Optical flow evaluation results Statistics:     Average   SD   R2.5   R5.0   R10.0   A90   A95   A99  
Error type:   endpoint   angle   interpolation   normalized interpolation  
Show images: below table   above table        
R2.5
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
PyrWarp [159]4.7 8.57 1 24.0 1 0.78 1 19.5 1 28.8 1 2.24 1 4.66 1 11.9 1 0.96 1 55.9 1 61.6 1 37.4 3 72.5 1 82.0 1 32.5 1 31.5 2 57.7 2 20.7 4 32.1 68 60.2 4 4.45 5 37.2 3 73.9 2 2.84 6
DAIN [158]13.4 11.1 5 30.7 7 2.01 79 23.2 4 34.6 8 3.46 101 6.24 3 16.0 3 1.19 2 56.0 2 63.0 2 35.4 1 74.5 3 84.0 3 34.1 3 31.5 2 60.4 5 19.6 3 32.0 53 62.4 8 3.96 2 38.7 12 77.3 8 2.38 3
NNF-Local [87]20.8 13.4 12 36.1 18 1.56 5 24.2 8 35.7 14 2.60 6 18.4 32 30.4 16 1.43 5 59.2 26 68.4 66 41.6 24 79.1 23 87.3 16 43.1 35 36.4 25 66.6 29 25.0 34 31.7 13 63.5 16 4.66 26 38.9 17 78.4 17 3.01 15
MEMC-Net+ [155]21.9 12.9 10 34.1 12 2.24 103 25.0 38 35.7 14 3.66 102 7.46 4 17.8 4 1.76 66 57.4 3 63.4 3 35.4 1 75.5 5 84.9 7 34.2 4 32.1 4 62.3 6 19.1 1 32.9 119 61.5 5 3.85 1 37.9 6 76.6 6 2.30 2
PH-Flow [101]22.9 13.7 31 37.1 45 1.77 34 24.3 11 35.5 13 2.58 4 18.5 35 30.6 19 1.54 19 58.8 6 66.8 13 41.6 24 79.0 16 87.2 14 42.9 22 36.3 18 67.1 70 24.6 15 31.6 3 63.7 25 4.64 17 39.0 25 78.6 26 3.10 44
NN-field [71]24.0 13.5 17 36.9 39 1.67 14 24.2 8 35.4 12 2.54 2 18.7 48 30.6 19 1.52 15 59.3 37 68.5 71 41.7 32 79.1 23 87.3 16 43.2 53 36.4 25 66.2 15 25.0 34 31.6 3 63.6 19 4.64 17 38.9 17 78.1 14 3.05 25
MDP-Flow2 [68]26.5 13.3 11 35.1 14 1.62 8 24.6 20 36.5 24 2.63 11 18.5 35 30.5 17 1.42 4 59.0 16 67.8 39 41.4 10 79.1 23 87.3 16 43.4 79 36.5 31 66.4 20 25.0 34 32.0 53 63.9 32 4.64 17 39.3 54 78.7 32 3.08 36
PMMST [114]27.4 13.4 12 35.0 13 1.70 21 25.1 42 37.1 37 2.73 21 18.5 35 30.5 17 1.39 3 58.9 10 67.4 25 41.5 18 79.2 44 87.4 24 43.4 79 36.3 18 66.2 15 24.9 26 31.8 23 63.8 27 4.67 29 39.2 46 78.7 32 3.09 40
COFM [59]27.6 13.6 25 36.0 16 1.89 60 24.6 20 36.4 22 2.71 19 18.5 35 30.3 14 1.59 30 58.8 6 66.8 13 41.1 8 79.0 16 87.4 24 42.6 16 35.8 10 67.2 79 24.1 7 31.2 2 61.6 6 4.89 104 38.5 8 78.1 14 3.34 109
CFRF [156]27.6 10.3 4 28.6 4 0.99 2 23.3 5 33.1 4 3.40 97 8.72 6 19.1 5 1.67 46 59.5 48 63.7 4 43.1 121 75.8 6 84.3 5 38.9 6 34.5 7 59.2 4 24.7 20 32.9 119 59.2 3 4.97 114 37.0 2 74.0 3 3.06 28
Layers++ [37]28.5 14.0 65 37.5 57 1.91 64 24.3 11 35.3 11 2.75 23 18.3 30 31.0 29 1.56 24 59.2 26 67.5 29 41.7 32 79.2 44 87.4 24 43.1 35 36.4 25 66.5 26 25.0 34 31.6 3 63.2 10 4.60 7 38.7 12 77.7 12 3.12 50
AGIF+OF [85]29.5 13.9 56 37.5 57 1.67 14 24.6 20 36.5 24 2.68 15 18.1 25 31.0 29 1.61 35 58.9 10 66.9 15 41.4 10 79.2 44 87.5 53 43.1 35 36.6 43 67.2 79 25.0 34 31.8 23 63.6 19 4.60 7 39.0 25 78.6 26 2.98 11
HAST [109]29.8 13.7 31 36.2 23 1.93 71 24.7 28 37.0 35 2.77 32 18.8 51 32.2 55 1.66 45 59.1 20 67.9 47 41.4 10 79.0 16 87.4 24 42.6 16 36.3 18 66.9 49 24.6 15 31.6 3 63.3 12 4.71 45 39.0 25 78.4 17 3.06 28
Sparse-NonSparse [56]30.2 13.8 43 37.3 50 1.81 38 24.4 14 36.0 18 2.61 7 18.0 23 31.2 34 1.52 15 59.0 16 67.1 20 42.0 55 79.2 44 87.4 24 43.1 35 36.7 59 66.7 33 25.3 77 31.7 13 63.6 19 4.63 12 38.9 17 78.5 24 3.08 36
nLayers [57]30.7 13.9 56 36.7 35 1.85 50 24.5 16 36.1 19 2.76 26 17.7 16 30.0 13 1.44 6 59.2 26 67.6 31 41.6 24 79.3 71 87.5 53 43.3 62 36.4 25 66.8 43 25.1 49 31.7 13 63.2 10 4.72 52 38.7 12 77.6 10 3.03 18
ProbFlowFields [128]33.8 13.5 17 36.6 29 1.82 41 24.4 14 36.4 22 2.68 15 18.5 35 31.2 34 1.49 10 59.2 26 67.2 21 42.1 60 79.3 71 87.5 53 43.6 108 36.5 31 67.0 59 25.2 60 31.6 3 63.5 16 4.64 17 39.0 25 78.4 17 3.06 28
2DHMM-SAS [92]35.2 14.1 75 38.9 106 1.82 41 25.5 63 38.0 51 2.77 32 17.2 12 30.9 26 1.56 24 58.9 10 66.5 10 41.7 32 79.1 23 87.4 24 42.9 22 36.5 31 66.6 29 24.9 26 31.7 13 63.9 32 4.68 37 39.2 46 79.0 47 3.07 32
OFLAF [77]35.3 13.5 17 36.1 18 1.62 8 24.3 11 35.8 16 2.62 10 18.7 48 31.5 41 1.47 8 59.1 20 67.8 39 41.2 9 79.3 71 87.4 24 43.4 79 36.6 43 67.4 95 25.0 34 31.9 38 64.3 44 4.79 81 38.9 17 78.7 32 3.10 44
LSM [39]35.5 13.9 56 38.0 75 1.78 36 24.6 20 36.5 24 2.61 7 18.1 25 32.0 49 1.55 22 59.2 26 67.6 31 42.1 60 79.2 44 87.4 24 43.1 35 36.7 59 66.9 49 25.3 77 31.7 13 63.6 19 4.65 25 38.9 17 78.6 26 3.07 32
FMOF [94]37.2 14.2 85 38.6 92 1.91 64 24.5 16 36.2 20 2.70 18 18.4 32 31.2 34 1.77 68 59.5 48 68.0 50 41.5 18 79.2 44 87.4 24 43.1 35 36.6 43 66.8 43 25.0 34 31.6 3 63.3 12 4.61 10 39.1 35 78.4 17 3.11 49
InterpCNN [160]37.5 14.5 108 31.5 10 5.94 154 24.5 16 34.1 7 5.10 133 10.4 8 20.2 7 2.68 135 61.6 140 66.9 15 39.4 5 74.3 2 83.7 2 33.5 2 30.8 1 58.8 3 19.1 1 33.3 130 58.8 2 4.44 4 37.7 5 74.6 4 2.74 5
CyclicGen [153]38.1 13.7 31 31.4 9 4.62 149 26.2 90 34.0 6 12.3 159 13.8 10 27.5 11 2.63 134 61.2 132 65.1 6 44.0 134 76.0 7 84.3 5 39.2 7 32.3 5 53.7 1 24.3 10 29.7 1 55.6 1 4.29 3 31.7 1 66.4 1 2.16 1
CombBMOF [113]39.0 13.6 25 36.4 26 1.71 24 24.5 16 36.9 33 2.58 4 18.1 25 31.5 41 1.81 76 59.5 48 68.2 56 41.6 24 79.1 23 87.3 16 43.0 28 36.8 71 66.5 26 25.0 34 33.9 145 65.2 99 4.68 37 39.1 35 78.4 17 2.92 8
ComponentFusion [96]39.2 13.4 12 36.1 18 1.72 27 24.6 20 36.8 32 2.57 3 18.9 58 32.9 69 1.69 49 59.1 20 67.8 39 41.4 10 79.2 44 87.4 24 43.6 108 36.5 31 66.3 17 25.1 49 32.0 53 64.8 71 4.76 75 39.1 35 78.7 32 3.10 44
IROF++ [58]40.3 13.8 43 37.8 67 1.72 27 24.6 20 36.6 29 2.61 7 18.6 44 31.3 37 1.64 42 58.8 6 66.7 12 41.8 40 79.0 16 87.3 16 42.7 19 36.5 31 66.6 29 25.0 34 32.0 53 65.0 79 4.74 66 39.5 82 79.2 64 3.30 104
MPRN [157]40.3 12.4 9 32.0 11 1.81 38 26.3 94 36.9 33 3.69 103 21.6 149 37.6 145 2.47 129 58.8 6 65.1 6 40.0 6 78.3 8 86.6 8 41.3 8 35.8 10 63.1 9 24.8 22 32.8 115 63.8 27 4.48 6 38.5 8 78.0 13 2.68 4
SepConv-v1 [127]41.2 9.23 2 28.0 3 1.08 3 20.5 3 32.4 3 3.35 93 8.95 7 20.5 8 2.08 104 60.8 119 66.9 15 44.2 136 79.1 23 87.1 12 43.2 53 35.6 8 62.4 7 25.1 49 32.2 85 62.3 7 5.34 141 37.6 4 76.4 5 3.28 100
Ramp [62]42.6 14.1 75 38.7 97 1.92 69 24.6 20 36.6 29 2.69 17 17.9 20 31.0 29 1.47 8 58.9 10 67.0 19 41.9 46 79.2 44 87.5 53 43.1 35 37.0 90 67.4 95 25.5 92 31.6 3 63.5 16 4.63 12 39.1 35 78.9 41 3.19 67
S2F-IF [123]43.4 13.5 17 36.6 29 1.70 21 24.9 35 37.9 47 2.77 32 18.8 51 32.7 65 1.54 19 59.1 20 67.7 35 41.6 24 79.3 71 87.5 53 43.3 62 36.5 31 67.1 70 25.0 34 31.9 38 64.7 67 4.74 66 39.3 54 79.1 57 3.10 44
OFRI [161]44.4 11.9 7 29.6 5 2.70 123 24.0 7 33.6 5 4.57 123 5.54 2 12.5 2 1.52 15 57.5 4 64.0 5 38.7 4 74.6 4 84.2 4 35.1 5 34.0 6 62.6 8 21.3 5 43.5 160 65.4 105 5.73 154 43.3 159 76.7 7 3.87 147
TV-L1-MCT [64]44.5 14.5 108 39.7 129 1.86 54 25.2 45 37.8 45 2.78 35 17.3 13 31.1 33 1.59 30 58.9 10 66.6 11 41.6 24 79.1 23 87.4 24 42.9 22 36.8 71 66.4 20 25.6 99 31.8 23 64.0 38 4.73 56 39.1 35 79.0 47 3.20 74
FlowFields+ [130]45.2 13.5 17 37.0 42 1.69 20 25.0 38 38.2 57 2.78 35 18.9 58 33.3 81 1.55 22 59.1 20 67.6 31 41.9 46 79.3 71 87.5 53 43.3 62 36.6 43 67.2 79 25.1 49 31.8 23 64.5 58 4.67 29 39.3 54 79.2 64 3.07 32
RNLOD-Flow [121]45.5 13.9 56 37.9 72 1.86 54 25.2 45 37.9 47 2.78 35 19.0 65 32.1 51 1.78 70 59.2 26 67.8 39 41.5 18 79.1 23 87.4 24 43.1 35 36.7 59 66.8 43 25.2 60 31.9 38 64.2 41 4.75 71 39.2 46 79.0 47 3.06 28
FC-2Layers-FF [74]46.1 14.0 65 38.6 92 1.84 46 24.2 8 35.1 9 2.82 45 17.9 20 31.3 37 1.51 13 59.3 37 67.7 35 42.1 60 79.3 71 87.6 91 43.3 62 36.7 59 67.4 95 25.3 77 31.6 3 63.6 19 4.67 29 39.1 35 78.7 32 3.19 67
Classic+NL [31]47.0 14.2 85 38.8 100 1.98 74 24.6 20 36.5 24 2.65 13 17.7 16 30.9 26 1.51 13 59.2 26 67.5 29 42.2 73 79.2 44 87.5 53 43.3 62 37.0 90 67.1 70 25.5 92 31.7 13 63.6 19 4.67 29 39.2 46 79.0 47 3.18 64
Classic+CPF [83]47.6 14.1 75 38.3 86 1.74 30 24.9 35 37.1 37 2.73 21 17.6 15 31.4 39 1.60 33 59.0 16 67.3 23 41.4 10 79.3 71 87.6 91 43.3 62 36.9 79 67.9 122 25.2 60 31.9 38 64.3 44 4.64 17 39.3 54 79.2 64 3.04 21
FlowFields [110]48.2 13.6 25 37.1 45 1.74 30 25.0 38 38.1 52 2.75 23 18.8 51 33.2 79 1.53 18 59.4 42 68.0 50 42.3 81 79.3 71 87.5 53 43.2 53 36.5 31 67.0 59 25.0 34 31.8 23 64.7 67 4.69 40 39.4 69 79.3 72 3.13 52
EAI-Flow [151]49.1 13.7 31 36.3 25 1.91 64 25.7 76 39.1 80 3.01 67 19.0 65 33.4 83 1.67 46 58.9 10 67.2 21 41.4 10 79.2 44 87.3 16 43.0 28 36.9 79 66.7 33 25.2 60 32.0 53 65.0 79 4.78 79 39.3 54 79.1 57 3.03 18
CtxSyn [136]50.2 9.68 3 27.4 2 1.15 4 20.4 2 31.4 2 2.64 12 8.05 5 19.1 5 1.57 27 58.6 5 65.2 8 42.5 97 79.0 16 87.1 12 43.0 28 37.9 129 65.3 12 25.7 110 38.4 159 67.7 138 5.17 130 42.3 154 78.4 17 3.48 129
NNF-EAC [103]51.1 14.2 85 37.3 50 2.09 87 25.3 51 37.6 43 2.76 26 18.9 58 30.6 19 1.61 35 59.8 74 68.5 71 43.3 125 79.1 23 87.3 16 43.1 35 36.5 31 66.5 26 25.0 34 32.1 68 64.3 44 4.73 56 39.4 69 79.0 47 3.14 54
LME [70]51.5 13.5 17 36.1 18 1.62 8 25.3 51 37.8 45 3.44 100 19.0 65 32.8 67 1.63 40 59.0 16 67.8 39 41.5 18 79.7 144 87.9 138 44.4 143 36.5 31 67.0 59 24.9 26 32.0 53 64.2 41 4.66 26 39.0 25 78.6 26 3.09 40
S2D-Matching [84]51.8 14.2 85 38.9 106 1.96 72 25.3 51 37.9 47 2.76 26 17.5 14 31.0 29 1.60 33 59.3 37 67.4 25 42.8 106 79.2 44 87.5 53 43.2 53 36.9 79 67.3 89 25.4 87 31.8 23 63.8 27 4.64 17 39.1 35 78.6 26 3.21 80
WLIF-Flow [93]52.1 13.8 43 37.4 54 1.73 29 24.9 35 37.1 37 2.81 42 18.5 35 30.9 26 1.49 10 59.4 42 67.8 39 42.5 97 79.2 44 87.4 24 43.8 133 37.2 105 67.5 102 25.9 118 31.8 23 63.9 32 4.64 17 39.4 69 78.9 41 3.14 54
FESL [72]54.6 14.4 103 39.1 113 1.83 44 25.0 38 37.4 41 2.76 26 18.2 29 31.6 43 1.70 51 59.7 62 68.5 71 41.7 32 79.3 71 87.6 91 43.3 62 36.9 79 67.9 122 25.2 60 31.8 23 63.8 27 4.61 10 39.3 54 78.8 38 3.04 21
FF++_ROB [145]56.8 13.5 17 36.6 29 1.68 17 25.4 60 38.6 68 2.89 55 19.1 78 33.5 85 1.74 61 59.3 37 68.0 50 41.8 40 79.3 71 87.5 53 43.4 79 37.1 97 66.9 49 25.9 118 31.7 13 64.3 44 4.73 56 39.3 54 79.1 57 3.20 74
JOF [140]57.2 14.4 103 39.1 113 2.17 91 24.7 28 36.3 21 2.87 52 18.1 25 30.6 19 1.54 19 59.7 62 67.9 47 43.2 123 79.3 71 87.5 53 43.6 108 36.9 79 67.0 59 25.4 87 31.6 3 63.4 15 4.66 26 39.1 35 78.7 32 3.29 101
PGM-C [120]59.2 13.8 43 37.7 66 1.85 50 25.1 42 38.1 52 2.90 56 19.1 78 33.6 86 1.59 30 59.3 37 68.2 56 41.9 46 79.3 71 87.5 53 43.5 91 36.6 43 67.2 79 25.2 60 31.9 38 64.8 71 4.67 29 39.5 82 79.4 81 3.22 82
PMF [73]59.5 13.7 31 37.1 45 1.66 13 25.5 63 39.3 83 2.71 19 19.0 65 34.9 116 1.74 61 59.4 42 68.4 66 41.8 40 79.4 105 87.6 91 43.3 62 37.3 109 66.9 49 26.2 128 31.9 38 64.3 44 4.73 56 39.3 54 78.8 38 2.93 9
MDP-Flow [26]60.0 13.4 12 36.1 18 1.67 14 24.8 32 37.2 40 2.79 39 18.8 51 32.0 49 1.70 51 59.8 74 68.9 100 42.1 60 79.3 71 87.6 91 43.5 91 36.7 59 67.7 113 25.2 60 32.5 104 65.5 111 4.77 78 39.1 35 79.0 47 3.09 40
SuperFlow [81]62.0 13.8 43 36.2 23 2.27 106 26.3 94 38.7 71 4.39 119 19.1 78 33.1 76 1.99 96 59.6 54 67.7 35 42.2 73 79.4 105 87.5 53 43.7 124 36.1 16 65.9 14 24.8 22 31.7 13 64.5 58 4.80 89 38.9 17 78.9 41 3.19 67
Efficient-NL [60]62.2 14.3 97 38.7 97 1.77 34 25.2 45 37.6 43 2.76 26 19.0 65 31.8 44 2.08 104 59.8 74 68.7 89 41.4 10 79.1 23 87.4 24 43.0 28 36.9 79 68.4 137 24.6 15 32.1 68 64.7 67 4.69 40 40.1 119 79.8 110 3.14 54
SVFilterOh [111]62.7 14.1 75 37.3 50 1.96 72 24.7 28 36.6 29 2.87 52 18.3 30 30.8 24 1.63 40 59.9 85 68.5 71 43.1 121 79.5 135 87.7 119 44.5 145 36.6 43 66.7 33 25.3 77 31.6 3 62.8 9 5.05 121 38.6 10 78.2 16 3.37 116
TC-Flow [46]62.7 13.7 31 36.9 39 1.91 64 25.3 51 38.5 65 3.05 72 19.3 97 34.1 103 1.73 57 59.2 26 67.8 39 42.2 73 79.3 71 87.5 53 43.5 91 37.1 97 68.0 126 25.6 99 31.9 38 64.3 44 4.71 45 39.0 25 79.0 47 3.13 52
AggregFlow [97]65.0 14.5 108 38.3 86 2.20 99 25.7 76 38.5 65 3.23 87 18.6 44 30.8 24 1.44 6 59.7 62 68.4 66 41.7 32 79.4 105 87.6 91 43.8 133 37.5 115 66.9 49 26.4 133 31.8 23 64.2 41 4.70 44 38.9 17 78.4 17 3.08 36
DMF_ROB [139]65.2 13.9 56 37.0 42 1.98 74 25.8 79 39.0 77 2.96 60 19.8 129 35.0 118 2.12 110 59.7 62 68.2 56 41.9 46 79.3 71 87.4 24 43.7 124 36.3 18 66.4 20 25.0 34 32.1 68 64.4 55 4.93 107 39.2 46 79.1 57 3.07 32
EPPM w/o HM [88]65.2 13.4 12 36.6 29 1.61 6 25.5 63 39.3 83 2.76 26 19.4 104 35.7 132 1.99 96 59.6 54 69.3 114 41.9 46 79.2 44 87.4 24 43.1 35 37.0 90 67.5 102 25.3 77 32.8 115 65.0 79 4.85 98 39.4 69 79.0 47 3.04 21
Second-order prior [8]65.5 14.0 65 37.1 45 2.11 88 26.2 90 39.3 83 2.93 58 19.4 104 35.1 122 2.16 115 59.4 42 67.8 39 41.8 40 79.1 23 87.3 16 43.1 35 36.5 31 66.7 33 25.0 34 32.3 92 65.4 105 4.74 66 39.5 82 79.6 98 3.19 67
PWC-Net_ROB [147]65.8 13.7 31 38.1 77 1.70 21 25.8 79 39.7 95 2.83 47 19.3 97 35.0 118 1.75 64 59.4 42 69.1 111 42.1 60 79.3 71 87.6 91 43.4 79 37.0 90 66.7 33 25.5 92 32.0 53 64.4 55 4.74 66 39.3 54 78.9 41 2.98 11
IROF-TV [53]65.8 14.0 65 38.1 77 1.99 76 24.7 28 36.5 24 2.65 13 19.1 78 34.2 104 1.78 70 59.1 20 67.4 25 42.4 91 79.4 105 87.7 119 43.6 108 36.0 12 66.4 20 24.4 11 32.1 68 64.6 63 4.75 71 39.8 104 79.9 116 3.35 112
SuperSlomo [132]66.1 12.3 8 30.3 6 2.92 133 24.8 32 35.2 10 6.60 148 13.6 9 25.5 10 2.01 98 60.5 111 65.5 9 43.9 132 78.3 8 86.6 8 41.9 10 37.4 113 64.3 10 26.4 133 35.3 156 63.9 32 5.33 140 40.3 128 77.6 10 3.51 132
DeepFlow2 [108]66.1 13.9 56 36.6 29 2.07 85 25.6 71 38.4 61 3.08 74 19.1 78 33.6 86 1.70 51 59.6 54 68.5 71 41.9 46 79.4 105 87.5 53 43.7 124 36.7 59 66.3 17 25.4 87 31.9 38 64.7 67 4.67 29 39.4 69 79.4 81 3.26 96
TF+OM [100]66.2 13.7 31 36.5 27 2.17 91 25.2 45 37.4 41 3.76 106 17.9 20 32.7 65 1.76 66 59.8 74 68.5 71 42.3 81 79.3 71 87.5 53 43.7 124 36.9 79 66.7 33 25.7 110 31.8 23 64.3 44 4.79 81 39.3 54 79.3 72 3.47 128
CPM-Flow [116]66.3 13.8 43 37.8 67 1.87 58 25.1 42 38.2 57 2.93 58 19.0 65 33.4 83 1.61 35 59.6 54 68.7 89 42.1 60 79.3 71 87.5 53 43.5 91 36.8 71 66.9 49 25.5 92 32.0 53 65.2 99 4.68 37 39.5 82 79.5 90 3.25 92
ProFlow_ROB [146]67.0 13.6 25 36.5 27 1.85 50 25.3 51 38.4 61 2.96 60 18.9 58 32.9 69 1.62 39 59.7 62 69.6 130 42.5 97 79.4 105 87.6 91 43.4 79 36.6 43 66.7 33 25.1 49 32.1 68 65.1 89 4.71 45 39.7 96 79.7 106 3.20 74
LiteFlowNet [142]67.9 13.8 43 38.6 92 1.68 17 26.0 82 40.1 107 2.84 48 19.2 89 35.3 128 1.64 42 59.8 74 69.4 120 42.3 81 79.1 23 87.4 24 42.9 22 36.6 43 67.6 108 24.4 11 32.9 119 65.8 117 4.81 91 39.6 89 78.9 41 3.03 18
EpicFlow [102]68.1 13.8 43 37.6 61 1.87 58 25.5 63 38.9 74 2.96 60 18.9 58 33.7 89 1.64 42 59.5 48 68.5 71 42.3 81 79.4 105 87.6 91 43.5 91 36.5 31 67.5 102 24.9 26 32.0 53 65.1 89 4.74 66 39.4 69 79.4 81 3.22 82
TriFlow [95]68.2 14.2 85 39.0 110 2.20 99 26.6 101 39.3 83 4.59 124 19.0 65 33.7 89 1.71 56 59.9 85 68.7 89 41.4 10 79.2 44 87.5 53 43.5 91 36.7 59 67.1 70 25.2 60 31.8 23 63.9 32 4.69 40 39.1 35 79.0 47 3.23 88
DeepFlow [86]68.5 13.7 31 35.7 15 2.03 82 25.6 71 38.2 57 3.30 91 19.2 89 33.9 96 1.74 61 59.7 62 68.0 50 42.2 73 79.4 105 87.5 53 43.7 124 37.3 109 66.4 20 26.2 128 31.8 23 64.8 71 4.63 12 39.3 54 79.3 72 3.26 96
SimpleFlow [49]68.6 14.1 75 38.9 106 1.92 69 25.5 63 37.9 47 2.85 50 19.0 65 32.3 57 2.26 120 59.2 26 67.3 23 42.4 91 79.2 44 87.5 53 43.2 53 36.7 59 67.6 108 25.1 49 32.0 53 66.1 124 5.29 136 39.3 54 79.2 64 3.15 57
SRR-TVOF-NL [91]68.6 14.2 85 37.6 61 2.07 85 26.1 86 39.8 99 3.30 91 19.4 104 33.9 96 1.82 77 59.8 74 68.6 83 41.0 7 79.1 23 87.5 53 42.9 22 36.0 12 66.9 49 24.1 7 32.9 119 64.8 71 4.81 91 39.6 89 79.4 81 3.22 82
CostFilter [40]69.6 13.6 25 37.4 54 1.63 12 25.5 63 39.7 95 2.75 23 19.0 65 36.0 135 1.79 72 59.4 42 68.8 95 42.0 55 79.4 105 87.6 91 43.7 124 38.6 140 67.1 70 28.1 150 31.9 38 64.6 63 4.81 91 39.0 25 78.5 24 3.00 13
OFH [38]69.8 14.1 75 38.2 83 2.03 82 25.6 71 38.4 61 3.01 67 19.4 104 35.1 122 1.79 72 59.5 48 68.8 95 42.3 81 79.1 23 87.4 24 43.1 35 36.7 59 67.6 108 25.2 60 32.1 68 65.1 89 4.79 81 39.2 46 79.2 64 3.15 57
Complementary OF [21]70.3 13.7 31 37.8 67 1.71 24 25.2 45 38.6 68 2.81 42 19.8 129 33.7 89 2.38 125 59.9 85 69.2 112 42.8 106 79.2 44 87.5 53 43.1 35 36.6 43 67.4 95 25.2 60 32.3 92 65.4 105 4.79 81 38.8 15 78.9 41 3.29 101
RFlow [90]70.7 13.8 43 37.8 67 2.02 80 26.0 82 39.1 80 2.85 50 19.0 65 33.1 76 1.86 80 59.7 62 68.4 66 42.2 73 79.2 44 87.6 91 43.4 79 36.1 16 66.8 43 24.5 14 32.2 85 65.1 89 4.82 97 39.7 96 79.8 110 3.34 109
Aniso. Huber-L1 [22]70.9 14.3 97 38.5 90 2.17 91 26.6 101 39.5 92 3.21 86 19.2 89 32.5 63 1.83 79 59.7 62 68.7 89 41.9 46 79.2 44 87.4 24 43.2 53 36.3 18 67.1 70 24.6 15 32.2 85 64.9 77 4.71 45 39.7 96 79.6 98 3.24 91
DPOF [18]71.0 14.2 85 39.1 113 2.19 98 24.8 32 37.0 35 2.80 40 19.3 97 31.9 45 2.01 98 60.2 102 69.5 125 42.3 81 79.1 23 87.4 24 43.1 35 36.7 59 67.1 70 24.6 15 32.4 98 65.3 103 4.81 91 39.5 82 79.5 90 3.18 64
TC/T-Flow [76]71.6 14.3 97 38.8 100 1.84 46 25.3 51 38.6 68 2.81 42 18.9 58 32.4 62 1.58 28 59.9 85 69.5 125 42.1 60 79.3 71 87.5 53 43.5 91 37.1 97 68.0 126 25.2 60 32.1 68 65.2 99 4.81 91 39.2 46 79.4 81 3.00 13
OAR-Flow [125]71.6 14.0 65 36.9 39 2.05 84 25.3 51 38.1 52 3.11 78 19.1 78 34.0 102 1.70 51 59.2 26 68.6 83 41.9 46 79.4 105 87.6 91 43.5 91 36.9 79 67.8 117 25.3 77 32.0 53 65.1 89 4.75 71 39.3 54 79.3 72 3.18 64
Brox et al. [5]72.6 14.0 65 37.4 54 1.90 62 26.4 97 40.1 107 3.08 74 19.3 97 35.0 118 1.97 92 59.7 62 68.2 56 41.7 32 79.4 105 87.6 91 43.6 108 36.6 43 66.9 49 25.1 49 31.9 38 64.8 71 4.73 56 39.4 69 79.5 90 3.15 57
Sparse Occlusion [54]72.8 14.2 85 38.6 92 1.99 76 25.8 79 39.2 82 2.78 35 19.3 97 32.3 57 1.80 75 59.8 74 68.8 95 41.7 32 79.3 71 87.5 53 43.2 53 37.1 97 68.4 137 25.3 77 32.1 68 64.4 55 4.60 7 39.7 96 79.6 98 3.15 57
ContinualFlow_ROB [152]73.5 14.6 113 40.3 136 2.11 88 26.6 101 40.8 122 3.96 110 19.8 129 36.5 142 1.98 93 59.6 54 69.0 105 42.3 81 79.2 44 87.5 53 43.4 79 36.0 12 66.8 43 24.4 11 31.9 38 64.3 44 4.64 17 39.2 46 79.4 81 3.04 21
TOF-M [154]73.8 11.7 6 31.3 8 1.68 17 23.9 6 35.8 16 5.13 136 14.0 11 25.4 9 2.53 131 60.9 121 66.9 15 43.8 129 78.9 12 87.0 11 43.4 79 38.3 135 65.3 12 26.7 138 37.9 158 65.0 79 5.51 147 43.2 157 79.5 90 3.94 149
ComplOF-FED-GPU [35]74.2 14.0 65 38.0 75 1.91 64 25.3 51 38.5 65 2.90 56 20.2 136 34.6 112 2.16 115 59.5 48 68.5 71 42.5 97 79.2 44 87.4 24 43.2 53 36.6 43 67.4 95 25.0 34 32.2 85 65.4 105 4.75 71 39.7 96 79.8 110 3.19 67
Aniso-Texture [82]74.5 13.6 25 36.6 29 1.82 41 26.2 90 39.3 83 3.20 85 19.6 115 33.0 73 1.96 91 59.7 62 68.5 71 42.6 102 79.4 105 87.6 91 43.6 108 37.0 90 68.4 137 25.7 110 31.9 38 63.8 27 4.63 12 39.4 69 79.3 72 3.16 61
LFNet_ROB [149]74.8 13.8 43 37.5 57 1.80 37 27.0 115 41.7 135 3.08 74 19.6 115 35.5 129 1.87 81 59.2 26 67.4 25 41.7 32 79.1 23 87.4 24 42.8 21 36.8 71 67.3 89 24.8 22 33.2 128 65.7 116 4.79 81 40.4 134 80.0 120 3.26 96
GraphCuts [14]75.0 15.1 127 39.3 118 2.68 121 26.4 97 39.4 90 4.50 121 19.2 89 30.7 23 2.69 136 60.7 117 68.6 83 42.8 106 79.0 16 87.4 24 42.5 14 35.6 8 66.7 33 23.7 6 32.0 53 65.0 79 5.04 120 39.0 25 79.2 64 3.48 129
Fusion [6]75.5 13.8 43 38.4 89 1.84 46 25.3 51 38.1 52 2.88 54 19.1 78 32.2 55 1.90 85 60.9 121 69.8 131 41.8 40 79.1 23 87.9 138 42.1 11 36.0 12 67.8 117 24.1 7 32.7 112 66.3 128 4.88 103 39.5 82 80.4 137 3.26 96
Classic++ [32]75.9 14.0 65 38.1 77 2.17 91 25.7 76 38.8 72 2.96 60 19.3 97 33.9 96 1.93 87 59.7 62 67.9 47 42.8 106 79.2 44 87.5 53 43.3 62 37.4 113 67.0 59 26.6 137 31.8 23 64.3 44 4.78 79 39.4 69 79.5 90 3.36 113
DF-Auto [115]76.4 14.2 85 36.7 35 2.25 104 26.5 99 39.0 77 4.23 114 18.8 51 31.4 39 1.58 28 60.1 97 69.3 114 41.6 24 79.3 71 87.5 53 43.6 108 36.6 43 67.0 59 25.1 49 32.3 92 65.1 89 4.81 91 39.9 105 80.1 125 3.22 82
Steered-L1 [118]76.9 13.7 31 37.5 57 1.84 46 25.5 63 38.9 74 3.17 84 19.7 124 33.1 76 2.40 126 60.2 102 68.5 71 42.8 106 79.4 105 87.7 119 43.5 91 36.6 43 67.0 59 25.6 99 31.8 23 64.6 63 4.96 113 38.6 10 79.0 47 3.36 113
ALD-Flow [66]77.1 14.1 75 37.9 72 2.17 91 25.4 60 38.4 61 3.14 80 19.1 78 33.9 96 1.73 57 59.6 54 69.0 105 42.6 102 79.4 105 87.6 91 43.6 108 37.0 90 67.5 102 25.6 99 31.7 13 64.0 38 4.69 40 39.4 69 79.5 90 3.20 74
p-harmonic [29]78.1 13.5 17 36.7 35 1.85 50 26.7 110 39.9 103 3.25 89 19.4 104 35.2 124 2.10 107 60.1 97 68.7 89 42.2 73 79.3 71 87.5 53 43.3 62 36.7 59 66.7 33 25.3 77 32.6 108 65.8 117 4.76 75 39.4 69 79.5 90 3.17 62
AugFNG_ROB [143]82.1 14.6 113 39.7 129 2.31 108 27.3 122 41.3 129 4.30 118 19.5 111 37.8 147 1.92 86 59.8 74 68.8 95 42.1 60 79.4 105 87.7 119 43.3 62 36.3 18 66.4 20 24.9 26 32.7 112 65.6 114 4.73 56 38.8 15 78.6 26 2.84 6
Shiralkar [42]82.2 14.2 85 39.0 110 2.02 80 26.8 111 40.3 113 2.98 64 18.5 35 38.0 150 2.48 130 60.1 97 67.7 35 41.8 40 78.8 11 87.2 14 42.3 13 37.7 122 67.2 79 26.2 128 33.2 128 67.1 134 4.94 109 39.4 69 79.3 72 3.10 44
HBM-GC [105]84.9 14.7 117 39.4 124 2.41 114 25.4 60 38.1 52 3.07 73 18.0 23 29.8 12 1.56 24 59.8 74 68.2 56 42.8 106 80.1 150 88.0 144 45.9 154 37.5 115 68.2 133 26.1 126 31.9 38 63.3 12 4.99 116 39.3 54 79.1 57 3.30 104
FlowNet2 [122]85.3 15.9 141 41.4 140 2.76 125 27.1 117 40.2 110 4.29 116 19.6 115 34.3 106 1.88 82 60.0 91 70.2 135 42.0 55 79.4 105 87.7 119 43.3 62 36.4 25 66.3 17 24.9 26 32.1 68 64.5 58 4.71 45 39.6 89 79.2 64 3.08 36
CLG-TV [48]86.4 14.3 97 38.8 100 2.17 91 26.6 101 39.8 99 3.24 88 19.5 111 33.9 96 2.11 109 60.0 91 69.0 105 42.4 91 79.3 71 87.6 91 43.5 91 36.6 43 66.9 49 25.1 49 32.1 68 65.1 89 4.71 45 39.9 105 80.0 120 3.20 74
SIOF [67]86.5 14.7 117 39.5 126 2.23 102 27.1 117 40.3 113 4.25 115 19.1 78 32.9 69 1.82 77 59.8 74 68.6 83 42.1 60 79.1 23 87.4 24 43.0 28 37.1 97 67.1 70 25.5 92 32.4 98 64.9 77 4.79 81 40.1 119 79.9 116 3.40 120
MLDP_OF [89]86.8 13.9 56 38.1 77 1.81 38 25.6 71 38.9 74 2.80 40 18.8 51 32.3 57 1.61 35 59.6 54 68.3 63 42.3 81 79.3 71 87.6 91 43.9 137 39.6 153 68.7 144 28.5 152 33.0 126 65.3 103 5.09 124 39.6 89 79.2 64 3.51 132
EPMNet [133]86.8 15.7 139 42.3 144 2.55 117 26.9 112 39.5 92 4.05 112 19.6 115 34.3 106 1.88 82 60.1 97 70.4 141 42.0 55 79.4 105 87.7 119 43.3 62 36.5 31 66.9 49 24.9 26 32.1 68 64.5 58 4.71 45 40.0 112 79.3 72 3.05 25
Local-TV-L1 [65]90.0 14.9 122 37.3 50 3.21 139 27.3 122 39.5 92 4.67 125 18.9 58 32.3 57 1.70 51 61.3 134 68.6 83 47.1 152 79.3 71 87.6 91 43.6 108 39.0 145 66.7 33 28.9 154 31.7 13 64.3 44 4.79 81 39.3 54 79.1 57 3.41 123
F-TV-L1 [15]90.5 15.0 123 39.3 118 2.88 132 27.2 120 40.2 110 3.69 103 19.2 89 34.5 111 2.19 117 59.7 62 68.4 66 42.8 106 78.9 12 87.4 24 42.7 19 37.3 109 67.0 59 25.6 99 32.1 68 64.5 58 4.89 104 40.1 119 80.0 120 3.42 124
3DFlow [135]90.6 14.1 75 38.7 97 1.71 24 25.2 45 38.2 57 2.84 48 19.0 65 32.3 57 1.69 49 59.9 85 69.0 105 42.4 91 79.6 139 87.6 91 45.1 150 37.7 122 69.2 151 25.4 87 33.7 140 66.7 131 4.86 101 39.9 105 79.8 110 3.12 50
IAOF [50]90.8 15.5 135 39.2 117 2.93 135 29.4 140 43.0 142 5.18 137 17.8 18 33.0 73 2.04 101 60.8 119 68.9 100 42.2 73 79.2 44 87.4 24 43.3 62 36.8 71 67.2 79 25.1 49 32.7 112 65.6 114 4.67 29 40.0 112 80.0 120 3.20 74
OFRF [134]91.0 16.1 143 39.8 131 3.51 143 27.6 126 40.2 110 4.76 127 18.4 32 34.3 106 1.75 64 60.4 107 68.9 100 43.0 117 79.2 44 87.5 53 43.1 35 38.1 131 68.1 130 26.4 133 32.2 85 65.0 79 4.79 81 39.0 25 79.1 57 3.05 25
TCOF [69]91.2 14.4 103 39.3 118 1.83 44 27.3 122 40.9 125 3.35 93 18.7 48 32.1 51 1.50 12 60.2 102 70.2 135 42.1 60 79.3 71 87.6 91 43.2 53 36.9 79 68.5 140 24.8 22 33.3 130 65.8 117 4.72 52 41.2 145 81.4 150 3.46 126
BriefMatch [124]93.3 14.0 65 37.0 42 2.17 91 25.6 71 38.8 72 3.98 111 19.7 124 33.0 73 2.69 136 61.1 129 69.0 105 46.4 149 79.3 71 87.6 91 43.8 133 40.5 156 67.9 122 30.6 156 31.8 23 64.0 38 4.94 109 39.0 25 78.8 38 3.34 109
Adaptive [20]94.2 14.5 108 39.6 128 2.31 108 27.1 117 40.4 116 3.35 93 18.6 44 33.7 89 1.98 93 59.6 54 68.2 56 42.4 91 79.4 105 87.6 91 43.4 79 37.1 97 67.5 102 25.7 110 32.4 98 64.8 71 4.73 56 40.0 112 80.1 125 3.38 118
IIOF-NLDP [131]94.4 14.1 75 38.2 83 1.62 8 26.1 86 39.9 103 2.98 64 19.3 97 32.1 51 1.77 68 60.6 116 69.4 120 43.2 123 79.3 71 87.5 53 43.6 108 37.8 127 68.6 141 25.6 99 34.1 149 69.5 152 5.66 153 39.9 105 79.6 98 3.01 15
CNN-flow-warp+ref [117]94.9 13.8 43 36.0 16 2.35 111 26.6 101 39.8 99 3.83 107 20.0 134 35.5 129 2.34 122 60.9 121 68.9 100 43.0 117 79.4 105 87.6 91 43.7 124 36.8 71 67.0 59 25.6 99 32.1 68 66.2 126 4.94 109 39.4 69 79.5 90 3.19 67
FlowNetS+ft+v [112]95.5 14.7 117 38.1 77 2.80 128 27.5 125 40.6 121 4.81 129 19.6 115 34.9 116 2.07 103 60.1 97 69.5 125 42.2 73 79.4 105 87.7 119 43.4 79 36.6 43 67.1 70 25.2 60 32.0 53 65.4 105 4.73 56 39.6 89 79.7 106 3.21 80
SPSA-learn [13]97.2 14.8 121 37.8 67 2.72 124 27.6 126 40.1 107 4.71 126 20.5 138 33.7 89 2.97 143 60.4 107 67.6 31 41.5 18 79.3 71 87.5 53 43.5 91 36.8 71 67.2 79 25.2 60 33.4 133 70.8 159 6.21 159 39.7 96 79.6 98 3.19 67
LDOF [28]97.7 15.0 123 38.8 100 2.92 133 28.0 133 41.1 127 5.03 132 19.7 124 34.8 115 2.15 113 60.0 91 68.9 100 42.6 102 79.4 105 87.6 91 43.5 91 36.9 79 66.8 43 25.5 92 31.9 38 65.1 89 4.73 56 39.5 82 79.6 98 3.23 88
AdaConv-v1 [126]97.7 16.5 147 42.3 144 4.36 148 30.4 146 43.8 146 9.06 155 20.6 141 36.3 139 4.45 155 64.5 153 71.3 151 45.3 144 78.4 10 86.7 10 42.2 12 36.3 18 64.9 11 25.4 87 32.5 104 63.7 25 5.53 148 38.0 7 77.4 9 3.53 135
ResPWCR_ROB [144]97.9 13.9 56 38.2 83 1.89 60 26.5 99 40.4 116 3.42 99 19.9 133 35.6 131 1.95 89 60.5 111 70.2 135 43.3 125 78.9 12 87.4 24 42.5 14 42.1 157 67.7 113 32.5 157 33.9 145 65.4 105 4.85 98 40.1 119 79.7 106 3.17 62
CRTflow [80]98.1 14.4 103 38.9 106 2.38 112 26.0 82 39.0 77 3.14 80 20.2 136 36.2 138 2.37 124 60.5 111 69.5 125 44.1 135 79.3 71 87.5 53 43.4 79 37.1 97 67.3 89 25.7 110 32.0 53 64.6 63 4.85 98 39.6 89 79.6 98 3.45 125
ROF-ND [107]98.3 15.1 127 37.9 72 1.86 54 26.3 94 40.5 120 3.12 79 19.6 115 32.8 67 1.68 48 60.9 121 71.1 150 41.9 46 79.3 71 87.5 53 43.5 91 37.0 90 68.2 133 24.9 26 34.3 151 68.3 145 5.28 135 40.5 139 80.5 140 3.25 92
HBpMotionGpu [43]98.6 15.8 140 40.2 135 3.66 145 29.5 141 42.8 141 6.27 142 18.5 35 31.9 45 1.73 57 61.3 134 69.9 132 43.9 132 79.1 23 87.6 91 43.0 28 37.6 121 67.6 108 25.9 118 32.0 53 64.3 44 4.67 29 40.0 112 79.9 116 3.75 144
Occlusion-TV-L1 [63]98.7 14.3 97 39.1 113 2.21 101 26.6 101 40.0 105 3.14 80 19.2 89 34.2 104 2.15 113 60.0 91 68.5 71 42.8 106 79.3 71 87.5 53 43.6 108 37.5 115 67.0 59 26.2 128 32.9 119 65.1 89 5.16 129 40.0 112 79.8 110 3.30 104
Modified CLG [34]99.5 14.1 75 37.6 61 2.33 110 28.5 137 41.4 133 5.68 138 19.6 115 35.8 134 2.31 121 60.2 102 68.6 83 42.1 60 79.4 105 87.5 53 43.5 91 36.7 59 67.2 79 25.2 60 32.3 92 66.0 122 4.76 75 40.2 125 80.4 137 3.40 120
CBF [12]100.6 13.7 31 37.2 49 2.15 90 26.0 82 39.4 90 3.28 90 19.1 78 32.1 51 1.79 72 61.0 127 70.0 134 45.8 146 79.6 139 87.8 136 44.9 149 36.8 71 67.4 95 25.2 60 32.2 85 65.5 111 5.22 132 40.0 112 80.2 132 3.99 152
TriangleFlow [30]102.7 14.7 117 40.0 134 2.29 107 26.6 101 40.8 122 3.03 70 19.4 104 33.3 81 2.10 107 60.4 107 69.9 132 42.8 106 79.0 16 87.4 24 42.6 16 37.7 122 68.3 136 25.3 77 33.1 127 67.8 140 5.24 134 40.4 134 80.6 142 3.32 108
2D-CLG [1]103.5 14.5 108 37.6 61 2.76 125 29.8 143 42.4 138 6.69 149 19.7 124 35.2 124 2.74 139 60.7 117 68.7 89 41.5 18 79.4 105 87.7 119 43.5 91 36.6 43 67.0 59 25.1 49 32.5 104 66.7 131 4.90 106 40.2 125 80.1 125 3.25 92
ACK-Prior [27]103.6 13.8 43 38.1 77 1.74 30 25.5 63 39.3 83 2.82 45 19.6 115 33.8 95 2.45 128 60.5 111 70.3 138 42.3 81 80.2 151 88.0 144 45.8 153 38.2 132 67.8 117 26.9 142 32.6 108 66.2 126 5.35 142 38.9 17 79.7 106 3.60 140
Nguyen [33]104.0 15.6 136 38.5 90 3.62 144 30.1 145 43.2 143 6.04 140 19.6 115 36.3 139 2.25 119 61.1 129 69.4 120 42.0 55 79.2 44 87.5 53 43.1 35 36.4 25 67.2 79 24.7 20 34.3 151 67.4 137 5.00 117 40.2 125 80.3 133 3.29 101
SegOF [10]104.1 14.2 85 36.8 38 2.54 116 27.0 115 40.0 105 4.18 113 21.1 145 36.1 137 3.15 148 60.5 111 70.7 146 41.6 24 79.4 105 87.6 91 43.6 108 36.9 79 68.2 133 25.2 60 32.5 104 68.0 144 5.31 139 39.6 89 79.4 81 3.22 82
BlockOverlap [61]104.2 15.1 127 37.6 61 3.31 141 27.7 129 39.3 83 5.73 139 18.6 44 30.3 14 2.09 106 60.9 121 68.2 56 47.1 152 80.2 151 87.9 138 46.5 155 39.0 145 67.3 89 28.4 151 31.9 38 63.9 32 5.09 124 39.7 96 79.3 72 3.55 136
IAOF2 [51]104.9 15.6 136 41.3 139 2.58 119 27.6 126 41.4 133 4.29 116 17.8 18 33.6 86 1.94 88 61.2 132 70.8 147 42.8 106 79.4 105 87.7 119 43.3 62 37.2 105 67.5 102 25.6 99 32.3 92 65.0 79 4.63 12 40.6 140 80.4 137 3.40 120
TV-L1-improved [17]106.1 14.2 85 38.8 100 2.25 104 26.9 112 40.3 113 3.40 97 19.5 111 33.9 96 2.44 127 59.9 85 69.0 105 42.7 105 79.4 105 87.7 119 43.5 91 37.2 105 67.6 108 25.8 115 32.1 68 66.1 124 5.05 121 39.9 105 80.0 120 3.46 126
Correlation Flow [75]106.8 14.0 65 38.3 86 1.61 6 26.2 90 39.8 99 2.98 64 19.1 78 31.9 45 1.73 57 60.4 107 69.4 120 43.6 128 80.2 151 87.9 138 47.8 158 38.0 130 68.7 144 26.0 123 33.4 133 67.2 135 5.29 136 40.1 119 80.3 133 3.39 119
Dynamic MRF [7]107.1 13.9 56 38.6 92 1.90 62 26.1 86 40.4 116 3.08 74 20.0 134 37.7 146 2.73 138 61.3 134 69.3 114 44.6 139 79.1 23 87.6 91 43.0 28 37.7 122 68.0 126 25.9 118 32.6 108 67.2 135 5.08 123 40.4 134 80.5 140 3.49 131
StereoOF-V1MT [119]107.2 14.6 113 39.9 132 2.00 78 27.2 120 41.9 136 3.04 71 20.9 144 37.8 147 2.85 142 61.3 134 68.3 63 43.8 129 79.2 44 87.5 53 42.9 22 38.2 132 67.8 117 26.3 132 33.8 141 68.5 146 5.36 143 40.0 112 79.4 81 3.09 40
WRT [150]108.4 14.3 97 39.0 110 1.76 33 26.6 101 39.7 95 3.14 80 20.8 142 31.9 45 2.57 132 60.2 102 69.5 125 42.3 81 79.6 139 87.7 119 44.4 143 37.8 127 69.5 156 25.5 92 34.2 150 71.4 160 6.06 157 39.7 96 79.8 110 2.95 10
Rannacher [23]110.0 14.4 103 39.3 118 2.38 112 26.9 112 40.4 116 3.36 96 19.5 111 34.6 112 2.58 133 59.8 74 68.8 95 42.8 106 79.4 105 87.7 119 43.6 108 37.2 105 67.8 117 25.8 115 32.2 85 66.0 122 5.02 118 39.9 105 79.9 116 3.56 137
Black & Anandan [4]110.6 15.3 131 38.8 100 2.96 137 28.4 135 40.9 125 4.78 128 20.5 138 35.2 124 2.74 139 60.9 121 69.3 114 42.1 60 79.4 105 87.7 119 43.6 108 37.1 97 66.6 29 25.6 99 32.9 119 65.9 121 4.72 52 40.3 128 80.3 133 3.25 92
LocallyOriented [52]113.0 15.0 123 40.3 136 2.53 115 27.7 129 41.3 129 3.86 108 19.4 104 34.4 109 1.95 89 61.1 129 70.6 143 43.3 125 79.2 44 87.5 53 43.3 62 39.1 149 68.1 130 27.6 148 32.9 119 65.8 117 4.72 52 40.6 140 80.6 142 3.37 116
UnFlow [129]115.0 16.0 142 42.8 147 2.87 131 30.6 149 45.2 155 4.52 122 21.3 148 39.4 153 2.81 141 60.0 91 68.3 63 42.1 60 79.2 44 87.4 24 43.5 91 37.5 115 68.0 126 25.2 60 33.8 141 65.1 89 4.98 115 43.2 157 81.8 154 3.67 142
Filter Flow [19]119.5 15.0 123 39.4 124 2.78 127 28.4 135 40.8 122 6.31 143 18.5 35 32.9 69 2.14 111 61.7 141 69.3 114 45.3 144 79.7 144 88.0 144 44.5 145 37.3 109 67.7 113 26.1 126 32.1 68 65.2 99 4.93 107 40.3 128 80.7 145 3.97 151
StereoFlow [44]120.6 22.8 159 51.1 160 4.80 150 36.2 159 51.1 160 6.57 146 19.2 89 34.6 112 1.89 84 60.0 91 68.5 71 42.4 91 80.3 154 89.1 159 43.9 137 39.0 145 74.1 160 25.3 77 32.1 68 65.0 79 4.73 56 40.3 128 80.9 146 3.36 113
TVL1_ROB [138]120.6 16.2 144 39.3 118 4.14 147 30.4 146 43.4 144 6.39 144 19.0 65 35.0 118 2.05 102 61.4 139 69.2 112 43.0 117 79.5 135 87.7 119 43.8 133 37.5 115 67.2 79 26.0 123 32.4 98 66.3 128 4.95 112 40.1 119 80.3 133 3.30 104
WOLF_ROB [148]125.5 15.6 136 41.7 142 2.64 120 27.8 131 41.3 129 3.72 105 19.7 124 35.2 124 2.02 100 61.3 134 71.8 152 44.5 138 79.4 105 87.8 136 43.7 124 38.8 143 67.9 122 27.5 147 34.4 153 67.7 138 5.09 124 39.9 105 79.6 98 3.22 82
Ad-TV-NDC [36]125.7 17.2 150 39.9 132 5.26 152 29.6 142 42.1 137 6.18 141 19.2 89 33.7 89 1.98 93 62.4 143 70.3 138 45.2 143 79.6 139 87.9 138 43.9 137 38.3 135 67.3 89 27.2 146 32.3 92 65.5 111 4.80 89 40.3 128 80.1 125 3.58 139
Bartels [41]126.6 14.6 113 39.3 118 2.80 128 26.1 86 39.7 95 4.45 120 19.0 65 33.2 79 2.14 111 62.1 142 70.9 148 48.9 154 80.7 158 88.1 147 49.2 160 43.7 159 69.0 150 34.8 160 32.4 98 65.0 79 5.76 155 40.4 134 80.1 125 4.26 154
TI-DOFE [24]129.0 17.9 153 43.0 148 5.41 153 32.3 155 46.2 157 7.98 153 20.5 138 38.1 151 2.97 143 63.1 150 70.6 143 43.8 129 79.1 23 87.6 91 43.1 35 37.7 122 67.4 95 25.8 115 33.4 133 67.8 140 5.09 124 41.6 149 81.5 152 3.68 143
Horn & Schunck [3]129.4 15.3 131 40.4 138 2.69 122 29.0 138 42.7 139 5.10 133 21.1 145 37.9 149 3.33 149 62.5 145 70.3 138 43.0 117 79.3 71 87.7 119 43.6 108 37.5 115 67.3 89 25.9 118 33.9 145 68.5 146 5.03 119 41.2 145 81.2 149 3.57 138
GroupFlow [9]132.4 16.8 149 43.4 149 3.43 142 29.1 139 43.9 147 5.11 135 22.2 152 39.3 152 3.53 150 61.0 127 70.6 143 42.5 97 79.7 144 88.1 147 44.0 140 39.0 145 69.4 155 26.8 141 32.8 115 66.8 133 4.87 102 40.4 134 80.1 125 3.01 15
2bit-BM-tele [98]134.2 15.3 131 39.5 126 3.22 140 27.8 131 41.2 128 4.90 130 18.8 51 32.5 63 2.34 122 62.4 143 71.0 149 49.0 155 80.6 156 88.2 152 47.9 159 42.8 158 69.3 154 32.9 158 33.4 133 70.0 156 6.77 160 40.3 128 79.4 81 4.33 157
SLK [47]135.2 17.4 151 43.9 152 4.90 151 30.5 148 44.0 148 7.18 151 22.5 153 39.8 154 4.15 154 64.5 153 70.5 142 46.7 150 78.9 12 87.7 119 41.6 9 38.5 138 68.8 146 26.0 123 33.8 141 70.1 158 5.50 146 41.6 149 81.4 150 3.91 148
SILK [79]136.0 16.3 145 42.0 143 4.01 146 29.9 144 43.5 145 6.44 145 21.6 149 37.4 144 3.55 151 62.6 146 69.4 120 47.0 151 79.3 71 87.7 119 43.6 108 39.9 154 68.1 130 29.2 155 32.8 115 67.8 140 5.14 128 40.6 140 80.6 142 3.52 134
NL-TV-NCC [25]136.2 15.1 127 41.6 141 1.86 54 26.6 101 41.3 129 3.02 69 20.8 142 35.7 132 2.24 118 63.2 151 73.9 156 45.9 147 81.3 160 88.7 158 49.9 161 38.6 140 69.8 158 25.6 99 37.6 157 69.5 152 5.62 151 42.4 155 82.1 157 4.00 153
HCIC-L [99]137.8 23.2 160 49.0 159 11.0 160 32.1 154 44.4 149 9.93 156 23.2 155 36.4 141 3.02 145 64.4 152 72.1 153 44.9 140 80.6 156 88.5 156 46.6 156 39.1 149 68.9 148 27.1 144 32.4 98 65.0 79 5.53 148 39.1 35 79.3 72 3.65 141
H+S_ROB [137]139.2 16.5 147 43.7 150 2.94 136 31.9 152 44.6 150 6.58 147 24.0 156 42.8 157 4.50 156 65.7 156 69.3 114 44.4 137 79.4 105 88.2 152 43.1 35 38.5 138 68.6 141 25.6 99 34.6 154 70.0 156 5.57 150 43.1 156 81.5 152 3.83 146
Heeger++ [104]139.7 17.5 152 47.2 158 2.80 128 31.1 151 44.9 153 4.93 131 26.6 158 47.7 159 4.79 157 62.6 146 68.0 50 45.1 141 79.8 148 88.4 155 44.1 141 39.1 149 68.9 148 26.5 136 34.8 155 67.9 143 5.23 133 41.5 148 80.1 125 3.23 88
Learning Flow [11]141.8 15.3 131 42.7 146 2.55 117 28.0 133 42.7 139 3.95 109 21.1 145 37.0 143 3.03 147 63.0 149 73.3 155 46.2 148 80.0 149 88.2 152 45.1 150 38.2 132 68.6 141 26.7 138 33.8 141 68.5 146 5.21 131 41.9 153 82.3 158 3.95 150
FFV1MT [106]142.0 16.4 146 44.7 154 3.13 138 31.9 152 44.7 151 7.15 150 25.4 157 45.6 158 5.04 158 62.6 146 68.0 50 45.1 141 79.6 139 87.9 138 44.1 141 38.9 144 67.7 113 27.1 144 34.0 148 68.5 146 5.29 136 41.8 152 81.0 147 4.48 159
Adaptive flow [45]145.2 19.6 155 44.1 153 6.76 155 32.8 156 45.7 156 10.2 157 19.8 129 34.4 109 3.02 145 64.7 155 72.1 153 49.4 156 80.3 154 88.6 157 45.6 152 38.3 135 69.2 151 26.7 138 32.6 108 66.5 130 5.45 145 41.0 143 81.1 148 3.75 144
Pyramid LK [2]147.5 21.2 158 43.7 150 10.7 159 33.1 158 45.1 154 11.9 158 27.3 159 36.0 135 6.46 159 70.7 160 78.5 160 57.7 160 79.5 135 88.1 147 43.3 62 38.6 140 68.8 146 27.0 143 33.5 138 68.8 150 6.00 156 41.0 143 81.8 154 4.31 156
FOLKI [16]151.2 20.9 156 46.0 155 9.48 158 32.8 156 47.4 158 8.75 154 21.6 149 40.7 155 4.10 153 67.2 159 74.2 158 53.7 159 79.5 135 88.1 147 43.7 124 39.2 152 69.2 151 27.9 149 33.4 133 69.5 152 5.65 152 41.7 151 82.3 158 4.28 155
PGAM+LK [55]151.6 19.4 154 46.4 156 6.81 156 30.9 150 44.8 152 7.52 152 22.7 154 40.9 156 3.99 152 66.6 158 73.9 156 52.4 158 79.7 144 88.1 147 44.5 145 40.2 155 69.7 157 28.8 153 33.3 130 69.3 151 5.42 144 41.4 147 81.8 154 4.36 158
Periodicity [78]157.9 21.0 157 47.0 157 9.32 157 38.1 160 48.1 159 14.7 160 29.8 160 47.9 160 9.27 160 66.0 157 77.1 159 50.7 157 80.8 159 89.3 160 46.8 157 45.1 160 70.6 159 35.5 161 33.5 138 69.6 155 6.07 158 43.5 160 84.0 160 6.51 160
AVG_FLOW_ROB [141]160.2 51.4 161 76.8 161 29.6 161 67.5 161 74.0 161 36.6 161 51.8 161 59.9 161 20.1 161 84.7 161 90.1 161 64.7 161 81.8 161 91.0 161 44.5 145 51.7 161 87.4 161 33.4 159 57.5 161 81.8 161 10.1 161 63.2 161 86.1 161 26.5 161
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.