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        
R1.0
normalized interpolation
error
avg. Mequon
(Hidden texture)
im0   GT   im1
Schefflera
(Hidden texture)
im0   GT   im1
Urban
(Synthetic)
im0   GT   im1
Teddy
(Stereo)
im0   GT   im1
Backyard
(High-speed camera)
im0   GT   im1
Basketball
(High-speed camera)
im0   GT   im1
Dumptruck
(High-speed camera)
im0   GT   im1
Evergreen
(High-speed camera)
im0   GT   im1
rank all disc untext all disc untext all disc untext all disc untext all disc untext all disc untext all disc untext all disc untext
DAIN [158]13.1 6.97 20 8.45 6 9.50 111 6.38 12 11.3 12 7.99 105 3.89 1 6.48 3 5.20 1 19.3 2 16.5 4 36.6 1 22.4 3 18.9 3 35.9 4 21.7 3 12.5 3 29.2 3 5.66 1 10.9 5 9.00 2 5.98 3 11.7 4 5.81 2
MDP-Flow2 [68]20.3 6.71 6 9.78 13 8.39 13 6.36 11 11.5 19 6.23 12 7.12 10 9.73 14 5.42 3 21.2 10 18.8 19 41.2 17 30.1 17 26.5 14 45.6 65 27.7 29 19.5 40 36.0 28 6.48 9 13.8 15 10.3 9 8.19 40 16.4 32 7.11 42
PyrWarp [159]20.6 3.14 1 5.13 1 3.97 1 4.97 1 8.72 1 6.66 62 4.07 2 4.89 1 7.52 136 19.0 1 14.7 1 37.9 3 20.1 1 16.4 1 34.4 1 23.0 4 11.0 2 31.7 4 6.05 5 9.54 1 10.9 120 5.91 2 10.2 2 8.42 140
PMMST [114]22.0 6.84 14 9.80 14 8.50 23 6.74 30 11.7 21 6.36 29 7.10 9 9.45 10 5.41 2 21.1 6 18.6 13 41.1 9 30.2 22 26.5 14 45.7 80 27.3 10 18.1 15 36.0 28 6.51 11 13.9 20 10.3 9 8.22 44 16.5 41 7.15 53
NNF-Local [87]23.7 6.74 8 10.1 17 8.30 6 5.97 3 10.3 4 6.14 3 7.09 8 9.63 13 5.44 4 21.7 41 20.5 79 41.2 17 30.3 32 26.6 20 45.4 36 27.9 49 20.6 80 36.0 28 6.51 11 13.8 15 10.3 9 8.04 25 16.1 20 7.10 40
PH-Flow [101]23.8 7.05 31 10.9 28 8.50 23 6.10 5 10.6 8 6.14 3 7.18 11 9.83 15 5.55 7 21.1 6 18.5 11 41.1 9 30.1 17 26.6 20 45.2 26 27.9 49 21.7 109 35.7 17 6.60 25 14.3 33 10.3 9 8.11 30 16.3 28 7.14 50
MEMC-Net+ [155]24.2 7.92 114 8.45 6 10.8 138 7.35 59 11.8 24 8.36 108 5.00 3 7.50 5 6.07 76 19.9 3 16.1 3 36.6 1 23.6 5 20.3 5 35.7 2 21.3 1 13.0 5 28.2 1 5.67 2 11.3 6 8.70 1 6.10 4 12.1 7 5.77 1
CombBMOF [113]24.5 6.93 19 9.87 16 8.43 16 6.33 10 11.4 15 6.22 11 7.60 53 10.3 22 6.25 88 21.5 27 19.4 35 41.2 17 30.2 22 26.6 20 45.3 32 27.7 29 19.2 32 36.1 41 6.57 19 14.1 23 10.3 9 7.67 10 15.4 15 6.82 6
NN-field [71]28.1 6.88 15 10.8 24 8.48 20 5.99 4 10.3 4 6.13 2 7.65 62 9.54 11 5.81 35 21.9 58 21.0 102 41.4 32 30.2 22 26.6 20 45.4 36 27.8 40 20.0 56 36.0 28 6.48 9 13.7 14 10.3 9 8.02 22 16.1 20 7.04 30
IROF++ [58]34.6 7.07 34 11.3 39 8.50 23 6.64 25 12.0 27 6.19 7 7.54 44 10.6 36 5.84 40 21.2 10 18.6 13 41.5 36 30.2 22 26.9 33 45.0 19 27.6 25 18.8 22 36.2 51 6.69 45 14.7 47 10.5 72 8.17 37 16.4 32 7.33 91
Layers++ [37]34.7 7.17 46 11.1 34 8.79 69 6.14 8 10.3 4 6.41 34 7.34 24 10.3 22 5.69 21 21.3 15 19.0 25 41.3 24 30.5 50 27.1 56 45.4 36 28.1 69 21.1 94 36.2 51 6.51 11 13.8 15 10.2 3 8.20 41 16.4 32 7.13 48
nLayers [57]36.5 7.15 43 10.4 21 8.81 74 6.44 15 11.4 15 6.42 35 7.23 14 9.30 8 5.65 15 21.4 21 19.1 28 41.5 36 30.7 89 27.4 100 45.6 65 28.0 59 20.8 84 36.2 51 6.54 17 13.6 12 10.3 9 8.07 26 16.3 28 6.91 12
MPRN [157]36.5 6.90 16 9.11 10 9.02 89 7.48 66 12.5 39 8.49 110 8.65 131 13.9 132 6.98 124 21.3 15 18.4 9 40.1 6 28.4 9 24.5 10 43.6 8 26.5 7 14.3 7 36.1 41 6.39 6 13.2 10 10.3 9 6.59 6 12.8 9 6.84 8
CFRF [156]38.5 3.55 2 5.98 2 4.25 2 7.17 47 11.4 15 8.55 111 6.22 6 7.73 6 8.62 144 21.4 21 16.5 4 42.9 125 24.5 6 20.3 5 41.4 7 26.5 7 13.1 6 36.5 84 6.46 8 10.7 4 12.1 148 6.72 7 11.7 4 9.32 152
Sparse-NonSparse [56]38.8 7.09 35 11.3 39 8.57 33 6.53 20 11.8 24 6.21 9 7.40 32 10.5 32 5.64 11 21.5 27 19.0 25 41.7 56 30.4 39 27.0 43 45.4 36 28.3 82 21.5 105 36.4 75 6.66 35 14.4 36 10.3 9 8.23 45 16.6 46 7.09 37
ProbFlowFields [128]39.3 7.03 27 11.8 66 8.66 48 6.41 13 11.7 21 6.31 22 7.18 11 10.3 22 5.58 8 21.7 41 19.6 42 41.8 64 30.7 89 27.1 56 46.0 129 27.9 49 20.5 71 36.2 51 6.51 11 13.8 15 10.3 9 7.80 13 15.6 16 7.14 50
2DHMM-SAS [92]39.9 7.27 61 12.0 75 8.59 37 7.82 76 14.2 70 6.36 29 7.25 15 10.5 32 5.74 27 21.4 21 18.7 16 41.4 32 30.3 32 26.9 33 45.2 26 27.9 49 20.3 64 36.0 28 6.64 32 14.4 36 10.3 9 8.41 63 17.0 61 7.08 34
AGIF+OF [85]40.3 7.11 38 11.3 39 8.46 17 6.68 29 12.2 31 6.27 16 7.38 29 10.1 18 5.71 25 21.2 10 18.6 13 41.1 9 30.8 101 27.6 124 45.4 36 28.3 82 22.2 127 36.0 28 6.67 37 14.0 22 10.3 9 8.34 55 17.0 61 6.91 12
CtxSyn [136]40.5 3.84 3 6.67 3 4.59 3 5.06 2 9.01 2 6.57 54 5.42 4 6.73 4 8.44 142 20.9 5 16.7 6 42.3 99 28.9 11 24.9 11 45.0 19 27.5 17 15.3 8 37.0 118 7.55 135 13.2 10 12.1 148 7.09 8 12.4 8 9.27 151
NNF-EAC [103]40.8 7.35 73 11.1 34 8.79 69 6.92 41 12.5 39 6.29 20 7.52 43 10.1 18 5.76 30 21.8 48 19.5 40 42.8 121 30.2 22 26.6 20 45.4 36 27.5 17 18.9 25 36.0 28 6.58 21 14.2 28 10.4 40 8.32 52 16.8 50 7.19 63
FlowFields [110]40.9 7.02 26 11.5 49 8.54 31 6.66 28 12.5 39 6.44 39 7.45 36 11.5 66 5.64 11 21.9 58 20.5 79 41.8 64 30.6 68 27.0 43 45.5 52 27.7 29 20.2 59 36.0 28 6.59 23 14.3 33 10.4 40 8.00 21 16.2 25 7.08 34
InterpCNN [160]41.5 12.7 153 9.11 10 18.5 157 7.86 77 11.2 11 11.5 141 7.31 21 8.52 7 11.0 157 21.9 58 17.9 7 39.6 4 22.9 4 19.6 4 35.8 3 21.4 2 12.9 4 28.8 2 5.97 4 10.4 3 10.4 40 6.10 4 11.1 3 7.58 119
FMOF [94]41.6 7.36 75 12.0 75 8.73 58 6.42 14 11.3 12 6.30 21 7.63 58 10.4 27 6.02 72 21.8 48 19.9 49 41.2 17 30.5 50 27.0 43 45.4 36 28.0 59 20.5 71 36.1 41 6.51 11 13.8 15 10.2 3 8.30 50 16.7 49 7.12 44
FlowFields+ [130]41.7 6.99 22 11.4 46 8.47 19 6.61 24 12.3 33 6.48 42 7.42 34 11.5 66 5.67 18 21.7 41 20.3 66 41.6 45 30.7 89 27.2 71 45.6 65 27.8 40 20.3 64 36.1 41 6.61 26 14.4 36 10.4 40 7.99 20 16.2 25 7.03 28
S2F-IF [123]42.6 7.01 25 11.5 49 8.48 20 6.57 21 12.2 31 6.42 35 7.40 32 11.2 62 5.64 11 21.6 31 20.1 57 41.3 24 30.7 89 27.3 85 45.7 80 27.8 40 20.4 68 36.1 41 6.71 50 14.9 62 10.4 40 7.98 19 16.1 20 7.04 30
LSM [39]44.4 7.17 46 11.8 66 8.58 35 6.64 25 12.1 30 6.17 6 7.49 39 10.9 50 5.69 21 21.6 31 19.6 42 41.6 45 30.5 50 27.1 56 45.4 36 28.3 82 21.6 107 36.3 66 6.68 43 14.6 42 10.2 3 8.35 58 16.9 59 7.03 28
LME [70]44.8 6.72 7 9.86 15 8.36 9 6.97 43 12.4 36 7.40 98 7.51 41 11.8 75 5.70 24 21.3 15 19.2 31 41.3 24 31.0 130 27.6 124 46.6 143 27.8 40 20.5 71 36.0 28 6.45 7 13.6 12 10.2 3 8.08 28 16.3 28 7.12 44
WLIF-Flow [93]46.0 6.99 22 11.0 31 8.48 20 6.76 31 12.4 36 6.39 32 7.38 29 10.3 22 5.68 20 21.4 21 18.8 19 41.9 77 30.4 39 26.9 33 45.9 118 28.8 120 21.9 116 36.9 114 6.56 18 13.9 20 10.3 9 8.34 55 16.8 50 7.15 53
TV-L1-MCT [64]46.4 7.50 94 12.5 105 8.79 69 7.19 51 13.4 55 6.37 31 7.28 18 10.6 36 5.80 34 21.4 21 18.8 19 41.3 24 30.5 50 27.1 56 45.1 23 27.9 49 18.6 18 36.6 94 6.72 54 15.0 71 10.4 40 7.92 17 15.9 18 7.20 67
ComponentFusion [96]46.5 6.91 17 10.8 24 8.55 32 6.49 19 12.0 27 6.10 1 7.49 39 11.2 62 5.72 26 21.3 15 19.4 35 41.2 17 30.6 68 27.2 71 45.8 97 27.8 40 19.6 43 36.2 51 6.92 91 16.4 115 10.4 40 8.43 67 17.0 61 7.16 57
COFM [59]47.2 7.04 28 10.7 23 8.70 52 6.60 22 11.9 26 6.35 26 7.26 16 9.93 16 5.63 10 21.2 10 18.8 19 41.0 7 30.4 39 27.3 85 44.9 18 27.7 29 22.6 131 35.1 7 6.86 83 14.7 47 11.2 131 8.67 99 17.2 80 7.78 130
MDP-Flow [26]48.5 6.83 12 10.8 24 8.50 23 6.65 27 12.4 36 6.51 48 7.46 37 10.6 36 5.88 50 22.1 80 20.6 84 41.7 56 30.4 39 26.8 28 45.6 65 28.2 78 21.9 116 36.3 66 6.69 45 14.8 56 10.4 40 8.15 33 16.6 46 7.10 40
HAST [109]49.6 6.97 20 10.2 18 8.69 50 6.46 16 11.6 20 6.26 15 7.72 73 11.1 58 5.97 64 21.1 6 18.7 16 41.1 9 30.5 50 27.5 109 44.8 16 28.2 78 22.8 137 35.5 10 6.76 64 15.2 81 10.4 40 8.81 106 18.0 116 6.96 19
CyclicGen [153]49.8 10.0 147 9.16 12 13.8 150 9.67 125 10.4 7 19.8 159 7.89 87 11.0 54 9.80 150 22.8 115 17.9 7 43.7 133 25.6 7 21.4 7 41.3 6 25.2 6 10.7 1 35.5 10 5.75 3 10.1 2 10.2 3 4.50 1 7.84 1 6.41 3
EAI-Flow [151]50.0 7.33 70 11.6 54 8.83 78 7.42 62 13.7 57 7.00 85 7.69 69 12.0 81 5.85 44 21.6 31 19.9 49 41.2 17 30.5 50 27.0 43 45.5 52 27.7 29 18.8 22 36.2 51 6.76 64 15.0 71 10.5 72 7.67 10 15.3 14 7.01 24
RNLOD-Flow [121]50.1 7.12 40 11.5 49 8.64 45 7.38 60 14.0 65 6.35 26 7.55 45 11.2 62 5.83 39 21.3 15 19.0 25 41.1 9 30.5 50 27.2 71 45.4 36 28.3 82 21.6 107 36.2 51 6.62 27 14.2 28 10.4 40 8.70 102 17.7 102 7.02 26
OFLAF [77]50.4 6.81 11 10.2 18 8.40 14 6.10 5 10.7 9 6.21 9 7.36 26 10.6 36 5.54 6 21.1 6 18.8 19 41.0 7 30.8 101 27.4 100 45.7 80 28.1 69 21.9 116 36.0 28 7.02 103 16.1 111 10.4 40 8.90 114 18.1 124 7.16 57
LFNet_ROB [149]50.8 7.27 61 11.6 54 8.82 76 7.96 82 14.9 90 7.11 89 7.95 91 13.8 129 6.10 77 21.8 48 20.6 84 41.3 24 30.1 17 26.6 20 45.1 23 27.9 49 21.0 88 35.9 21 6.51 11 14.1 23 10.3 9 7.84 14 15.7 17 7.00 21
OFRI [161]51.3 8.13 120 6.88 4 11.7 145 8.13 89 11.3 12 13.9 153 5.71 5 5.92 2 10.4 155 20.0 4 15.9 2 39.7 5 22.1 2 18.3 2 37.4 5 25.1 5 15.9 10 34.0 5 15.7 160 12.1 7 45.2 161 7.85 15 11.7 4 15.4 160
Ramp [62]53.0 7.31 68 12.1 79 8.78 66 6.60 22 12.0 27 6.27 16 7.36 26 10.4 27 5.65 15 21.3 15 18.9 24 41.4 32 30.5 50 27.1 56 45.4 36 28.7 114 22.3 129 36.6 94 6.73 60 14.9 62 10.3 9 8.55 83 17.3 85 7.26 78
PGM-C [120]55.2 7.19 51 12.1 79 8.72 55 6.82 33 12.9 44 6.62 56 7.67 66 12.2 88 5.78 33 21.9 58 20.9 96 41.8 64 30.6 68 27.1 56 45.8 97 27.7 29 19.5 40 36.2 51 6.65 33 14.7 47 10.3 9 8.20 41 16.6 46 7.31 85
Second-order prior [8]55.5 7.30 67 11.3 39 8.90 86 8.52 101 15.6 104 6.74 69 8.32 116 13.6 125 6.42 106 21.8 48 20.0 53 41.5 36 30.1 17 26.5 14 45.5 52 27.5 17 19.0 27 36.0 28 6.67 37 14.6 42 10.3 9 8.25 46 16.8 50 7.11 42
FC-2Layers-FF [74]56.1 7.22 54 11.9 73 8.70 52 6.10 5 10.2 3 6.47 41 7.31 21 10.5 32 5.64 11 21.4 21 19.1 28 41.6 45 30.7 89 27.5 109 45.6 65 28.6 108 22.7 134 36.4 75 6.77 67 15.0 71 10.3 9 8.57 87 17.2 80 7.20 67
Classic+NL [31]56.5 7.44 88 12.3 90 8.86 80 6.78 32 12.3 33 6.28 18 7.32 23 10.4 27 5.69 21 21.6 31 19.4 35 41.8 64 30.5 50 27.1 56 45.5 52 28.6 108 21.8 112 36.6 94 6.72 54 14.7 47 10.3 9 8.50 76 17.2 80 7.24 75
DeepFlow2 [108]56.8 7.28 63 11.3 39 8.88 83 7.68 71 14.4 78 6.94 83 7.58 51 12.3 92 5.88 50 21.9 58 20.2 62 41.7 56 30.5 50 26.8 28 45.9 118 27.5 17 18.0 14 36.4 75 6.67 37 14.6 42 10.4 40 8.18 38 16.4 32 7.31 85
Aniso. Huber-L1 [22]57.0 7.61 101 12.2 87 9.19 99 8.99 113 15.7 107 7.12 91 7.73 75 11.0 54 5.86 47 21.8 48 20.0 53 41.6 45 30.2 22 26.6 20 45.5 52 27.4 12 19.6 43 35.7 17 6.68 43 14.6 42 10.3 9 8.34 55 16.8 50 7.29 84
SRR-TVOF-NL [91]58.0 7.42 86 11.5 49 8.86 80 7.79 73 14.8 86 7.08 88 7.62 55 11.5 66 5.85 44 21.7 41 19.6 42 41.1 9 30.3 32 27.1 56 45.2 26 27.5 17 20.5 71 35.3 9 6.72 54 14.8 56 10.4 40 8.97 121 18.3 130 7.17 60
PWC-Net_ROB [147]58.1 7.22 54 12.8 115 8.51 29 7.26 55 14.0 65 6.49 45 7.78 80 12.7 105 5.95 60 21.6 31 20.3 66 41.6 45 30.8 101 27.5 109 45.6 65 28.2 78 20.2 59 36.5 84 6.58 21 14.2 28 10.4 40 8.02 22 16.3 28 6.87 10
FF++_ROB [145]58.4 7.00 24 11.4 46 8.50 23 7.08 45 13.3 53 6.62 56 7.67 66 11.9 78 5.93 57 21.9 58 20.6 84 41.5 36 30.8 101 27.4 100 45.7 80 28.4 94 20.5 71 36.9 114 6.67 37 14.6 42 10.4 40 8.08 28 16.4 32 7.09 37
LiteFlowNet [142]59.1 7.24 58 12.4 100 8.60 40 7.17 47 13.7 57 6.52 49 7.70 71 13.1 114 5.84 40 22.5 106 22.5 136 41.9 77 30.4 39 27.0 43 45.2 26 27.8 40 21.3 100 35.5 10 6.90 90 15.6 96 10.4 40 7.86 16 16.0 19 6.80 5
FESL [72]59.2 7.36 75 11.7 60 8.65 47 6.82 33 12.6 42 6.33 24 7.51 41 10.7 44 5.89 53 21.6 31 19.6 42 41.3 24 30.9 124 27.5 109 45.7 80 28.4 94 22.1 125 36.2 51 6.70 48 14.8 56 10.2 3 8.59 88 17.4 92 7.08 34
Classic+CPF [83]59.5 7.22 54 11.6 54 8.52 30 6.90 40 12.6 42 6.28 18 7.37 28 10.6 36 5.76 30 21.2 10 18.7 16 41.1 9 31.1 135 27.9 133 45.5 52 28.7 114 23.1 142 36.3 66 6.92 91 15.3 85 10.3 9 8.75 104 17.9 111 6.99 20
CPM-Flow [116]59.7 7.21 52 12.2 87 8.71 54 6.83 36 12.9 44 6.65 60 7.61 54 11.7 72 5.88 50 22.2 87 21.4 117 41.8 64 30.6 68 27.1 56 45.8 97 27.9 49 19.1 28 36.6 94 6.67 37 14.7 47 10.3 9 8.16 35 16.5 41 7.34 94
DF-Auto [115]60.0 7.54 96 11.1 34 9.32 108 8.42 97 14.5 81 8.82 116 7.35 25 10.3 22 5.65 15 22.0 71 20.2 62 41.5 36 30.4 39 26.7 27 45.8 97 27.5 17 18.7 20 36.1 41 6.82 75 15.3 85 10.5 72 8.43 67 17.1 69 7.20 67
S2D-Matching [84]61.6 7.37 78 12.3 90 8.80 72 7.62 70 14.2 70 6.43 38 7.28 18 10.4 27 5.74 27 21.6 31 19.1 28 42.2 96 30.6 68 27.3 85 45.4 36 28.6 108 22.5 130 36.4 75 6.76 64 14.5 40 10.3 9 8.46 70 17.0 61 7.32 88
IROF-TV [53]61.8 7.33 70 12.3 90 8.82 76 6.83 36 12.3 33 6.23 12 7.70 71 12.9 111 5.93 57 21.5 27 19.5 40 42.0 86 30.8 101 27.3 85 45.9 118 27.5 17 20.2 59 35.6 14 6.75 62 15.1 79 10.5 72 8.18 38 16.4 32 7.37 98
DeepFlow [86]62.5 7.21 52 11.0 31 8.88 83 7.79 73 14.3 73 7.33 97 7.64 60 12.6 101 5.95 60 22.1 80 20.1 57 42.0 86 30.6 68 26.8 28 46.1 134 28.0 59 17.9 13 37.2 124 6.57 19 14.1 23 10.4 40 8.07 26 16.2 25 7.32 88
EPPM w/o HM [88]62.5 6.77 10 10.4 21 8.32 7 7.00 44 13.4 55 6.16 5 8.19 108 13.6 125 6.26 89 21.7 41 20.3 66 41.5 36 30.5 50 27.2 71 45.5 52 28.5 102 21.8 112 36.5 84 6.84 77 15.6 96 10.6 99 8.41 63 17.1 69 6.95 18
Efficient-NL [60]63.5 7.28 63 11.6 54 8.61 42 7.24 54 13.3 53 6.35 26 8.21 110 10.8 46 6.39 103 21.7 41 19.6 42 41.2 17 30.4 39 27.0 43 45.3 32 28.3 82 22.8 137 35.6 14 6.86 83 15.6 96 10.4 40 9.10 128 18.3 130 7.14 50
Brox et al. [5]63.6 7.28 63 11.4 46 8.76 60 7.86 77 14.6 83 6.92 82 8.03 98 13.1 114 6.34 97 21.9 58 19.9 49 41.4 32 30.6 68 27.0 43 45.8 97 27.7 29 19.5 40 36.2 51 6.80 70 15.4 91 10.4 40 8.16 35 16.5 41 7.19 63
p-harmonic [29]63.6 7.04 28 11.3 39 8.62 43 8.81 107 15.8 110 6.98 84 7.76 78 13.1 114 6.18 85 22.4 100 20.7 89 41.9 77 30.5 50 27.0 43 45.5 52 27.8 40 19.2 32 36.4 75 6.71 50 15.1 79 10.3 9 8.29 49 16.8 50 7.12 44
ProFlow_ROB [146]64.2 7.15 43 11.3 39 8.77 63 7.31 56 14.3 73 6.72 67 7.57 50 11.5 66 5.76 30 22.0 71 21.2 114 42.2 96 30.8 101 27.4 100 45.6 65 27.6 25 18.7 20 36.1 41 6.81 73 15.3 85 10.3 9 8.55 83 17.3 85 7.31 85
JOF [140]64.3 7.63 102 12.3 90 9.19 99 6.48 17 11.4 15 6.48 42 7.27 17 10.1 18 5.67 18 21.8 48 19.2 31 42.6 117 30.8 101 27.3 85 45.8 97 28.7 114 22.2 127 36.6 94 6.59 23 14.1 23 10.3 9 8.55 83 17.1 69 7.46 105
SepConv-v1 [127]65.1 4.07 4 8.88 8 4.61 4 6.87 38 13.0 48 7.47 99 6.42 7 9.58 12 9.25 148 23.4 132 20.0 53 44.0 137 30.2 22 26.3 13 45.7 80 27.9 49 16.5 12 37.4 130 7.61 138 15.6 96 12.9 155 7.71 12 13.8 11 9.78 154
SuperSlomo [132]67.1 6.74 8 9.03 9 8.40 14 9.03 116 13.1 50 12.7 149 8.09 105 10.5 32 9.15 146 22.7 113 18.4 9 43.7 133 28.3 8 24.4 8 44.1 10 28.5 102 15.3 8 38.7 146 7.11 108 12.9 8 12.9 155 7.43 9 13.2 10 9.86 155
EpicFlow [102]67.2 7.18 48 12.0 75 8.72 55 7.42 62 14.4 78 6.72 67 7.68 68 12.1 85 5.92 56 22.1 80 21.1 107 42.0 86 30.7 89 27.1 56 45.8 97 27.5 17 19.9 54 35.9 21 6.79 69 15.2 81 10.4 40 8.40 62 17.1 69 7.33 91
DPOF [18]67.7 7.58 100 13.2 125 9.07 90 6.27 9 11.0 10 6.54 51 8.10 106 10.6 36 6.27 91 22.0 71 20.5 79 41.9 77 30.2 22 26.8 28 45.4 36 28.0 59 21.2 96 35.8 20 6.84 77 15.0 71 10.7 109 8.62 92 17.4 92 7.26 78
PMF [73]68.1 6.83 12 10.3 20 8.37 11 6.96 42 13.1 50 6.19 7 7.86 84 13.1 114 6.03 73 21.5 27 19.4 35 41.3 24 31.0 130 27.7 128 45.8 97 28.7 114 20.5 71 37.2 124 6.80 70 15.0 71 10.5 72 8.87 110 18.2 127 7.00 21
ComplOF-FED-GPU [35]69.2 7.23 57 11.8 66 8.72 55 7.20 52 13.9 62 6.62 56 8.43 121 12.6 101 6.45 107 21.9 58 20.8 95 42.3 99 30.4 39 26.9 33 45.4 36 27.7 29 20.1 57 36.1 41 6.86 83 15.4 91 10.5 72 8.55 83 17.3 85 7.28 83
TOF-M [154]69.9 5.20 5 7.84 5 6.44 5 8.53 102 13.7 57 11.0 137 7.96 92 10.9 50 10.2 153 22.5 106 18.5 11 43.6 132 29.1 12 25.0 12 45.5 52 28.8 120 16.4 11 38.7 146 7.15 112 13.1 9 12.9 155 8.03 24 14.4 12 10.2 157
DMF_ROB [139]70.0 7.31 68 11.8 66 8.81 74 7.86 77 14.9 90 6.69 65 8.52 124 13.8 129 6.40 105 22.1 80 20.7 89 41.5 36 30.5 50 26.9 33 46.0 129 27.4 12 18.9 25 36.1 41 6.95 98 14.3 33 11.2 131 8.13 31 16.4 32 7.19 63
Sparse Occlusion [54]70.4 7.37 78 12.3 90 8.87 82 8.04 85 15.3 97 6.48 42 7.58 51 10.8 46 5.87 48 22.0 71 20.4 73 41.5 36 30.6 68 27.2 71 45.5 52 28.3 82 21.8 112 36.4 75 6.80 70 15.3 85 10.3 9 8.74 103 17.7 102 7.18 62
TC/T-Flow [76]70.8 7.37 78 11.8 66 8.59 37 7.31 56 14.0 65 6.42 35 7.47 38 11.1 58 5.81 35 21.8 48 20.5 79 41.7 56 30.8 101 27.5 109 45.7 80 28.1 69 20.9 85 36.2 51 7.03 105 16.0 109 10.6 99 8.62 92 17.6 99 7.13 48
AggregFlow [97]73.0 7.71 108 12.6 108 9.11 92 7.50 67 13.9 62 7.06 86 7.19 13 9.98 17 5.53 5 21.9 58 20.4 73 41.6 45 30.8 101 27.3 85 46.1 134 29.0 125 19.7 50 37.9 138 6.75 62 14.7 47 10.5 72 8.32 52 16.8 50 7.40 102
CLG-TV [48]73.5 7.52 95 12.3 90 9.14 94 8.67 105 15.8 110 7.11 89 7.97 94 12.7 105 6.26 89 22.1 80 20.3 66 42.0 86 30.5 50 26.9 33 45.7 80 27.6 25 19.1 28 36.2 51 6.71 50 14.9 62 10.4 40 8.53 82 17.3 85 7.24 75
RFlow [90]73.5 7.24 58 12.1 79 8.90 86 8.42 97 15.6 104 6.49 45 7.72 73 12.2 88 6.01 71 22.0 71 20.6 84 41.7 56 30.4 39 27.1 56 45.7 80 27.4 12 19.8 53 35.6 14 6.84 77 15.9 107 10.5 72 8.91 117 18.0 116 7.47 109
SuperFlow [81]73.8 7.43 87 11.5 49 9.30 105 8.55 103 14.8 86 9.15 120 7.91 89 12.0 81 6.31 94 22.1 80 19.9 49 42.0 86 30.7 89 27.2 71 45.9 118 27.3 10 18.4 17 35.9 21 6.86 83 15.8 103 10.6 99 8.15 33 16.5 41 7.16 57
TCOF [69]74.4 7.36 75 12.1 79 8.68 49 9.41 122 16.6 124 7.17 93 7.38 29 10.7 44 5.61 9 21.8 48 20.4 73 41.8 64 30.4 39 27.0 43 45.6 65 28.1 69 21.8 112 35.9 21 6.85 80 15.7 102 10.4 40 9.30 138 19.0 143 7.61 124
TC-Flow [46]74.5 7.18 48 11.8 66 8.78 66 7.46 65 14.6 83 6.77 73 7.86 84 12.6 101 5.89 53 21.8 48 20.3 66 41.9 77 30.7 89 27.4 100 45.7 80 28.3 82 21.0 88 36.6 94 6.73 60 14.8 56 10.5 72 8.51 78 17.3 85 7.24 75
SIOF [67]74.6 7.66 105 12.6 108 9.09 91 9.45 123 16.6 124 8.48 109 7.65 62 11.9 78 5.98 65 21.9 58 20.1 57 41.8 64 30.0 15 26.5 14 45.3 32 28.1 69 19.7 50 36.6 94 6.63 30 14.7 47 10.5 72 8.82 107 17.9 111 7.46 105
3DFlow [135]74.8 7.09 35 11.7 60 8.46 17 6.89 39 13.0 48 6.39 32 8.03 98 10.6 36 5.98 65 21.6 31 19.3 33 41.9 77 30.8 101 27.1 56 47.6 151 29.0 125 23.9 149 36.4 75 7.05 106 16.2 112 10.5 72 8.83 109 18.0 116 7.15 53
IAOF [50]75.6 8.70 133 12.9 118 10.3 130 12.4 148 19.2 154 9.77 132 7.74 76 12.0 81 6.21 86 22.8 115 20.2 62 42.0 86 30.2 22 26.5 14 45.5 52 27.7 29 19.6 43 36.1 41 6.67 37 15.0 71 10.3 9 8.41 63 17.1 69 7.12 44
OAR-Flow [125]76.3 7.45 90 11.7 60 8.98 88 7.57 69 14.4 78 6.91 81 7.62 55 12.4 95 5.82 37 21.6 31 20.3 66 41.6 45 30.9 124 27.5 109 45.8 97 28.0 59 20.5 71 36.4 75 6.97 100 15.6 96 10.5 72 8.46 70 17.1 69 7.34 94
ContinualFlow_ROB [152]77.3 7.92 114 13.9 137 9.35 110 8.28 91 15.5 100 8.57 112 8.24 112 14.0 135 6.28 92 21.9 58 21.0 102 41.9 77 30.6 68 27.3 85 45.5 52 27.4 12 20.1 57 35.5 10 6.65 33 14.4 36 10.4 40 8.65 96 17.9 111 6.94 16
ALD-Flow [66]78.4 7.54 96 12.1 79 9.14 94 7.43 64 14.3 73 6.85 78 7.66 65 12.5 98 5.87 48 21.8 48 20.4 73 42.3 99 30.8 101 27.4 100 45.9 118 28.1 69 19.9 54 36.6 94 6.62 27 14.2 28 10.5 72 8.68 100 17.5 98 7.46 105
SVFilterOh [111]79.0 7.18 48 10.9 28 8.76 60 6.48 17 11.7 21 6.45 40 7.62 55 10.2 21 5.99 68 21.7 41 19.4 35 42.5 111 31.3 140 28.0 139 46.6 143 28.6 108 22.0 121 36.5 84 6.92 91 14.1 23 11.4 137 8.97 121 17.8 108 8.09 136
OFH [38]79.7 7.39 83 12.1 79 8.88 83 8.07 86 15.0 94 6.66 62 8.03 98 13.8 129 5.96 63 21.9 58 21.1 107 42.1 93 30.5 50 27.3 85 45.4 36 27.8 40 20.4 68 36.2 51 7.11 108 16.4 115 10.5 72 8.61 91 17.6 99 7.19 63
MLDP_OF [89]80.8 7.10 37 11.2 38 8.64 45 7.33 58 13.7 57 6.31 22 7.44 35 10.9 50 5.75 29 22.0 71 19.8 48 42.3 99 30.6 68 27.3 85 46.2 140 31.0 153 22.6 131 40.0 153 6.93 94 15.2 81 11.0 125 8.65 96 17.4 92 7.79 132
ResPWCR_ROB [144]81.5 7.34 72 12.4 100 8.76 60 7.92 81 15.1 95 7.28 95 8.37 118 13.4 122 6.22 87 22.7 113 22.2 131 43.1 128 29.7 13 26.5 14 44.6 13 32.9 157 21.5 105 43.1 157 6.66 35 14.9 62 10.3 9 8.50 76 17.4 92 7.00 21
CostFilter [40]81.7 6.91 17 11.1 34 8.37 11 6.82 33 12.9 44 6.25 14 7.99 95 13.9 132 6.10 77 21.9 58 20.6 84 41.7 56 31.1 135 27.9 133 45.9 118 29.8 140 20.3 64 39.1 149 6.94 96 15.8 103 10.6 99 8.82 107 18.1 124 7.09 37
Fusion [6]81.8 7.13 42 12.3 90 8.60 40 7.18 50 13.1 50 6.56 52 7.63 58 10.9 50 6.13 82 22.5 106 21.1 107 41.5 36 30.7 89 28.2 144 44.3 11 28.1 69 23.8 147 35.2 8 7.22 119 17.9 130 10.6 99 9.64 146 19.9 150 7.32 88
Modified CLG [34]83.5 7.63 102 11.6 54 9.65 113 10.7 136 17.2 133 10.7 136 8.25 113 14.3 139 6.60 113 22.4 100 21.1 107 41.8 64 30.6 68 26.9 33 45.8 97 27.7 29 19.2 32 36.3 66 6.69 45 14.9 62 10.4 40 8.41 63 17.0 61 7.35 97
IIOF-NLDP [131]84.0 7.04 28 10.9 28 8.36 9 7.81 75 14.8 86 6.64 59 8.07 104 11.0 54 6.12 79 22.3 94 20.0 53 42.8 121 30.4 39 27.0 43 45.9 118 29.1 131 23.2 143 36.5 84 8.40 155 24.6 157 11.3 134 8.78 105 17.8 108 6.86 9
F-TV-L1 [15]84.6 8.24 123 13.1 122 9.92 122 9.28 118 16.3 119 7.48 100 8.00 96 13.2 120 6.35 99 22.3 94 20.9 96 42.3 99 29.9 14 26.9 33 44.8 16 27.9 49 19.4 37 36.5 84 6.87 87 15.4 91 10.5 72 8.46 70 16.8 50 7.58 119
FlowNet2 [122]85.5 9.30 138 14.6 142 10.5 133 8.42 97 14.6 83 9.24 124 8.03 98 12.5 98 6.14 83 22.2 87 21.9 127 41.8 64 30.9 124 27.5 109 45.8 97 28.0 59 20.5 71 36.0 28 6.72 54 14.9 62 10.4 40 8.31 51 16.9 59 7.01 24
EPMNet [133]85.6 9.02 136 14.8 145 10.2 128 8.29 93 14.1 69 8.78 115 8.03 98 12.5 98 6.15 84 22.8 115 23.9 150 41.7 56 30.9 124 27.5 109 45.8 97 28.0 59 21.4 102 35.9 21 6.72 54 14.9 62 10.4 40 8.21 43 16.8 50 6.83 7
Complementary OF [21]85.7 7.11 38 12.1 79 8.50 23 7.17 47 14.0 65 6.58 55 8.76 133 12.0 81 6.55 110 22.3 94 21.4 117 42.6 117 30.6 68 27.5 109 45.2 26 28.1 69 20.9 85 36.4 75 7.15 112 16.7 119 10.5 72 9.09 126 18.7 137 7.38 99
SimpleFlow [49]86.1 7.37 78 12.4 100 8.74 59 7.88 80 14.3 73 6.50 47 8.59 127 11.5 66 6.51 108 21.6 31 19.3 33 41.8 64 30.6 68 27.3 85 45.5 52 28.5 102 22.9 140 36.2 51 7.66 141 20.5 150 10.8 119 8.89 113 18.2 127 7.15 53
AugFNG_ROB [143]86.3 8.05 116 12.5 105 9.84 121 8.99 113 15.9 115 9.32 127 8.37 118 15.6 145 6.30 93 22.5 106 22.3 132 41.9 77 31.2 138 28.0 139 45.6 65 27.8 40 20.2 59 35.9 21 6.83 76 15.0 71 10.4 40 7.96 18 16.4 32 6.68 4
TF+OM [100]87.3 7.41 84 12.1 79 9.19 99 7.21 53 12.9 44 7.83 102 7.55 45 12.3 92 5.82 37 22.2 87 21.0 102 41.9 77 30.8 101 27.5 109 46.0 129 28.3 82 20.5 71 36.8 109 6.97 100 16.3 113 10.5 72 8.65 96 17.3 85 7.75 128
LDOF [28]87.7 8.08 117 12.3 90 9.79 119 8.94 111 14.9 90 9.18 122 8.23 111 13.5 124 6.52 109 22.3 94 21.1 107 42.4 109 30.6 68 27.0 43 45.8 97 27.9 49 18.8 22 36.6 94 6.77 67 15.3 85 10.4 40 8.44 69 17.1 69 7.38 99
ROF-ND [107]87.8 7.46 91 11.0 31 8.77 63 7.96 82 15.4 98 6.76 72 7.55 45 11.0 54 5.85 44 23.3 129 23.5 148 41.6 45 30.6 68 27.1 56 45.8 97 28.3 82 22.7 134 35.9 21 7.52 133 17.5 125 11.4 137 9.25 136 18.7 137 7.27 80
TriFlow [95]89.1 7.77 110 13.7 134 9.28 103 8.98 112 15.7 107 9.30 126 7.65 62 12.4 95 5.84 40 22.0 71 20.9 96 41.1 9 30.9 124 27.7 128 45.7 80 28.4 94 21.3 100 36.3 66 6.85 80 15.5 95 10.4 40 8.69 101 17.4 92 7.23 74
Local-TV-L1 [65]89.7 8.46 129 12.6 108 10.4 131 9.68 126 16.0 117 8.93 118 7.56 49 11.2 62 5.84 40 23.1 124 20.4 73 46.0 150 30.6 68 27.1 56 45.9 118 30.1 145 19.1 28 39.9 152 6.72 54 14.9 62 10.5 72 8.13 31 16.1 20 7.58 119
Classic++ [32]89.9 7.49 93 12.5 105 9.11 92 8.07 86 15.2 96 6.67 64 7.89 87 12.6 101 6.04 74 22.3 94 20.7 89 42.2 96 30.6 68 27.2 71 45.7 80 29.0 125 21.0 88 37.6 133 6.81 73 15.2 81 10.5 72 8.62 92 17.4 92 7.46 105
Occlusion-TV-L1 [63]90.3 7.44 88 12.3 90 9.14 94 8.91 110 16.5 122 6.85 78 7.83 82 12.8 108 6.32 95 22.6 112 21.5 122 42.5 111 30.5 50 26.9 33 45.8 97 28.4 94 19.6 43 37.1 120 7.15 112 14.8 56 10.7 109 8.51 78 17.1 69 7.34 94
Nguyen [33]91.9 9.74 143 12.6 108 12.4 147 12.3 146 18.6 149 11.1 138 8.27 115 14.8 141 6.69 115 23.4 132 21.7 124 41.8 64 30.3 32 26.8 28 45.3 32 27.4 12 19.6 43 35.7 17 7.24 121 18.3 133 10.5 72 8.37 60 17.0 61 7.22 73
2D-CLG [1]92.0 8.44 127 12.3 90 10.6 135 11.9 144 18.0 143 12.3 147 8.94 136 13.9 132 7.33 132 23.1 124 21.2 114 41.3 24 30.5 50 26.9 33 45.8 97 27.6 25 19.2 32 36.2 51 7.14 110 17.2 123 10.5 72 8.37 60 16.5 41 7.20 67
FlowNetS+ft+v [112]92.2 7.81 112 11.7 60 9.63 112 9.77 128 16.8 126 9.16 121 8.06 103 13.4 122 6.36 100 22.1 80 20.7 89 42.1 93 30.8 101 27.4 100 45.8 97 27.7 29 19.4 37 36.3 66 7.01 102 16.4 115 10.5 72 8.51 78 17.2 80 7.33 91
Aniso-Texture [82]92.8 7.16 45 11.6 54 8.78 66 8.84 108 16.5 122 6.86 80 8.38 120 11.8 75 5.99 68 22.4 100 21.4 117 42.5 111 31.0 130 27.5 109 46.0 129 29.0 125 24.2 151 36.7 104 6.70 48 14.7 47 10.3 9 8.90 114 18.0 116 7.27 80
Adaptive [20]93.3 7.71 108 13.2 125 9.21 102 9.40 121 16.8 126 7.07 87 7.87 86 12.4 95 6.12 79 22.0 71 20.3 66 41.8 64 30.7 89 27.3 85 45.6 65 28.4 94 20.7 83 36.8 109 6.95 98 16.0 109 10.4 40 8.87 110 17.9 111 7.55 116
Shiralkar [42]93.5 7.48 92 12.8 115 8.80 72 9.00 115 15.8 110 6.65 60 8.52 124 16.1 146 6.84 120 23.4 132 22.3 132 41.6 45 30.0 15 27.0 43 44.5 12 28.7 114 21.1 94 37.1 120 7.49 131 18.7 141 10.6 99 8.64 95 17.7 102 6.93 14
CNN-flow-warp+ref [117]97.5 7.35 73 10.8 24 9.30 105 8.87 109 16.2 118 8.14 107 8.60 128 14.1 137 6.62 114 23.7 137 21.9 127 42.7 119 30.8 101 27.3 85 45.9 118 28.0 59 19.1 28 36.7 104 7.37 126 18.5 138 10.6 99 8.33 54 16.8 50 7.27 80
CRTflow [80]97.8 7.69 106 12.6 108 9.28 103 8.45 100 15.5 100 6.81 76 8.55 126 14.0 135 7.29 130 22.4 100 20.7 89 43.8 136 30.7 89 27.2 71 45.7 80 28.1 69 19.6 43 36.7 104 6.87 87 15.8 103 10.6 99 8.59 88 17.2 80 7.65 125
Black & Anandan [4]98.1 8.54 130 12.8 115 10.2 128 10.9 138 17.3 136 9.40 128 9.06 138 13.6 125 6.99 125 22.9 121 21.3 116 41.7 56 30.7 89 27.2 71 45.9 118 28.0 59 18.6 18 36.7 104 6.93 94 15.9 107 10.4 40 8.46 70 17.0 61 7.20 67
HBpMotionGpu [43]98.5 9.39 140 14.6 142 11.3 141 11.7 143 18.9 151 11.5 141 7.55 45 11.1 58 6.00 70 23.3 129 22.3 132 43.5 131 30.3 32 27.2 71 45.2 26 28.7 114 20.9 85 37.1 120 6.62 27 14.2 28 10.5 72 8.99 123 17.8 108 8.04 135
GraphCuts [14]98.8 8.65 132 14.1 140 9.83 120 8.28 91 14.2 70 9.28 125 9.89 148 10.6 36 7.38 133 23.0 122 21.1 107 42.5 111 30.3 32 27.3 85 44.7 15 27.2 9 21.4 102 34.7 6 7.42 129 17.8 128 11.0 125 9.32 139 18.9 141 7.66 126
StereoOF-V1MT [119]99.1 7.65 104 13.5 131 8.77 63 8.69 106 15.9 115 6.52 49 9.43 144 15.4 143 7.23 128 24.4 143 22.3 132 43.2 129 30.5 50 27.2 71 45.0 19 28.9 123 21.2 96 37.2 124 7.77 144 19.4 145 11.0 125 8.26 48 16.4 32 6.93 14
HBM-GC [105]99.9 7.91 113 12.6 108 9.75 117 7.51 68 13.9 62 6.80 75 7.29 20 9.43 9 5.94 59 22.0 71 19.7 47 42.3 99 32.1 151 28.6 148 48.0 153 30.0 142 24.6 154 37.8 135 7.14 110 14.8 56 11.6 141 8.95 120 17.7 102 8.28 138
CBF [12]100.0 7.41 84 11.9 73 9.31 107 8.07 86 14.9 90 7.14 92 7.69 69 11.1 58 5.95 60 22.8 115 20.7 89 45.1 146 30.8 101 27.3 85 47.0 148 28.2 78 20.6 80 36.5 84 7.17 116 16.6 118 11.2 131 9.16 133 17.9 111 8.83 146
TriangleFlow [30]100.4 7.79 111 13.0 121 9.16 98 8.36 96 15.5 100 6.69 65 8.20 109 11.9 78 6.59 111 22.5 106 21.0 102 42.5 111 30.1 17 27.0 43 45.0 19 28.9 123 22.6 131 36.5 84 7.42 129 18.3 133 11.0 125 9.49 143 19.3 145 7.47 109
Steered-L1 [118]100.4 7.06 33 12.2 87 8.59 37 7.40 61 14.3 73 6.83 77 8.48 123 11.7 72 6.69 115 22.8 115 20.9 96 42.7 119 31.2 138 28.1 142 45.8 97 28.3 82 21.2 96 36.7 104 7.25 122 17.8 128 10.9 120 9.00 124 18.3 130 7.58 119
Correlation Flow [75]100.6 7.05 31 11.7 60 8.32 7 8.29 93 15.6 104 6.56 52 7.64 60 10.8 46 5.89 53 22.2 87 20.1 57 42.9 125 31.7 145 27.7 128 49.9 158 29.6 136 23.8 147 37.2 124 7.62 139 19.0 143 11.3 134 9.22 135 18.6 136 7.51 115
IAOF2 [51]103.5 8.43 126 13.6 133 9.76 118 9.86 130 17.4 137 8.67 113 7.74 76 12.2 88 6.33 96 23.1 124 21.7 124 42.3 99 31.0 130 27.9 133 45.6 65 28.5 102 21.0 88 36.6 94 6.71 50 15.0 71 10.3 9 9.14 131 18.4 133 7.49 113
WRT [150]105.9 7.26 60 11.8 66 8.58 35 8.55 103 14.8 86 6.77 73 9.36 143 10.8 46 6.71 119 22.2 87 20.1 57 42.0 86 31.3 140 28.0 139 46.9 147 29.4 133 25.4 156 36.5 84 9.01 158 28.1 160 11.7 144 9.92 150 21.0 154 6.94 16
BriefMatch [124]106.0 7.38 82 12.0 75 8.85 79 7.71 72 14.5 81 7.86 103 8.77 134 11.7 72 7.25 129 24.2 141 22.0 129 46.2 151 30.8 101 27.4 100 46.1 134 31.8 156 21.7 109 41.3 156 6.85 80 15.3 85 10.7 109 8.51 78 17.1 69 7.56 118
SegOF [10]108.0 8.16 121 12.4 100 10.1 127 9.10 117 15.5 100 8.83 117 9.48 145 14.1 137 7.46 135 22.8 115 23.0 146 41.6 45 30.8 101 27.4 100 45.8 97 28.3 82 22.0 121 36.3 66 7.83 145 21.5 152 11.0 125 8.46 70 17.1 69 7.17 60
TV-L1-improved [17]108.7 7.55 98 12.9 118 9.15 97 9.36 119 16.9 128 7.19 94 8.63 130 12.2 88 6.92 122 22.2 87 21.0 102 42.3 99 30.8 101 27.5 109 45.6 65 28.5 102 21.4 102 36.8 109 7.38 127 18.6 140 10.7 109 8.94 118 18.0 116 7.75 128
BlockOverlap [61]109.5 8.81 134 12.4 100 11.1 140 10.0 133 15.8 110 10.6 135 7.84 83 10.4 27 6.59 111 23.3 129 20.4 73 46.3 152 31.9 148 27.9 133 48.8 156 30.3 149 19.7 50 39.8 151 7.08 107 14.5 40 11.7 144 8.35 58 16.1 20 8.60 144
OFRF [134]109.8 9.30 138 13.4 130 11.0 139 9.59 124 15.7 107 9.07 119 7.92 90 12.7 105 6.05 75 22.3 94 20.2 62 42.8 121 31.0 130 27.9 133 45.4 36 29.6 136 23.3 144 37.4 130 7.34 124 17.7 127 10.5 72 9.09 126 18.8 140 7.05 32
Dynamic MRF [7]110.3 7.29 66 13.1 122 8.69 50 8.20 90 16.3 119 6.74 69 9.18 140 16.4 149 7.22 127 24.5 145 23.1 147 44.4 140 30.3 32 27.2 71 45.1 23 29.2 132 23.4 145 37.2 124 7.64 140 19.8 148 10.7 109 9.14 131 18.0 116 7.48 112
LocallyOriented [52]110.9 8.08 117 13.1 122 9.72 115 9.73 127 17.0 131 7.88 104 8.34 117 12.8 108 6.34 97 23.0 122 22.1 130 43.0 127 30.6 68 27.2 71 45.6 65 30.0 142 21.9 116 38.5 145 7.02 103 15.8 103 10.5 72 9.05 125 18.4 133 7.39 101
AdaConv-v1 [126]111.8 9.81 145 13.9 137 11.6 144 12.1 145 17.6 139 16.0 155 11.4 155 16.1 146 13.1 158 26.5 154 24.4 155 45.3 147 28.4 9 24.4 8 44.6 13 28.4 94 18.1 15 37.7 134 7.74 143 16.3 113 13.1 158 8.25 46 15.1 13 10.1 156
Rannacher [23]111.8 7.69 106 13.2 125 9.32 108 9.37 120 16.9 128 7.28 95 8.67 132 13.0 112 6.91 121 22.2 87 21.1 107 42.4 109 30.8 101 27.5 109 45.7 80 28.5 102 21.2 96 36.9 114 7.35 125 18.5 138 10.7 109 8.90 114 18.0 116 7.78 130
SPSA-learn [13]112.0 8.28 124 12.9 118 9.95 124 9.92 131 16.3 119 9.49 129 9.15 139 12.8 108 7.30 131 23.1 124 20.5 79 41.6 45 30.8 101 27.5 109 45.7 80 28.0 59 20.4 68 36.3 66 8.81 157 27.1 159 11.8 146 10.0 152 21.0 154 7.20 67
ACK-Prior [27]113.2 7.12 40 11.7 60 8.57 33 7.08 45 13.8 61 6.34 25 8.81 135 11.8 75 6.69 115 22.5 106 21.4 117 42.3 99 32.6 155 29.3 154 48.2 154 30.7 152 25.6 157 38.1 141 7.95 148 18.8 142 12.0 147 10.8 157 21.8 157 8.53 143
Ad-TV-NDC [36]113.6 10.8 150 13.9 137 13.4 149 11.6 142 17.6 139 11.2 139 7.77 79 12.3 92 6.12 79 24.0 139 21.6 123 44.4 140 31.1 135 27.6 124 46.1 134 29.0 125 19.3 36 38.0 139 6.87 87 15.4 91 10.5 72 8.59 88 17.0 61 7.71 127
TVL1_ROB [138]115.1 10.2 149 13.8 135 12.5 148 12.6 152 19.1 152 11.9 144 8.01 97 13.7 128 6.37 101 23.6 135 21.4 117 42.3 99 30.8 101 27.3 85 46.0 129 28.6 108 20.2 59 37.1 120 7.29 123 17.6 126 10.7 109 8.49 75 17.1 69 7.40 102
Horn & Schunck [3]115.3 8.45 128 13.3 129 10.0 125 11.4 141 18.1 146 9.84 133 9.65 146 16.1 146 7.89 139 24.6 146 22.8 140 42.8 121 30.6 68 27.2 71 45.6 65 28.3 82 19.4 37 36.8 109 7.41 128 18.0 132 10.6 99 8.94 118 17.7 102 7.55 116
UnFlow [129]117.5 9.13 137 15.0 147 10.7 136 10.9 138 18.1 146 9.23 123 9.21 141 16.9 152 7.18 126 22.4 100 21.8 126 41.8 64 30.8 101 27.6 124 45.9 118 28.6 108 22.7 134 36.0 28 6.94 96 15.6 96 10.5 72 10.0 152 19.3 145 7.47 109
StereoFlow [44]118.2 13.8 156 20.2 159 14.0 151 14.1 156 21.3 159 12.0 146 7.79 81 13.3 121 5.98 65 22.4 100 20.9 96 42.1 93 33.7 159 32.3 159 46.1 134 30.5 150 31.8 160 36.3 66 6.63 30 14.7 47 10.4 40 9.98 151 21.0 154 7.42 104
TI-DOFE [24]118.9 11.8 152 14.7 144 14.8 153 13.9 155 20.3 156 13.5 152 9.26 142 16.5 151 7.69 138 25.1 148 22.8 140 43.3 130 30.2 22 27.1 56 45.4 36 28.4 94 19.6 43 36.8 109 7.22 119 17.2 123 10.7 109 9.21 134 18.1 124 7.59 123
WOLF_ROB [148]124.1 8.87 135 16.2 152 9.71 114 9.97 132 16.9 128 7.77 101 8.60 128 13.0 112 6.39 103 23.2 128 23.5 148 43.7 133 30.9 124 27.9 133 45.8 97 30.1 145 22.8 137 38.2 143 7.54 134 19.0 143 10.7 109 9.12 130 18.7 137 7.06 33
Filter Flow [19]124.3 8.30 125 13.2 125 10.0 125 10.8 137 17.1 132 11.7 143 7.96 92 12.1 85 6.38 102 23.7 137 20.9 96 44.5 144 31.5 144 28.2 144 46.7 145 28.8 120 21.0 88 37.3 129 7.15 112 17.0 122 10.7 109 9.54 145 18.9 141 8.47 142
NL-TV-NCC [25]126.8 7.56 99 12.7 114 8.62 43 8.00 84 15.4 98 6.74 69 8.46 122 13.1 114 6.70 118 24.2 141 24.0 153 45.0 145 32.8 156 28.3 147 52.0 161 29.4 133 24.1 150 36.9 114 7.89 147 17.9 130 12.4 153 10.1 154 19.9 150 8.92 147
Bartels [41]128.7 8.10 119 13.8 135 9.94 123 8.35 95 15.8 110 8.75 114 8.11 107 12.1 85 6.97 123 24.1 140 22.7 137 47.6 154 32.4 152 27.8 132 51.1 160 35.4 159 23.0 141 46.5 160 7.18 117 14.9 62 12.3 152 9.36 142 18.0 116 9.76 153
SILK [79]128.8 9.77 144 15.1 148 11.8 146 12.3 146 18.7 150 11.2 139 10.3 149 16.4 149 8.14 141 25.2 149 22.8 140 45.9 149 30.8 101 27.5 109 45.7 80 30.6 151 20.3 64 40.1 154 7.19 118 16.8 120 10.9 120 8.87 110 17.6 99 7.50 114
H+S_ROB [137]129.0 9.52 141 14.2 141 11.3 141 12.4 148 18.0 143 11.9 144 12.0 156 20.2 157 9.80 150 28.3 156 22.7 137 44.0 137 30.6 68 27.7 128 45.6 65 28.4 94 21.0 88 36.5 84 8.24 151 22.1 153 11.4 137 9.34 141 17.7 102 7.84 133
SLK [47]134.7 11.4 151 15.4 149 14.4 152 12.4 148 18.0 143 12.6 148 10.9 153 17.6 154 8.85 145 27.8 155 25.2 156 46.6 153 30.6 68 28.1 142 43.6 8 29.0 125 21.9 116 37.0 118 8.25 152 22.4 154 11.3 134 9.33 140 18.5 135 7.91 134
GroupFlow [9]135.5 10.1 148 16.9 154 11.3 141 10.4 135 17.8 142 10.0 134 10.8 152 17.5 153 9.21 147 23.6 135 23.9 150 42.5 111 31.9 148 29.3 154 46.2 140 30.1 145 24.5 153 37.8 135 7.55 135 18.4 136 10.6 99 9.52 144 19.8 149 6.89 11
Heeger++ [104]136.1 9.81 145 17.3 155 10.4 131 11.3 140 17.2 133 9.67 130 13.6 158 23.8 159 10.2 153 26.3 152 22.8 140 44.4 140 31.8 147 28.9 153 46.3 142 29.6 136 22.0 121 37.5 132 8.17 150 19.8 148 10.9 120 9.10 128 18.2 127 7.02 26
Learning Flow [11]137.5 8.21 122 14.8 145 9.74 116 9.78 129 17.6 139 8.11 106 9.68 147 15.5 144 7.56 137 25.0 147 24.3 154 45.4 148 31.9 148 28.7 151 47.3 149 29.4 133 22.0 121 37.8 135 7.49 131 18.3 133 10.9 120 10.2 155 20.2 152 8.31 139
2bit-BM-tele [98]138.7 8.61 131 13.5 131 10.5 133 10.0 133 17.5 138 9.73 131 8.26 114 11.5 66 7.40 134 24.4 143 22.7 137 48.1 155 32.5 154 28.6 148 50.2 159 34.7 158 24.3 152 44.9 158 9.35 159 26.2 158 13.8 159 9.25 136 17.3 85 10.2 157
FFV1MT [106]143.5 9.53 142 16.7 153 10.7 136 12.6 152 18.2 148 12.8 150 13.3 157 23.5 158 10.5 156 26.3 152 22.8 140 44.4 140 31.4 143 28.2 144 46.1 134 29.8 140 20.6 80 38.1 141 8.32 154 20.5 150 11.0 125 10.5 156 20.4 153 8.45 141
Adaptive flow [45]146.4 13.2 155 15.9 150 16.2 155 14.2 157 19.9 155 16.4 156 9.02 137 13.1 114 8.03 140 26.0 151 22.8 140 48.6 156 32.4 152 29.4 156 47.9 152 30.1 145 24.6 154 38.0 139 7.55 135 16.9 121 12.2 150 9.85 147 19.5 147 8.97 150
FOLKI [16]147.0 15.0 158 17.4 156 19.4 158 14.3 158 20.9 158 14.4 154 10.7 151 19.2 156 9.99 152 29.8 158 26.8 157 53.1 159 31.3 140 28.6 148 45.8 97 30.0 142 21.7 109 38.9 148 7.85 146 19.4 145 11.6 141 9.85 147 19.2 144 8.80 145
Pyramid LK [2]149.4 16.3 159 16.1 151 21.6 159 16.0 159 20.3 156 18.2 158 16.7 159 15.3 142 14.3 159 35.7 160 36.7 160 56.5 160 32.8 156 31.2 158 45.7 80 29.7 139 22.1 125 38.2 143 8.31 153 23.1 155 11.6 141 11.8 158 25.0 158 8.22 137
PGAM+LK [55]150.4 12.7 153 18.1 157 15.3 154 12.4 148 19.1 152 13.0 151 11.1 154 18.6 155 9.33 149 29.2 157 27.5 158 51.6 158 31.7 145 28.8 152 46.7 145 31.3 154 23.7 146 40.1 154 7.67 142 19.4 145 11.4 137 9.89 149 19.5 147 8.95 148
HCIC-L [99]151.5 18.0 160 18.7 158 23.1 160 12.7 154 17.2 133 17.0 157 10.5 150 14.4 140 8.49 143 25.7 150 23.9 150 44.3 139 33.2 158 29.8 157 49.0 157 31.7 155 26.4 159 39.2 150 7.98 149 18.4 136 12.4 153 12.4 159 25.3 160 8.96 149
Periodicity [78]158.4 14.9 157 20.8 160 18.2 156 20.1 160 22.0 160 21.5 160 17.7 160 26.4 160 16.1 160 29.8 158 34.8 159 49.7 157 35.4 160 34.2 160 48.7 155 37.1 160 25.8 158 47.4 161 8.68 156 23.6 156 12.2 150 13.3 160 25.1 159 11.6 159
AVG_FLOW_ROB [141]160.4 46.4 161 51.9 161 42.8 161 44.1 161 40.8 161 44.3 161 39.2 161 37.3 161 32.5 161 57.2 161 58.7 161 63.7 161 41.6 161 42.4 161 47.5 150 46.3 161 59.4 161 45.4 159 25.5 161 33.4 161 16.2 160 33.6 161 39.6 161 34.2 161
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.