.3 106 87.6 130 68.9 52 54.5 60 82.2 65 28.5 67 44.7 54 42.4 15 36.9 73 56.5 89 50.4 92
Sparse Occlusion [54]65.7 36.5 71 33.7 79 48.2 66 27.6 76 34.1 83 37.3 37 29.3 51 31.8 52 38.8 13 52.8 76 48.1 77 80.5 38 72.3 75 67.1 75 87.4 43 69.2 91 56.1 126 82.0 16 28.5 67 45.4 91 42.3 3 37.2 101 57.0 111 50.2 58
TC/T-Flow [76]65.8 36.6 80 33.7 79 47.9 21 26.8 44 32.9 53 37.1 18 29.1 33 31.9 57 38.8 13 52.7 65 48.5 97 80.4 20 72.5 118 67.4 119 87.5 94 69.1 80 55.2 95 82.1 42 28.6 86 45.4 91 42.5 71 37.0 81 56.9 105 50.0 16
LFNet_ROB [150]65.8 36.6 80 33.5 68 48.4 86 28.4 96 35.0 108 38.4 93 29.9 98 34.8 118 39.5 89 52.4 34 47.8 63 80.6 45 72.1 29 66.9 35 87.4 43 68.9 52 54.8 77 82.1 42 28.3 22 44.4 39 42.6 103 36.6 38 55.4 29 50.4 92
TF+OM [100]66.4 36.3 46 33.0 51 48.5 90 26.9 50 32.2 39 39.2 101 28.6 5 32.4 64 38.9 29 52.8 76 48.2 88 80.7 65 72.3 75 67.1 75 87.4 43 69.0 67 54.5 60 82.3 86 28.4 45 45.1 82 42.5 71 37.0 81 56.4 86 50.6 119
SRR-TVOF-NL [91]66.8 36.6 80 33.5 68 48.2 66 27.7 79 34.3 87 37.9 82 29.5 76 33.2 86 39.1 55 53.1 93 48.1 77 80.2 5 72.2 44 67.1 75 87.3 22 68.9 52 55.7 115 81.8 5 28.5 67 44.9 67 42.4 15 37.5 117 58.0 133 50.1 36
ALD-Flow [66]68.1 36.7 90 33.9 90 48.6 97 27.0 57 33.2 61 37.9 82 29.3 51 33.4 93 38.9 29 52.5 40 48.0 70 80.9 88 72.4 102 67.2 90 87.6 130 68.9 52 54.4 55 82.2 65 28.2 13 43.6 11 42.4 15 37.0 81 56.6 94 50.3 78
CLG-TV [48]68.6 36.6 80 33.4 65 48.5 90 28.2 89 34.4 93 38.2 89 29.7 86 33.6 96 39.4 81 52.8 76 48.0 70 80.9 88 72.2 44 66.9 35 87.5 94 68.7 23 54.0 35 82.1 42 28.4 45 45.1 82 42.4 15 37.0 81 56.5 89 50.2 58
SIOF [67]68.9 36.7 90 34.1 99 48.2 66 29.1 109 35.4 117 39.7 104 29.4 66 32.9 76 39.1 55 52.7 65 47.7 57 80.9 88 71.9 9 66.6 7 87.4 43 69.1 80 54.3 50 82.4 100 28.3 22 44.6 49 42.4 15 37.3 108 56.8 101 50.3 78
SimpleFlow [49]69.1 36.5 71 34.2 103 48.2 66 27.2 65 32.8 52 37.3 37 30.1 109 31.7 47 39.4 81 52.0 14 46.3 13 80.7 65 72.3 75 67.2 90 87.4 43 69.0 67 55.4 106 82.0 16 28.7 99 47.1 119 42.6 103 37.0 81 56.8 101 50.1 36
Complementary OF [21]69.6 36.1 24 33.3 60 47.8 11 26.7 39 33.2 61 37.3 37 30.4 115 32.9 76 39.5 89 52.8 76 48.7 106 81.1 110 72.3 75 67.2 90 87.3 22 68.8 41 54.7 71 82.2 65 28.7 99 45.6 100 42.5 71 36.8 58 56.7 97 50.3 78
ContinualFlow_ROB [153]69.8 37.6 115 36.8 134 49.1 110 28.6 102 35.6 119 40.5 109 30.4 115 36.3 133 39.5 89 52.8 76 49.3 122 80.6 45 72.3 75 67.3 106 87.4 43 68.4 4 54.0 35 81.9 7 28.3 22 44.4 39 42.3 3 36.3 11 55.9 54 49.9 8
AggregFlow [97]70.1 37.1 106 34.8 118 48.5 90 27.3 68 33.2 61 38.1 87 28.7 11 30.2 11 38.5 2 52.9 85 48.6 100 80.3 10 72.4 102 67.2 90 87.6 130 69.3 103 54.5 60 82.6 119 28.3 22 44.2 34 42.5 71 36.7 51 56.0 59 50.4 92
LDOF [28]71.0 37.1 106 33.7 79 48.8 104 29.5 114 35.3 115 40.6 113 30.0 102 34.3 107 39.7 103 52.8 76 47.9 67 80.9 88 72.2 44 66.9 35 87.4 43 68.8 41 53.6 18 82.3 86 28.3 22 44.5 46 42.4 15 36.6 38 55.8 51 50.4 92
SuperSlomo [132]71.0 32.1 3 27.9 3 42.8 3 29.9 122 32.2 39 48.3 147 31.1 132 30.8 26 49.0 150 53.4 106 45.1 2 82.8 139 70.9 3 65.3 3 86.6 4 69.5 114 51.7 3 84.0 147 32.2 150 42.2 3 55.7 151 37.2 101 52.2 3 57.3 151
MLDP_OF [89]71.3 36.2 31 32.9 49 48.0 34 27.0 57 32.7 51 37.2 32 29.1 33 31.8 52 38.8 13 52.6 55 47.3 39 80.8 79 72.3 75 67.1 75 87.5 94 70.5 144 56.6 135 83.6 143 28.6 86 44.8 63 42.8 122 36.9 73 56.1 68 50.5 108
F-TV-L1 [15]71.3 37.4 111 34.6 112 49.2 112 28.8 106 34.9 106 38.3 91 29.7 86 34.1 105 39.5 89 52.7 65 47.6 51 81.0 100 71.7 6 66.5 6 87.4 43 68.8 41 53.5 16 82.4 100 28.3 22 44.3 37 42.4 15 37.1 92 56.3 81 50.6 119
Aniso-Texture [82]72.3 36.1 24 32.4 26 48.2 66 28.0 86 34.4 93 37.6 66 30.0 102 32.8 75 39.3 76 52.7 65 48.1 77 80.7 65 72.4 102 67.3 106 87.3 22 69.2 91 56.2 132 82.3 86 28.4 45 44.5 46 42.4 15 37.2 101 57.0 111 50.2 58
IAOF [50]72.7 38.0 126 34.2 103 49.8 123 31.7 133 37.9 134 41.1 118 28.9 21 32.6 68 39.4 81 53.7 112 48.1 77 80.8 79 72.0 14 66.7 13 87.5 94 68.9 52 54.1 40 82.2 65 28.3 22 45.1 82 42.3 3 36.8 58 56.1 68 50.2 58
Classic++ [32]73.0 36.4 59 33.5 68 48.4 86 27.4 71 33.7 75 37.6 66 29.6 81 33.6 96 39.2 69 52.7 65 47.3 39 80.9 88 72.2 44 67.0 51 87.5 94 69.1 80 54.5 60 82.5 111 28.5 67 44.9 67 42.6 103 36.8 58 56.2 76 50.3 78
CostFilter [40]75.2 35.9 11 32.7 40 47.6 5 26.8 44 33.5 71 37.1 18 29.7 86 35.6 132 39.2 69 52.9 85 49.4 124 80.3 10 72.6 131 67.6 129 87.4 43 69.6 120 54.8 77 83.1 136 28.6 86 45.6 100 42.6 103 37.0 81 56.7 97 49.9 8
Fusion [6]75.4 36.0 18 32.7 40 47.8 11 26.8 44 32.1 37 37.5 58 29.5 76 31.5 44 39.5 89 53.5 107 48.6 100 80.7 65 72.6 131 68.0 141 87.1 7 69.3 103 57.6 146 81.8 5 28.7 99 47.1 119 42.5 71 38.2 135 59.9 148 50.0 16
Shiralkar [42]75.6 36.5 71 34.6 112 48.1 50 28.3 93 34.3 87 37.2 32 29.8 92 36.9 138 40.0 114 53.9 118 49.0 116 80.5 38 71.8 7 66.6 7 87.2 14 69.2 91 55.1 91 82.4 100 29.2 124 48.0 130 42.5 71 36.6 38 55.7 44 50.1 36
FlowNetS+ft+v [112]75.7 36.8 96 33.0 51 48.7 100 29.5 114 35.6 119 40.5 109 29.8 92 34.3 107 39.5 89 52.8 76 48.2 88 80.8 79 72.2 44 67.0 51 87.4 43 68.7 23 53.9 30 82.1 42 28.6 86 45.9 104 42.5 71 36.7 51 56.0 59 50.4 92
TriFlow [95]75.9 37.0 103 35.3 125 48.8 104 28.7 104 34.5 96 41.0 116 29.2 41 33.4 93 38.8 13 53.0 89 48.8 112 80.4 20 72.3 75 67.3 106 87.4 43 69.2 91 55.5 109 82.1 42 28.5 67 44.8 63 42.4 15 36.9 73 56.4 86 50.1 36
SVFilterOh [111]75.9 36.3 46 32.2 21 48.1 50 26.2 24 30.9 20 37.4 49 29.2 41 30.6 19 39.3 76 52.6 55 47.5 47 81.0 100 72.6 131 67.6 129 87.6 130 69.3 103 55.9 120 82.3 86 28.5 67 43.7 16 43.3 140 37.3 108 57.1 115 51.0 129
Occlusion-TV-L1 [63]76.0 36.6 80 33.8 85 48.5 90 28.4 96 34.8 103 37.7 72 29.5 76 33.0 81 39.5 89 53.0 89 48.1 77 81.1 110 72.1 29 66.8 23 87.5 94 68.9 52 53.4 12 82.4 100 29.0 118 44.7 54 42.6 103 36.8 58 55.6 40 50.4 92
CNN-flow-warp+ref [117]77.5 36.3 46 31.7 12 48.7 100 28.5 99 34.7 99 39.5 102 30.4 115 35.0 122 39.8 107 54.0 122 48.1 77 81.2 115 72.3 75 67.0 51 87.4 43 68.6 9 53.2 8 82.4 100 28.8 107 47.1 119 42.5 71 36.6 38 55.7 44 50.3 78
3DFlow [135]77.5 36.4 59 33.8 85 47.9 21 26.5 34 32.1 37 37.1 18 29.8 92 31.7 47 39.1 55 52.5 40 47.7 57 80.6 45 72.5 118 67.3 106 88.0 144 70.0 127 57.8 147 82.2 65 29.1 122 47.4 123 42.5 71 37.4 114 57.6 126 49.9 8
CRTflow [80]77.6 36.7 90 33.8 85 48.5 90 27.7 79 33.8 78 37.4 49 30.7 124 35.3 127 40.9 133 52.9 85 48.1 77 81.8 128 72.2 44 66.9 35 87.4 43 68.9 52 54.1 40 82.3 86 28.4 45 44.9 67 42.5 71 36.8 58 56.1 68 50.5 108
FlowNet2 [122]79.8 39.4 136 38.2 140 50.4 130 29.2 112 34.8 103 41.9 127 30.0 102 34.6 114 39.4 81 53.3 101 51.0 140 80.6 45 72.5 118 67.4 119 87.4 43 68.8 41 54.3 50 82.0 16 28.4 45 45.0 75 42.3 3 36.5 28 55.7 44 49.8 3
ResPWCR_ROB [145]80.5 36.4 59 34.0 95 48.1 50 28.2 89 34.9 106 38.9 98 30.7 124 35.0 122 39.7 103 53.7 112 50.6 134 81.5 122 71.8 7 66.6 7 87.0 6 70.7 146 56.0 123 84.3 148 28.4 45 44.9 67 42.5 71 36.5 28 55.9 54 50.0 16
Modified CLG [34]80.7 36.9 101 32.8 44 49.4 116 30.9 130 36.3 128 42.8 129 30.0 102 34.8 118 39.9 110 53.0 89 47.9 67 80.7 65 72.2 44 66.9 35 87.5 94 68.7 23 53.8 21 82.2 65 28.4 45 45.1 82 42.5 71 36.9 73 56.2 76 50.5 108
TCOF [69]80.7 36.6 80 33.9 90 48.1 50 29.1 109 35.7 121 38.3 91 29.0 25 31.4 39 38.7 8 52.8 76 48.7 106 80.6 45 72.2 44 67.1 75 87.4 43 69.3 103 56.0 123 82.1 42 28.7 99 46.2 109 42.5 71 38.2 135 58.7 144 50.5 108
Adaptive [20]80.8 36.8 96 34.4 108 48.5 90 28.8 106 35.2 114 37.7 72 29.4 66 33.2 86 39.2 69 52.6 55 47.6 51 80.6 45 72.3 75 67.0 51 87.5 94 69.1 80 54.7 71 82.3 86 28.7 99 46.0 106 42.4 15 37.3 108 56.9 105 50.4 92
AugFNG_ROB [144]81.3 37.7 117 35.6 127 49.6 119 29.5 114 36.0 123 41.4 123 30.4 115 37.7 141 40.1 119 53.5 107 50.5 133 81.0 100 72.5 118 67.5 126 87.4 43 68.7 23 54.4 55 82.0 16 28.5 67 44.4 39 42.4 15 35.3 3 54.0 6 49.4 2
EPMNet [133]81.4 38.9 132 38.5 143 49.9 126 29.0 108 34.2 86 41.2 119 30.0 102 34.6 114 39.4 81 53.9 118 52.7 146 80.6 45 72.5 118 67.4 119 87.4 43 69.0 67 55.5 109 82.0 16 28.4 45 45.0 75 42.3 3 36.3 11 55.3 24 49.8 3
IIOF-NLDP [131]82.0 36.3 46 33.3 60 47.7 8 27.6 76 34.3 87 37.4 49 29.8 92 31.7 47 39.2 69 53.3 101 48.7 106 81.2 115 72.2 44 67.0 51 87.5 94 69.8 124 56.8 136 82.2 65 29.4 129 50.8 147 42.8 122 37.1 92 56.8 101 49.9 8
Steered-L1 [118]83.1 36.0 18 32.9 49 47.9 21 27.0 57 33.3 65 37.7 72 30.3 114 32.3 61 39.9 110 53.2 95 48.0 70 81.0 100 72.5 118 67.5 126 87.5 94 68.9 52 55.0 89 82.2 65 28.8 107 46.7 116 42.7 117 37.0 81 57.3 119 50.3 78
Nguyen [33]84.8 39.6 137 33.9 90 52.6 141 32.5 138 37.9 134 43.3 132 30.0 102 35.5 130 40.2 120 54.1 126 49.0 116 80.9 88 72.0 14 66.8 23 87.4 43 68.6 9 53.8 21 82.0 16 28.8 107 47.8 128 42.4 15 36.8 58 56.1 68 50.3 78
BriefMatch [124]85.0 36.3 46 33.3 60 48.0 34 27.2 65 33.4 68 38.5 97 30.6 122 32.6 68 40.6 126 54.0 122 48.6 100 82.8 139 72.4 102 67.3 106 87.3 22 70.2 138 55.6 114 83.9 146 28.3 22 44.3 37 42.7 117 36.6 38 55.7 44 50.5 108
StereoOF-V1MT [119]85.9 36.8 96 35.3 125 48.1 50 28.3 93 35.1 111 36.9 3 31.4 135 36.6 135 40.5 123 54.6 136 48.6 100 81.3 117 72.0 14 66.8 23 87.2 14 69.5 114 54.9 81 82.6 119 29.7 139 48.8 137 42.7 117 36.5 28 55.1 15 50.1 36
SPSA-learn [13]89.2 37.4 111 33.6 75 49.4 116 29.8 121 35.1 111 41.4 123 30.9 128 33.2 86 40.7 127 53.5 107 47.2 37 80.4 20 72.2 44 67.0 51 87.4 43 68.8 41 54.1 40 82.2 65 29.5 134 52.2 151 42.9 128 37.1 92 57.0 111 50.3 78
GraphCuts [14]89.5 38.0 126 35.1 122 49.5 118 28.4 96 33.9 81 41.3 121 31.3 134 30.8 26 40.7 127 53.7 112 48.3 91 81.0 100 72.1 29 67.1 75 87.1 7 68.6 9 54.9 81 81.7 2 28.8 107 46.3 110 42.8 122 37.7 121 58.5 140 50.4 92
HBpMotionGpu [43]90.5 38.8 131 35.9 129 50.9 135 32.1 135 38.2 136 44.4 136 29.2 41 31.7 47 39.3 76 53.9 118 49.6 126 81.5 122 72.1 29 67.0 51 87.1 7 69.5 114 54.9 81 82.4 100 28.3 22 44.4 39 42.5 71 37.3 108 56.5 89 51.1 130
Dynamic MRF [7]90.8 36.2 31 34.1 99 48.0 34 27.5 74 34.6 98 37.4 49 30.9 128 36.8 137 40.4 122 54.5 134 49.3 122 81.9 129 71.9 9 66.8 23 87.2 14 69.4 112 55.5 109 82.5 111 29.0 118 47.8 128 42.5 71 37.5 117 56.8 101 50.5 108
ROF-ND [107]90.8 37.0 103 32.8 44 48.1 50 27.7 79 34.7 99 37.6 66 29.7 86 32.3 61 39.1 55 54.2 128 51.4 141 80.4 20 72.4 102 67.3 106 87.4 43 69.5 114 56.9 137 82.0 16 29.7 139 49.0 140 43.2 137 37.8 125 57.7 130 50.2 58
2D-CLG [1]91.2 37.9 121 33.5 68 50.5 133 32.5 138 37.4 131 45.0 138 30.8 127 34.8 118 40.7 127 53.7 112 48.3 91 80.5 38 72.3 75 67.1 75 87.6 130 68.6 9 53.2 8 82.2 65 28.8 107 46.7 116 42.5 71 36.9 73 55.6 40 50.3 78
Black & Anandan [4]91.8 37.9 121 34.1 99 49.6 119 30.7 128 36.0 123 41.2 119 31.0 130 34.7 117 40.3 121 53.9 118 48.6 100 80.7 65 72.3 75 67.0 51 87.4 43 69.0 67 53.8 21 82.5 111 28.8 107 46.5 111 42.4 15 37.0 81 56.1 68 50.4 92
TV-L1-improved [17]91.8 36.6 80 34.1 99 48.4 86 28.6 102 35.1 111 37.8 78 30.5 120 33.2 86 40.0 114 52.7 65 48.0 70 80.9 88 72.3 75 67.2 90 87.4 43 69.1 80 54.9 81 82.3 86 28.8 107 47.3 122 42.6 103 37.2 101 56.7 97 50.6 119
CBF [12]92.9 36.4 59 32.5 32 48.9 107 27.5 74 33.8 78 37.9 82 29.3 51 31.6 45 39.1 55 53.2 95 48.1 77 82.6 136 72.4 102 67.2 90 87.7 140 69.2 91 55.3 100 82.3 86 28.7 99 46.1 107 42.9 128 37.9 127 57.7 130 51.7 139
Rannacher [23]95.2 36.7 90 34.5 110 48.7 100 28.7 104 35.3 115 38.1 87 30.5 120 34.0 103 39.9 110 52.7 65 48.0 70 80.8 79 72.4 102 67.2 90 87.5 94 69.0 67 54.6 67 82.3 86 28.8 107 47.0 118 42.6 103 37.1 92 56.4 86 50.6 119
Correlation Flow [75]95.8 36.2 31 33.4 65 47.7 8 27.7 79 34.3 87 37.3 37 29.4 66 31.4 39 38.8 13 53.1 93 48.5 97 81.3 117 72.8 138 67.6 129 88.6 152 70.1 131 57.1 140 82.6 119 29.4 129 48.8 137 43.0 131 37.7 121 57.9 132 50.5 108
UnFlow [129]96.3 39.2 135 37.9 137 50.6 134 32.3 136 38.9 142 41.3 121 31.6 140 38.5 144 40.8 132 53.2 95 48.7 106 80.9 88 72.0 14 66.7 13 87.4 43 69.5 114 54.6 67 82.4 100 28.2 13 43.2 6 42.4 15 39.3 148 58.3 137 51.2 131
HBM-GC [105]97.1 37.7 117 34.7 115 49.8 123 27.1 61 32.4 45 37.9 82 28.8 14 29.6 5 39.2 69 52.6 55 47.3 39 80.8 79 73.2 144 68.1 142 88.2 147 70.0 127 57.3 142 82.7 123 28.9 117 45.0 75 43.5 141 37.6 119 57.2 117 51.3 133
TriangleFlow [30]97.7 37.0 103 34.9 119 48.5 90 28.0 86 34.7 99 37.5 58 30.2 111 33.0 81 39.9 110 53.2 95 49.0 116 81.1 110 72.0 14 66.9 35 87.1 7 69.8 124 56.1 126 82.4 100 29.2 124 48.5 134 42.8 122 38.1 133 58.5 140 50.5 108
TVL1_ROB [139]98.2 40.0 140 35.0 120 52.6 141 33.3 141 38.6 140 45.0 138 29.6 81 34.3 107 39.8 107 54.0 122 48.2 88 81.3 117 72.2 44 66.9 35 87.5 94 69.0 67 53.7 19 82.5 111 28.8 107 46.5 111 42.6 103 36.8 58 56.0 59 50.5 108
SegOF [10]98.3 37.6 115 33.2 57 50.0 127 29.1 109 34.7 99 41.0 116 31.4 135 35.3 127 40.7 127 53.6 110 50.7 137 80.6 45 72.3 75 67.2 90 87.5 94 69.0 67 55.3 100 82.2 65 29.0 118 48.6 135 42.7 117 36.7 51 55.8 51 50.4 92
BlockOverlap [61]99.1 38.5 130 33.2 57 51.3 136 30.0 124 34.4 93 42.8 129 29.4 66 30.4 17 40.0 114 53.2 95 46.9 29 83.0 141 72.9 141 67.6 129 88.3 148 69.7 122 54.1 40 83.3 141 28.7 99 44.1 31 43.5 141 37.1 92 55.3 24 51.8 140
WRT [151]100.1 36.6 80 34.0 95 47.9 21 28.1 88 33.8 78 37.5 58 31.1 132 31.4 39 39.6 99 53.2 95 48.4 96 80.8 79 72.7 136 67.7 136 87.8 142 70.1 131 58.7 150 82.2 65 29.8 142 53.4 153 42.9 128 37.6 119 58.4 138 49.8 3
IAOF2 [51]101.3 37.9 121 35.9 129 49.1 110 29.6 118 36.1 126 40.0 106 29.3 51 33.4 93 40.0 114 54.1 126 50.2 132 81.0 100 72.4 102 67.4 119 87.4 43 69.2 91 54.9 81 82.4 100 28.6 86 45.5 96 42.4 15 37.9 127 57.6 126 50.6 119
Ad-TV-NDC [36]102.1 40.4 143 35.1 122 53.1 143 31.9 134 36.7 130 43.8 134 29.4 66 32.9 76 39.1 55 54.5 134 49.2 121 82.1 131 72.5 118 67.3 106 87.5 94 69.3 103 53.9 30 82.7 123 28.6 86 45.4 91 42.4 15 37.2 101 56.2 76 50.6 119
OFRF [134]107.2 38.9 132 36.1 132 50.4 130 29.5 114 35.0 108 40.5 109 29.6 81 34.4 110 39.0 46 53.3 101 48.9 113 81.1 110 72.6 131 67.7 136 87.3 22 70.1 131 57.3 142 82.5 111 29.1 122 47.6 126 42.6 103 37.4 114 58.0 133 50.0 16
AdaConv-v1 [126]110.1 37.2 110 36.6 133 47.5 4 34.3 145 39.1 145 51.1 150 36.1 149 39.4 146 52.9 153 58.2 148 53.1 147 83.8 146 70.9 3 65.4 4 86.6 4 69.7 122 54.7 71 84.4 149 38.6 152 46.5 111 77.4 154 38.2 135 54.6 8 60.3 153
LocallyOriented [52]110.4 37.5 113 35.9 129 49.2 112 29.6 118 36.2 127 39.1 100 30.1 109 33.8 100 39.5 89 53.7 112 50.0 130 81.3 117 72.3 75 67.2 90 87.5 94 70.2 138 56.2 132 82.9 132 28.8 107 45.6 100 42.5 71 37.7 121 57.6 126 50.5 108
ACK-Prior [27]111.5 36.4 59 33.7 79 48.1 50 26.7 39 33.1 59 37.1 18 30.7 124 33.3 90 39.7 103 53.6 110 50.0 130 81.0 100 73.5 148 68.6 145 88.3 148 70.8 147 59.8 151 82.7 123 29.7 139 48.7 136 43.6 144 39.5 149 62.1 151 51.3 133
StereoFlow [44]112.6 46.3 152 45.9 153 54.3 145 38.3 152 45.4 153 45.7 141 29.3 51 33.8 100 39.1 55 52.9 85 47.7 57 81.0 100 74.4 152 70.5 153 87.6 130 72.0 151 66.3 153 82.4 100 28.4 45 45.0 75 42.4 15 38.0 131 59.1 145 50.5 108
Horn & Schunck [3]112.7 37.9 121 35.1 122 49.6 119 31.4 131 37.7 133 41.8 126 31.7 141 37.4 140 41.5 135 55.8 140 50.6 134 81.3 117 72.2 44 67.0 51 87.4 43 69.2 91 54.2 47 82.7 123 29.5 134 48.9 139 42.6 103 37.8 125 57.2 117 50.9 128
WOLF_ROB [149]112.8 38.3 128 38.0 138 49.0 109 29.6 118 35.9 122 39.0 99 30.6 122 35.0 122 39.8 107 54.4 132 52.4 145 82.0 130 72.5 118 67.6 129 87.4 43 69.8 124 55.4 106 82.8 129 29.4 129 49.1 141 42.5 71 37.1 92 56.6 94 50.2 58
Filter Flow [19]113.4 37.8 120 34.6 112 49.8 123 30.8 129 36.0 123 44.3 135 29.4 66 32.4 64 39.5 89 54.2 128 48.1 77 82.2 133 72.7 136 67.7 136 87.6 130 69.2 91 55.1 91 82.5 111 28.7 99 46.5 111 42.6 103 38.3 141 58.4 138 51.4 136
TI-DOFE [24]115.5 42.0 146 37.5 136 54.8 148 35.2 147 41.1 150 46.8 145 31.4 135 37.7 141 41.6 137 56.1 142 50.6 134 81.6 124 72.0 14 66.9 35 87.2 14 69.4 112 54.4 55 82.6 119 29.2 124 47.6 126 42.6 103 38.2 135 57.5 123 50.8 127
SILK [79]117.2 39.6 137 38.1 139 51.5 140 32.4 137 38.5 139 43.6 133 32.4 142 37.2 139 41.5 135 55.4 138 49.7 127 83.0 141 72.2 44 67.0 51 87.4 43 70.0 127 54.7 71 83.4 142 29.0 118 46.5 111 42.8 122 37.4 114 56.7 97 50.7 125
Bartels [41]118.4 37.1 106 35.0 120 49.3 115 28.2 89 34.8 103 40.5 109 29.9 98 33.1 84 40.5 123 54.2 128 49.7 127 83.9 147 73.0 142 67.6 129 88.7 153 71.8 150 56.1 126 85.6 152 28.6 86 43.9 20 43.6 144 38.1 133 57.0 111 53.2 147
SLK [47]126.8 41.6 145 38.7 144 54.4 146 33.0 140 38.3 138 45.5 140 33.3 143 38.6 145 42.8 139 57.8 146 51.8 143 83.5 145 72.1 29 67.3 106 86.5 3 70.1 131 55.8 118 82.7 123 30.0 143 51.4 149 43.0 131 38.2 135 57.5 123 51.5 137
NL-TV-NCC [25]127.6 37.1 106 35.7 128 48.0 34 27.8 83 35.0 108 37.6 66 31.0 130 35.4 129 40.0 114 56.0 141 54.2 150 82.6 136 73.8 150 68.6 145 89.1 154 70.6 145 58.4 149 82.5 111 30.4 147 50.0 144 44.0 148 39.8 151 60.2 149 52.4 145
GroupFlow [9]129.4 40.3 142 40.1 146 51.3 136 31.5 132 38.9 142 42.6 128 33.5 145 39.5 147 43.8 141 54.7 137 52.3 144 81.0 100 73.2 144 68.6 145 87.6 130 70.4 143 57.3 142 83.0 134 29.3 127 48.1 131 42.5 71 37.9 127 58.2 136 50.1 36
FFV1MT [106]129.6 39.6 137 40.8 148 50.3 129 34.8 146 38.8 141 46.6 144 36.5 150 45.8 151 44.6 144 56.2 143 49.0 116 81.7 125 72.5 118 67.4 119 87.4 43 70.2 138 54.7 71 83.2 137 30.1 145 49.3 143 42.8 122 38.6 142 57.6 126 51.3 133
Learning Flow [11]130.0 37.7 117 37.0 135 49.2 112 29.9 122 37.5 132 39.7 104 31.5 139 36.3 133 40.7 127 55.4 138 51.6 142 82.6 136 72.8 138 67.8 140 87.8 142 69.6 120 55.7 115 82.8 129 29.3 127 48.4 132 42.7 117 39.2 146 59.8 147 51.2 131
Heeger++ [104]130.5 40.6 144 42.5 150 50.4 130 33.5 142 38.2 136 43.0 131 37.6 151 48.1 152 44.9 145 56.2 143 49.0 116 81.7 125 73.4 147 68.9 150 87.5 94 70.1 131 56.0 123 82.8 129 30.3 146 49.1 141 42.6 103 37.7 121 56.9 105 50.3 78
2bit-BM-tele [98]131.0 37.9 121 34.7 115 50.1 128 30.1 125 36.4 129 41.7 125 30.2 111 32.2 59 41.3 134 54.3 131 49.4 124 84.1 149 73.3 146 68.1 142 88.3 148 72.2 152 57.2 141 85.5 151 30.4 147 52.7 152 44.4 149 38.2 135 56.3 81 54.1 149
H+S_ROB [138]132.0 40.0 140 38.4 142 51.4 138 34.1 144 38.9 142 44.9 137 35.1 148 42.5 150 44.2 143 59.3 150 49.7 127 82.3 135 72.5 118 67.7 136 87.1 7 70.0 127 55.7 115 82.7 123 30.4 147 50.9 148 43.2 137 39.6 150 57.5 123 51.9 141
FOLKI [16]135.7 44.6 150 40.4 147 58.4 151 35.7 148 42.3 151 47.3 146 33.3 143 40.7 148 44.9 145 59.4 152 53.6 148 86.5 152 72.5 118 67.6 129 87.3 22 70.1 131 55.8 118 83.2 137 29.4 129 48.4 132 43.1 135 38.7 143 58.5 140 52.0 142
Pyramid LK [2]140.3 46.1 151 38.9 145 61.0 152 36.7 151 40.4 148 50.9 149 39.9 152 36.6 135 49.4 151 64.1 153 61.2 153 87.7 153 73.1 143 68.6 145 87.4 43 70.1 131 56.1 126 83.0 134 29.6 137 50.7 146 43.2 137 39.2 146 61.2 150 51.5 137
Adaptive flow [45]141.1 43.8 148 38.2 140 56.5 149 35.8 149 40.5 149 50.2 148 31.4 135 34.5 112 42.5 138 56.5 145 50.9 139 83.9 147 73.5 148 68.7 149 88.1 146 70.2 138 57.4 145 82.9 132 29.4 129 47.5 124 43.6 144 39.0 145 59.1 145 52.0 142
PGAM+LK [55]141.9 42.5 147 41.4 149 54.6 147 33.7 143 40.1 147 45.8 142 33.9 146 41.1 149 43.3 140 59.3 150 55.2 151 85.5 151 72.8 138 68.1 142 87.5 94 70.8 147 56.9 137 83.6 143 29.6 137 50.0 144 43.0 131 38.7 143 58.6 143 52.1 144
HCIC-L [99]145.8 49.1 153 42.6 151 63.0 153 35.8 149 39.4 146 52.5 151 34.6 147 37.7 141 43.9 142 58.0 147 53.9 149 81.7 125 74.0 151 69.3 151 88.5 151 71.7 149 60.5 152 83.2 137 29.5 134 47.5 124 43.9 147 40.5 152 62.9 152 52.8 146
Periodicity [78]150.3 44.4 149 43.3 152 56.9 150 42.8 153 43.4 152 56.2 153 40.9 153 49.1 153 49.5 152 58.9 149 58.6 152 84.9 150 74.4 152 70.2 152 88.0 144 73.1 153 57.9 148 86.2 153 30.0 143 51.5 150 43.5 141 41.8 153 63.4 153 53.7 148
AVG_FLOW_ROB [142]151.3 76.9 154 76.7 154 78.2 154 71.8 154 68.8 154 76.4 154 64.0 154 60.2 154 65.8 154 82.9 154 80.9 154 91.0 154 80.9 154 79.6 154 87.5 94 83.9 154 84.0 154 86.6 154 53.7 154 65.3 154 47.7 150 62.3 154 71.1 154 70.4 154
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.