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        
SD
interpolation
error
avg. Mequon
(Hidden texture)
im0   GT   im1
Schefflera
(Hidden texture)
im0   GT   im1
Urban
(Synthetic)
im0   GT   im1
Teddy
(Stereo)
im0   GT   im1
Backyard
(High-speed camera)
im0   GT   im1
Basketball
(High-speed camera)
im0   GT   im1
Dumptruck
(High-speed camera)
im0   GT   im1
Evergreen
(High-speed camera)
im0   GT   im1
rank all disc untext all disc untext all disc untext all disc untext all disc untext all disc untext all disc untext all disc untext
CtxSyn [137]15.7 5.47 2 7.82 2 2.18 113 5.98 1 7.57 2 3.23 22 19.2 91 10.6 2 17.0 95 7.49 1 9.86 1 5.67 2 20.6 3 23.3 3 8.83 4 12.8 2 18.0 2 4.72 1 19.4 3 29.6 3 6.47 2 16.8 2 21.0 2 7.43 16
CyclicGen [154]16.8 5.21 1 5.40 1 4.82 146 6.70 2 7.19 1 6.53 139 8.79 1 10.2 1 7.42 9 10.2 2 10.7 2 11.3 75 18.5 1 21.0 1 7.91 2 7.38 1 10.4 1 4.92 3 13.6 1 20.8 1 6.46 1 10.5 1 13.2 1 6.74 9
SuperSlomo [132]16.8 7.27 3 10.8 3 4.59 144 7.40 10 9.09 5 5.23 101 11.7 2 11.2 3 6.26 3 11.1 3 12.9 3 10.9 67 21.1 4 23.9 4 8.02 3 14.2 6 20.0 6 4.85 2 19.9 4 30.4 4 6.69 6 17.4 4 21.8 4 6.97 10
MDP-Flow2 [68]27.1 8.94 35 13.9 33 1.56 1 7.36 9 9.57 13 2.68 3 15.9 40 20.5 87 16.3 91 13.4 20 18.2 35 8.87 35 24.1 11 27.3 11 12.1 21 17.6 34 25.0 39 6.01 28 22.8 15 34.9 16 7.11 8 20.9 22 26.2 22 7.78 21
CBF [12]27.5 8.02 5 12.4 5 1.75 41 8.18 58 10.3 39 4.84 83 13.2 7 15.5 19 9.17 22 11.5 4 15.0 5 7.94 20 22.8 5 25.8 5 12.2 33 16.2 14 22.9 15 6.24 47 23.4 25 35.8 25 8.06 62 21.2 28 26.7 30 8.24 62
PMMST [114]29.3 8.93 31 14.0 40 1.57 2 7.21 5 9.34 8 2.66 1 14.1 20 16.2 30 11.0 41 15.6 85 21.2 97 14.7 134 24.1 11 27.3 11 12.1 21 16.3 16 23.0 16 5.91 18 22.6 13 34.5 13 7.52 23 19.9 9 25.0 9 8.18 50
FGIK [136]30.8 8.10 7 12.4 5 4.36 143 7.78 20 9.20 7 7.01 142 17.9 81 11.5 4 14.4 77 12.9 14 15.5 8 12.6 119 19.9 2 22.6 2 9.56 7 13.5 4 19.0 4 5.15 5 17.7 2 27.1 2 8.23 66 17.9 6 22.4 6 6.70 5
DeepFlow [86]32.9 8.76 20 13.7 23 1.59 3 8.08 50 10.4 50 4.72 73 13.8 15 18.1 55 7.83 14 12.1 5 15.2 7 8.41 26 28.8 67 32.7 67 12.1 21 17.0 28 24.1 30 5.99 25 21.6 7 33.0 8 7.45 16 23.0 78 28.9 79 7.80 23
SepConv-v1 [127]33.4 8.19 8 12.5 7 5.08 148 7.89 30 9.09 5 8.08 146 20.8 106 12.1 5 18.4 108 12.5 10 14.8 4 11.7 97 23.5 8 26.6 8 9.13 5 14.1 5 19.9 5 5.12 4 21.1 5 32.2 5 8.37 70 17.2 3 21.5 3 6.71 6
DeepFlow2 [108]37.2 8.59 15 13.4 16 1.64 8 8.06 48 10.4 50 4.45 63 13.8 15 18.6 59 8.12 16 12.4 7 16.1 10 10.6 60 28.5 61 32.3 60 12.3 45 16.7 20 23.6 20 5.92 20 23.0 19 35.0 17 7.49 19 22.6 61 28.4 62 8.47 122
SuperFlow [81]38.0 8.83 24 13.8 25 1.68 12 8.19 59 10.3 39 5.04 91 16.0 42 16.6 33 10.9 39 14.9 64 15.1 6 7.67 14 23.0 7 26.0 6 12.5 52 18.0 47 25.5 49 6.55 75 23.6 32 35.8 25 9.80 97 20.7 17 26.0 18 8.12 39
SIOF [67]41.9 8.78 22 13.5 18 1.80 60 8.97 97 11.2 103 4.51 65 16.7 55 23.2 110 11.6 46 13.2 15 17.7 25 9.61 46 23.7 10 26.8 10 11.8 12 17.8 39 25.2 42 5.98 24 23.4 25 35.9 29 7.33 11 22.1 48 27.8 50 8.15 43
CLG-TV [48]42.0 8.34 12 12.9 11 1.98 92 8.74 86 10.8 77 4.75 79 14.0 18 16.0 25 9.23 23 12.4 7 16.1 10 9.95 50 29.7 96 33.7 96 12.0 18 16.8 21 23.9 24 5.46 8 22.2 9 32.9 7 8.02 57 21.8 45 27.4 46 8.33 90
AdaConv-v1 [126]43.5 9.51 71 14.1 47 4.99 147 9.04 101 9.51 12 9.70 149 18.8 90 13.8 8 18.3 107 14.5 51 16.5 15 15.2 139 25.9 23 29.4 23 7.81 1 13.1 3 18.4 3 5.63 11 21.4 6 32.7 6 7.45 16 17.4 4 21.8 4 6.73 8
Aniso. Huber-L1 [22]43.9 8.22 9 12.7 9 1.84 71 9.12 112 11.1 98 5.11 94 13.6 11 16.3 32 7.58 12 12.2 6 16.1 10 9.16 38 29.8 102 33.8 101 12.7 58 16.9 24 23.9 24 5.57 10 23.2 22 35.5 22 7.30 10 21.8 45 27.4 46 8.32 88
CombBMOF [113]44.4 9.74 87 14.3 57 3.85 140 7.82 24 10.2 29 3.81 38 16.2 44 19.1 68 12.8 62 13.8 29 18.5 40 10.2 55 26.5 25 30.0 25 12.2 33 17.8 39 25.2 42 6.09 37 23.1 20 35.2 20 7.64 31 21.3 31 26.7 30 8.21 59
LME [70]45.1 8.97 37 14.0 40 1.62 5 8.07 49 10.5 58 3.69 31 16.9 58 17.7 48 9.29 25 14.5 51 19.6 65 9.68 47 29.2 86 33.1 86 15.3 109 18.1 49 25.7 51 6.15 40 22.7 14 34.7 14 7.37 13 21.0 25 26.4 26 8.20 55
MDP-Flow [26]45.5 8.27 10 12.8 10 1.74 36 7.26 7 9.42 9 3.90 41 17.2 65 16.1 29 15.0 84 13.6 24 18.0 31 10.9 67 28.8 67 32.7 67 15.3 109 17.9 45 25.2 42 7.36 107 23.6 32 36.1 33 12.2 120 20.6 14 25.9 15 8.07 28
NN-field [71]45.6 9.03 43 14.1 47 1.74 36 7.01 4 9.05 4 2.74 8 18.3 85 19.1 68 12.6 60 16.8 112 22.7 118 15.8 141 24.2 13 27.5 13 12.1 21 17.8 39 25.1 40 6.07 33 23.1 20 35.4 21 7.69 36 20.6 14 25.9 15 8.36 104
WLIF-Flow [93]46.5 8.64 16 13.4 16 1.69 17 7.89 30 10.2 29 3.94 43 17.0 61 22.0 97 14.5 79 13.7 27 18.4 38 11.5 87 26.7 28 30.3 28 12.3 45 19.8 111 28.0 112 8.12 130 22.4 11 34.2 11 7.58 28 21.1 27 26.4 26 7.62 18
NNF-Local [87]46.8 8.84 26 13.8 25 1.61 4 7.25 6 9.44 10 2.76 9 14.6 26 19.3 74 14.5 79 16.0 99 21.6 105 15.8 141 24.2 13 27.5 13 12.2 33 18.4 57 26.0 60 6.42 67 24.2 42 37.1 46 9.54 90 20.3 12 25.4 12 8.27 74
p-harmonic [29]46.8 8.89 29 13.9 33 1.68 12 8.86 91 10.9 84 5.20 100 13.4 8 17.5 45 6.45 4 13.7 27 17.9 30 10.0 52 28.9 72 32.8 73 12.8 59 17.6 34 24.9 36 6.53 74 22.9 16 35.0 17 8.88 76 22.5 58 28.3 60 8.10 34
IROF-TV [53]47.1 8.93 31 13.9 33 1.82 67 8.15 55 10.6 63 4.01 47 13.9 17 17.6 46 8.70 19 13.3 17 18.0 31 9.04 36 28.5 61 32.3 60 15.3 109 18.5 63 26.2 65 6.57 76 24.7 56 37.8 58 6.81 7 22.2 52 28.0 55 6.71 6
Second-order prior [8]47.5 8.06 6 12.5 7 1.93 83 8.80 87 11.0 89 4.80 82 12.8 4 16.2 30 7.51 11 12.6 11 16.7 16 6.25 6 28.9 72 32.8 73 12.2 33 18.1 49 25.7 51 6.10 38 23.3 23 35.5 22 9.35 87 22.7 67 28.6 72 8.45 120
ALD-Flow [66]48.2 10.4 106 16.0 105 1.76 44 7.99 40 10.3 39 3.78 36 14.1 20 19.3 74 6.64 5 16.1 101 21.9 109 5.92 4 26.5 25 30.0 25 14.0 75 16.9 24 23.9 24 6.23 46 22.5 12 34.4 12 7.50 21 23.2 87 29.2 92 8.09 31
OAR-Flow [125]48.6 9.14 52 14.0 40 1.71 24 7.90 32 10.1 23 4.04 50 14.3 23 18.8 62 5.59 1 16.6 109 22.6 115 6.23 5 27.7 47 31.4 47 15.3 109 15.9 10 22.4 10 6.89 88 24.2 42 36.4 37 7.80 44 22.9 76 28.8 77 8.15 43
Ad-TV-NDC [36]49.5 9.09 48 13.8 25 2.24 119 9.50 130 11.1 98 6.94 141 14.2 22 15.4 18 6.85 7 14.5 51 18.6 41 9.51 44 27.4 40 31.1 40 12.3 45 18.2 54 25.8 57 6.40 66 22.9 16 34.7 14 7.43 14 20.6 14 25.8 14 8.26 70
IROF++ [58]50.7 8.58 14 13.3 14 1.68 12 7.99 40 10.4 50 3.84 39 17.3 67 18.5 57 12.5 58 12.4 7 16.7 16 9.15 37 28.4 60 32.3 60 15.3 109 19.5 102 27.6 103 6.06 31 23.3 23 35.6 24 8.55 75 23.4 96 29.4 101 7.79 22
DF-Auto [115]51.2 9.30 63 14.4 66 1.99 93 8.37 69 10.6 63 4.99 87 15.5 37 22.3 99 8.88 20 13.3 17 17.6 24 9.86 49 25.7 20 29.1 20 13.9 74 18.1 49 25.7 51 5.96 23 25.2 64 38.6 69 10.8 110 20.7 17 25.9 15 8.09 31
Brox et al. [5]52.4 9.33 65 14.7 71 1.62 5 7.86 28 10.1 23 4.14 54 15.9 40 16.0 25 10.4 34 13.5 23 17.7 25 8.77 33 26.8 29 30.4 29 11.9 13 19.1 88 27.0 89 9.52 146 28.6 121 43.6 119 23.0 152 19.9 9 25.0 9 8.05 27
Modified CLG [34]52.7 7.87 4 12.2 4 1.68 12 8.96 95 10.7 73 5.94 134 16.8 56 16.7 35 15.9 90 13.3 17 16.4 14 12.6 119 27.6 45 31.3 45 11.9 13 18.8 76 26.6 77 6.50 72 22.3 10 34.0 10 7.67 33 22.2 52 27.9 52 8.64 126
NNF-EAC [103]54.7 9.00 40 14.0 40 1.99 93 7.79 22 10.2 29 2.85 12 17.5 73 25.1 127 19.2 112 15.4 78 20.6 85 11.6 92 29.9 106 33.9 106 12.1 21 16.5 17 23.4 17 5.99 25 22.9 16 35.1 19 7.51 22 20.9 22 26.2 22 8.42 117
F-TV-L1 [15]58.3 10.4 106 16.2 107 1.94 85 9.02 99 11.2 103 4.72 73 14.6 26 16.7 35 11.0 41 14.2 38 18.9 50 10.3 56 27.5 42 31.2 43 12.3 45 16.0 11 22.6 11 6.38 63 23.9 39 36.6 42 9.23 83 21.3 31 26.7 30 10.2 141
Local-TV-L1 [65]58.4 8.65 17 13.3 14 1.90 79 9.07 106 11.0 89 5.04 91 13.1 6 15.3 17 8.62 18 12.8 12 17.0 18 7.89 19 30.8 135 35.0 136 15.5 142 18.4 57 26.0 60 6.98 92 23.9 39 36.5 41 7.66 32 21.4 34 26.9 35 8.40 113
JOF [141]58.7 9.04 44 14.1 47 1.81 64 7.55 14 9.71 18 5.06 93 15.1 31 16.7 35 12.0 48 14.4 47 19.4 60 11.4 81 29.2 86 33.2 88 15.4 133 19.8 111 28.0 112 6.36 62 23.4 25 35.8 25 7.68 35 21.4 34 26.8 34 8.30 82
PH-Flow [101]59.2 9.30 63 14.3 57 1.70 21 7.70 18 10.1 23 2.82 11 14.9 30 20.6 90 14.8 82 14.3 42 19.4 60 11.5 87 25.0 17 28.3 17 12.2 33 21.6 146 30.7 147 9.38 145 25.0 62 38.3 64 7.76 41 22.6 61 28.4 62 8.15 43
FMOF [94]60.3 9.22 59 13.9 33 1.96 89 7.58 15 9.87 19 2.87 13 19.5 95 22.4 100 17.7 102 15.3 74 20.6 85 12.5 118 24.5 15 27.7 15 13.7 69 19.3 96 27.3 97 6.05 30 24.6 54 37.7 56 6.64 5 23.4 96 29.4 101 6.98 11
Filter Flow [19]61.3 9.35 67 14.5 67 1.79 56 9.19 114 11.1 98 5.50 120 17.6 77 16.8 39 12.2 52 14.0 34 18.0 31 11.3 75 24.6 16 27.9 16 12.2 33 18.4 57 26.0 60 7.54 113 24.8 57 37.9 60 7.77 42 21.5 37 27.0 38 8.40 113
DMF_ROB [140]61.8 9.66 79 15.1 83 1.75 41 8.12 53 10.3 39 4.86 84 17.2 65 22.7 105 11.7 47 14.5 51 19.3 56 9.60 45 27.3 37 31.0 37 15.3 109 17.9 45 25.4 47 8.26 133 23.5 29 35.9 29 6.49 3 23.2 87 29.2 92 8.32 88
CRTflow [80]63.9 8.75 19 13.6 20 2.04 97 9.27 119 11.5 126 5.28 102 16.2 44 22.5 102 9.27 24 12.8 12 17.0 18 11.5 87 27.0 32 30.6 32 15.3 109 17.6 34 24.9 36 6.06 31 27.8 106 42.7 109 7.62 30 23.4 96 29.4 101 8.16 48
TC/T-Flow [76]64.2 9.42 69 14.6 70 2.39 124 8.67 83 11.2 103 4.00 45 13.6 11 16.0 25 8.03 15 17.5 124 23.5 128 10.8 64 27.3 37 31.0 37 15.3 109 17.4 32 24.6 32 5.89 17 25.8 79 37.8 58 9.59 92 22.8 71 28.7 75 8.13 41
ComplOF-FED-GPU [35]64.4 9.91 96 15.5 96 1.77 51 7.74 19 10.1 23 4.25 59 19.8 99 17.7 48 17.0 95 15.3 74 20.7 88 11.8 98 28.2 59 32.0 59 14.5 81 16.2 14 22.8 14 5.95 22 26.2 85 39.6 86 9.25 84 22.7 67 28.4 62 8.25 67
CNN-flow-warp+ref [117]64.6 8.33 11 13.0 12 2.06 102 8.26 65 10.3 39 5.85 131 18.3 85 22.7 105 11.1 43 13.6 24 16.0 9 11.1 71 29.1 84 33.0 83 15.3 109 15.7 8 22.1 8 6.96 91 28.2 112 43.1 113 7.67 33 21.8 45 27.3 44 8.49 123
Sparse Occlusion [54]64.8 9.75 88 15.2 86 2.05 100 8.71 84 11.2 103 4.19 56 13.5 9 15.9 24 7.80 13 14.6 57 19.7 70 7.51 13 30.4 118 34.5 119 15.3 109 16.1 12 22.7 12 6.27 53 26.9 97 41.1 99 7.45 16 23.2 87 29.2 92 8.12 39
COFM [59]64.9 8.95 36 13.8 25 1.90 79 7.42 11 9.61 14 3.19 21 15.3 34 22.1 98 16.3 91 15.4 78 20.9 91 14.6 130 26.8 29 30.4 29 12.2 33 21.4 142 30.3 142 6.26 52 26.3 86 40.4 89 10.4 106 20.8 19 26.1 19 8.36 104
2DHMM-SAS [92]65.9 8.83 24 13.6 20 1.76 44 8.88 93 11.3 111 4.29 60 17.5 73 20.9 91 12.5 58 14.5 51 19.6 65 11.3 75 30.1 111 34.1 110 15.1 97 17.6 34 24.9 36 5.84 13 25.2 64 38.7 71 8.23 66 23.1 82 29.1 85 8.16 48
CPM-Flow [116]66.0 9.82 92 15.4 94 1.69 17 7.60 16 9.90 20 3.04 18 15.6 39 15.7 22 7.43 10 16.9 114 23.0 123 12.0 104 27.6 45 31.3 45 15.3 109 18.5 63 26.2 65 7.13 101 23.4 25 35.8 25 9.99 99 23.8 114 29.9 116 8.37 107
Horn & Schunck [3]66.7 8.92 30 13.6 20 1.73 29 9.79 140 11.4 116 6.31 137 24.1 127 18.7 60 18.6 110 15.8 95 19.4 60 11.1 71 28.0 52 31.8 53 10.4 9 17.8 39 25.2 42 5.54 9 25.3 69 38.4 67 9.70 94 22.1 48 27.7 49 8.27 74
PMF [73]66.7 9.35 67 14.5 67 1.77 51 7.80 23 10.1 23 2.68 3 24.0 126 28.7 139 22.5 134 15.3 74 20.6 85 11.6 92 25.7 20 29.2 21 12.1 21 19.1 88 27.0 89 5.92 20 27.6 103 42.4 106 9.09 81 23.1 82 29.0 84 6.47 1
LDOF [28]67.0 8.85 27 13.8 25 2.04 97 10.2 148 9.70 17 10.8 153 17.0 61 20.4 86 12.0 48 13.4 20 17.4 22 12.3 113 22.9 6 26.0 6 11.9 13 18.9 80 26.7 80 6.27 53 30.1 136 46.3 138 16.0 133 19.7 7 24.7 7 8.89 132
2D-CLG [1]67.4 8.51 13 13.2 13 1.76 44 8.84 90 10.4 50 5.71 129 19.4 94 15.6 21 15.0 84 14.2 38 16.3 13 14.0 126 31.1 142 35.3 142 20.9 153 16.1 12 22.7 12 6.34 58 27.7 105 42.3 104 8.19 65 21.4 34 26.9 35 8.13 41
TC-Flow [46]67.5 10.9 116 17.1 118 1.71 24 8.86 91 11.6 129 4.00 45 13.0 5 16.0 25 6.24 2 15.6 85 21.1 94 8.58 27 27.9 49 31.7 51 15.1 97 18.7 72 26.4 73 6.72 80 24.6 54 37.6 54 7.95 54 23.4 96 29.4 101 8.28 78
FlowNetS+ft+v [112]67.8 9.02 42 14.1 47 2.07 106 10.0 146 11.0 89 9.60 148 16.3 47 14.4 9 13.5 69 13.8 29 17.7 25 13.3 123 29.7 96 33.8 101 15.3 109 16.8 21 23.8 22 6.25 50 27.8 106 42.6 107 7.83 48 20.4 13 25.5 13 8.24 62
Black & Anandan [4]68.1 9.24 61 14.1 47 1.95 87 9.65 137 11.4 116 5.28 102 28.3 136 24.2 117 20.2 122 14.8 61 18.7 45 10.5 58 27.7 47 31.5 48 9.57 8 19.0 84 27.0 89 6.35 60 24.2 42 36.7 43 8.42 72 21.0 25 26.3 25 6.55 2
PGM-C [120]68.4 9.70 82 15.2 86 1.69 17 7.84 27 10.2 29 3.70 32 21.2 109 17.2 41 12.3 55 17.4 121 23.6 129 8.69 29 28.0 52 31.8 53 15.3 109 16.6 18 23.4 17 6.17 41 26.4 87 40.5 93 8.04 59 24.3 130 30.5 132 8.34 93
EpicFlow [102]68.5 9.69 81 15.2 86 1.67 11 7.90 32 10.2 29 4.37 62 16.0 42 14.5 10 9.75 27 19.1 135 25.8 138 12.3 113 27.9 49 31.6 49 15.3 109 16.9 24 23.9 24 6.21 45 24.9 59 38.0 61 10.3 104 24.6 136 30.9 137 8.30 82
OFLAF [77]68.6 9.70 82 15.0 80 1.69 17 7.94 37 10.4 50 2.73 7 14.3 23 15.0 14 10.2 30 13.8 29 18.6 41 8.40 25 30.0 108 34.0 108 15.4 133 17.0 28 23.9 24 6.73 81 30.1 136 46.1 136 13.9 128 23.4 96 29.3 97 9.45 136
Fusion [6]68.6 8.82 23 13.8 25 2.62 128 7.96 38 10.1 23 4.47 64 16.5 50 13.6 7 17.3 101 14.0 34 18.1 34 9.97 51 29.8 102 33.8 101 12.8 59 19.4 98 27.4 98 10.1 149 26.4 87 40.4 89 8.14 63 21.7 41 27.2 42 10.1 140
MLDP_OF [89]68.7 9.06 45 14.1 47 1.83 70 8.81 88 11.3 111 4.78 81 14.0 18 17.6 46 8.56 17 15.5 82 20.3 80 15.8 141 29.7 96 33.7 96 13.6 65 19.1 88 27.0 89 5.86 15 23.8 35 36.3 35 8.15 64 23.2 87 29.1 85 8.25 67
ProFlow_ROB [147]68.8 9.82 92 15.5 96 1.71 24 8.15 55 10.6 63 3.63 28 15.4 36 12.7 6 8.98 21 19.8 139 26.9 140 7.99 21 29.0 82 33.0 83 15.3 109 15.6 7 22.0 7 5.16 6 26.4 87 40.3 88 7.90 52 24.4 133 30.5 132 11.5 145
EAI-Flow [152]69.9 11.1 119 15.4 94 6.27 152 8.02 44 10.2 29 4.70 72 16.9 58 20.1 84 15.0 84 14.8 61 19.8 72 4.86 1 29.2 86 33.1 86 14.8 88 16.6 18 23.5 19 5.99 25 23.8 35 36.4 37 20.9 148 22.6 61 28.4 62 11.1 143
TV-L1-MCT [64]70.1 9.18 55 14.2 55 1.78 53 8.53 76 11.1 98 3.70 32 17.7 79 23.3 113 13.6 70 14.4 47 19.5 64 11.6 92 30.5 124 34.6 122 13.8 71 18.1 49 25.7 51 6.02 29 25.8 79 39.5 83 15.0 131 21.7 41 27.3 44 7.99 25
AGIF+OF [85]70.2 9.07 46 14.0 40 1.78 53 7.93 36 10.3 39 3.78 36 14.4 25 17.8 50 12.4 56 14.9 64 20.2 77 11.4 81 28.9 72 32.8 73 15.3 109 20.0 114 28.3 114 6.98 92 25.5 72 39.0 74 7.74 39 23.9 121 30.1 125 8.28 78
LFNet_ROB [150]70.3 11.6 126 18.2 129 2.58 127 8.05 46 10.4 50 4.02 49 17.8 80 26.6 131 11.5 45 13.2 15 17.8 28 7.21 11 26.9 31 30.5 31 14.8 88 21.1 136 29.9 137 7.31 105 23.5 29 35.9 29 11.3 113 22.4 56 28.2 58 8.11 37
Bartels [41]71.5 12.7 138 20.1 139 2.13 112 8.52 75 11.0 89 4.96 86 13.5 9 14.5 10 10.2 30 14.4 47 18.9 50 10.8 64 23.5 8 26.6 8 12.9 62 19.0 84 26.9 86 6.94 90 24.5 49 37.5 52 19.7 145 23.4 96 29.4 101 8.31 85
S2F-IF [123]71.9 10.3 103 16.3 109 1.79 56 7.83 26 10.2 29 2.90 15 17.0 61 20.0 82 13.9 74 16.1 101 21.6 105 6.69 7 29.2 86 33.2 88 15.3 109 16.8 21 23.7 21 6.34 58 24.9 59 38.2 62 10.7 108 23.9 121 30.0 123 8.35 101
OFH [38]73.4 9.54 73 15.0 80 1.74 36 8.49 74 10.6 63 5.13 95 18.1 83 24.9 125 10.4 34 17.4 121 23.7 131 5.72 3 28.7 65 32.5 63 14.6 84 17.6 34 24.8 34 5.85 14 26.0 83 39.2 78 10.2 102 22.7 67 28.5 70 14.1 150
BlockOverlap [61]73.5 9.09 48 14.3 57 2.04 97 8.96 95 10.9 84 5.37 113 18.1 83 15.5 19 18.0 105 14.2 38 17.2 20 14.0 126 28.9 72 32.8 73 13.8 71 18.8 76 26.7 80 7.92 124 24.8 57 37.2 47 21.0 149 20.0 11 25.1 11 8.38 108
nLayers [57]73.8 9.15 53 14.3 57 1.76 44 7.42 11 9.62 15 3.57 26 27.8 134 29.9 142 25.8 144 15.9 97 21.5 103 11.9 99 30.2 112 34.3 113 14.7 87 20.3 121 28.8 122 6.45 69 23.5 29 36.0 32 7.87 50 21.6 38 27.1 39 8.10 34
TCOF [69]74.1 9.34 66 14.3 57 1.89 75 9.50 130 11.7 136 5.42 114 16.2 44 21.7 95 10.3 32 13.8 29 18.6 41 9.45 43 30.4 118 34.6 122 13.6 65 18.2 54 25.7 51 6.20 44 28.5 118 43.5 118 7.54 24 22.9 76 28.8 77 8.18 50
HAST [109]74.5 8.87 28 13.8 25 1.76 44 7.34 8 9.50 11 2.70 5 28.8 138 28.6 138 24.0 139 14.9 64 20.2 77 7.68 15 28.9 72 32.8 73 12.1 21 21.3 141 30.2 141 7.57 115 28.6 121 43.9 122 7.55 26 22.8 71 28.7 75 8.43 119
DPOF [18]75.3 11.0 117 17.4 122 3.88 141 7.78 20 10.2 29 3.01 16 18.7 88 18.1 55 18.4 108 16.5 106 22.4 112 14.6 130 28.8 67 32.7 67 12.1 21 18.9 80 26.7 80 6.18 43 25.2 64 38.4 67 7.59 29 23.6 109 29.6 109 8.07 28
Layers++ [37]76.0 8.93 31 14.0 40 1.76 44 6.74 3 8.61 3 2.71 6 18.3 85 25.8 129 19.3 113 15.3 74 20.8 89 11.3 75 33.1 150 37.6 150 19.8 150 21.6 146 30.6 146 8.73 139 24.4 47 37.4 50 7.81 46 21.6 38 27.1 39 8.09 31
FlowFields [110]76.2 9.98 97 15.7 99 2.08 108 7.96 38 10.4 50 3.62 27 23.1 119 23.2 110 20.3 124 16.0 99 21.5 103 7.08 10 27.0 32 30.6 32 14.2 77 19.2 94 27.1 94 6.08 36 24.4 47 37.4 50 10.2 102 23.2 87 29.2 92 8.35 101
Classic++ [32]76.9 9.48 70 14.9 75 1.80 60 8.59 77 11.0 89 4.61 68 13.7 14 15.0 14 9.57 26 14.4 47 19.0 52 8.76 32 29.9 106 33.9 106 13.6 65 20.2 119 28.7 120 6.87 87 27.4 101 42.0 101 9.63 93 23.8 114 29.9 116 8.34 93
SRR-TVOF-NL [91]77.8 9.65 77 14.8 72 1.82 67 8.21 63 10.6 63 4.76 80 22.7 117 28.1 136 21.9 129 15.6 85 20.9 91 9.18 39 28.9 72 32.8 73 15.3 109 20.7 130 29.3 130 5.91 18 24.5 49 37.6 54 6.56 4 22.5 58 28.2 58 8.34 93
NL-TV-NCC [25]78.2 9.19 57 14.3 57 2.18 113 9.02 99 11.6 129 4.13 53 14.8 28 16.7 35 10.9 39 20.8 140 28.1 142 8.19 23 26.5 25 30.0 25 13.1 63 18.9 80 26.7 80 6.43 68 26.6 92 40.4 89 15.1 132 23.7 113 29.7 113 8.29 81
HBM-GC [105]78.5 9.25 62 14.5 67 1.81 64 9.08 108 11.9 145 3.75 35 17.3 67 18.7 60 17.9 103 14.3 42 19.2 54 8.85 34 30.0 108 34.0 108 15.5 142 21.5 143 30.4 143 8.27 134 27.4 101 42.1 103 7.15 9 20.8 19 26.1 19 7.05 13
Nguyen [33]79.0 9.83 94 15.2 86 1.73 29 9.59 135 11.0 89 5.65 128 15.3 34 20.5 87 10.3 32 14.6 57 18.8 46 12.1 106 28.8 67 32.7 67 12.2 33 19.4 98 27.4 98 8.01 128 29.7 131 45.5 131 8.29 69 21.2 28 26.6 29 8.34 93
Complementary OF [21]79.2 11.4 125 18.1 128 1.70 21 9.23 117 12.1 146 4.19 56 31.6 143 19.0 66 23.6 136 19.5 138 26.5 139 6.72 8 28.1 57 31.8 53 14.6 84 17.3 31 24.4 31 6.38 63 26.1 84 39.0 74 8.92 78 22.3 54 27.9 52 7.57 17
Efficient-NL [60]80.0 8.71 18 13.5 18 1.68 12 8.66 81 11.2 103 3.65 29 22.5 113 20.0 82 19.9 118 14.3 42 19.3 56 11.0 69 30.5 124 34.7 129 15.0 92 20.1 115 28.4 116 6.27 53 28.5 118 43.7 120 8.92 78 23.8 114 29.9 116 6.66 4
FESL [72]80.5 9.09 48 13.9 33 1.74 36 7.90 32 10.3 39 3.35 24 16.5 50 21.9 96 12.0 48 15.1 71 20.3 80 11.4 81 30.8 135 35.0 136 15.4 133 19.6 105 27.8 108 6.48 70 27.8 106 42.6 107 7.75 40 23.9 121 30.0 123 8.39 109
ProbFlowFields [128]80.8 10.1 98 16.0 105 1.78 53 8.04 45 10.5 58 3.08 20 25.8 133 28.8 140 24.3 140 14.5 51 19.6 65 11.4 81 27.2 35 30.9 36 15.3 109 17.4 32 24.6 32 8.78 140 27.3 100 42.0 101 18.8 143 22.4 56 28.1 56 8.39 109
AggregFlow [97]81.1 12.9 140 20.3 140 1.75 41 8.34 67 10.8 77 4.14 54 20.0 100 24.4 120 19.5 117 16.5 106 22.3 111 12.2 110 25.2 18 28.6 18 12.2 33 16.9 24 23.9 24 6.60 77 29.0 128 43.9 122 16.7 136 23.0 78 28.9 79 8.03 26
FlowFields+ [130]82.3 9.67 80 15.2 86 3.33 139 7.86 28 10.3 39 3.02 17 23.3 121 24.6 122 20.8 125 17.0 116 23.0 123 6.91 9 27.3 37 31.0 37 15.4 133 19.0 84 26.9 86 6.24 47 25.3 69 38.8 72 13.1 124 23.2 87 29.1 85 8.39 109
RNLOD-Flow [121]82.5 8.93 31 13.8 25 1.65 9 8.48 73 11.0 89 4.06 51 16.3 47 23.2 110 12.8 62 14.1 36 19.1 53 11.1 71 29.7 96 33.7 96 15.6 144 20.3 121 28.7 120 8.92 143 25.7 75 39.4 81 16.4 135 24.2 127 30.4 129 8.20 55
PWC-Net_ROB [148]82.6 11.7 128 18.2 129 2.05 100 8.35 68 10.9 84 3.91 42 13.6 11 17.4 44 6.79 6 18.8 131 25.6 137 8.72 31 30.5 124 34.6 122 15.1 97 19.4 98 27.4 98 5.68 12 23.6 32 36.2 34 7.95 54 24.2 127 30.4 129 11.5 145
StereoOF-V1MT [119]82.7 11.1 119 17.3 120 1.73 29 8.61 78 10.6 63 5.28 102 23.4 123 17.3 42 17.1 99 16.6 109 19.9 74 12.3 113 27.4 40 31.1 40 15.0 92 17.0 28 23.8 22 6.80 85 30.2 138 46.2 137 12.3 121 21.6 38 26.9 35 9.58 137
FlowNet2 [122]82.8 15.6 149 23.6 150 1.96 89 9.34 121 12.1 146 4.72 73 17.3 67 19.2 72 13.0 65 17.1 119 23.1 125 10.1 53 28.0 52 31.8 53 12.3 45 18.6 66 26.3 69 6.35 60 26.7 93 40.8 96 8.04 59 21.7 41 27.2 42 8.30 82
Sparse-NonSparse [56]83.5 9.18 55 14.3 57 1.73 29 8.14 54 10.6 63 3.31 23 16.6 53 22.9 108 13.8 73 14.8 61 19.8 72 11.3 75 30.5 124 34.6 122 15.0 92 20.1 115 28.5 118 7.48 111 28.5 118 43.7 120 9.49 88 23.5 106 29.5 106 8.24 62
TI-DOFE [24]83.8 9.80 90 15.2 86 2.80 132 9.94 144 11.4 116 5.62 125 15.5 37 15.7 22 10.5 37 17.0 116 21.7 107 10.6 60 27.1 34 30.8 34 12.1 21 20.9 134 29.6 135 6.99 94 24.0 41 36.3 35 8.92 78 24.3 130 28.1 56 12.5 148
TVL1_ROB [139]83.9 10.5 109 16.3 109 1.88 74 9.86 142 11.6 129 5.57 121 20.0 100 19.0 66 14.6 81 15.6 85 20.3 80 11.9 99 26.4 24 29.9 24 12.1 21 19.0 84 26.8 84 7.12 100 28.2 112 42.9 110 12.9 123 20.9 22 26.2 22 8.34 93
ACK-Prior [27]84.4 9.81 91 15.1 83 2.07 106 8.01 43 10.4 50 3.86 40 25.1 130 19.1 68 22.0 131 15.1 71 20.1 76 10.1 53 30.4 118 34.4 116 15.4 133 19.1 88 26.9 86 7.57 115 25.8 79 39.3 79 19.5 144 22.3 54 27.9 52 7.73 19
3DFlow [135]84.6 9.65 77 14.9 75 1.89 75 7.82 24 10.0 22 4.94 85 16.9 58 19.9 79 13.7 71 16.5 106 22.4 112 15.8 141 29.3 91 33.2 88 12.5 52 18.4 57 25.9 59 8.56 137 27.2 98 41.4 100 10.1 101 23.4 96 29.3 97 8.87 130
LSM [39]84.7 9.10 51 14.2 55 1.73 29 8.33 66 10.9 84 3.40 25 16.6 53 22.7 105 12.2 52 15.0 70 20.3 80 11.0 69 30.5 124 34.7 129 15.1 97 20.7 130 29.4 131 6.17 41 28.1 111 43.0 112 11.5 115 23.8 114 29.9 116 8.27 74
Occlusion-TV-L1 [63]84.7 10.1 98 15.9 103 2.43 125 9.36 122 11.8 142 5.01 90 12.7 3 14.7 12 7.22 8 17.0 116 22.7 118 11.4 81 28.6 63 32.5 63 12.0 18 18.7 72 26.5 76 7.48 111 25.2 64 37.7 56 10.0 100 24.3 130 30.3 128 9.33 134
Classic+CPF [83]85.0 9.07 46 14.0 40 1.80 60 8.09 51 10.5 58 3.71 34 17.0 61 21.5 93 12.9 64 13.9 33 18.8 46 11.4 81 30.7 133 34.9 134 15.4 133 21.2 138 30.0 139 7.73 122 28.2 112 43.2 115 7.80 44 24.7 139 31.0 139 7.89 24
ResPWCR_ROB [145]85.8 11.2 124 17.7 124 1.95 87 8.98 98 11.7 136 4.11 52 15.2 33 17.3 42 10.0 28 23.1 146 31.2 148 10.5 58 30.5 124 34.6 122 14.2 77 20.3 121 28.8 122 5.27 7 23.8 35 36.4 37 7.54 24 25.4 145 31.9 149 7.76 20
TriFlow [95]86.2 13.1 141 20.8 141 2.06 102 9.53 133 12.2 148 5.29 108 16.5 50 18.5 57 10.1 29 17.2 120 22.8 120 7.74 16 27.9 49 31.6 49 15.1 97 19.4 98 27.4 98 6.07 33 24.5 49 37.2 47 10.9 111 23.8 114 29.8 115 8.15 43
CostFilter [40]86.4 10.8 115 17.0 117 1.80 60 7.90 32 10.3 39 2.66 1 24.6 129 27.7 135 21.9 129 18.7 130 25.4 135 13.7 125 27.5 42 31.1 40 12.6 55 18.2 54 25.8 57 5.87 16 28.9 126 44.2 127 9.34 86 24.4 133 30.7 135 8.20 55
FFV1MT [106]86.4 11.6 126 17.7 124 2.19 115 9.20 115 10.9 84 5.96 135 22.6 114 30.3 143 16.3 91 15.5 82 18.8 46 12.4 116 27.5 42 31.2 43 11.6 11 18.6 66 25.7 51 7.42 110 27.2 98 40.7 94 8.88 76 21.2 28 26.4 26 9.73 138
TF+OM [100]86.5 11.8 130 18.7 132 3.19 136 8.23 64 10.8 77 4.54 66 15.1 31 19.7 77 10.4 34 16.3 104 21.9 109 7.87 18 28.9 72 32.8 73 19.1 149 18.6 66 26.3 69 6.68 79 26.5 91 40.7 94 11.5 115 23.8 114 29.9 116 8.23 61
Ramp [62]87.2 9.22 59 14.3 57 1.73 29 8.19 59 10.7 73 4.24 58 21.9 112 28.8 140 21.1 128 14.2 38 19.2 54 11.6 92 30.6 131 34.8 132 14.8 88 20.4 125 29.0 128 7.40 108 28.0 110 42.9 110 7.57 27 23.0 78 28.9 79 8.28 78
H+S_ROB [138]87.2 9.20 58 14.1 47 1.72 27 8.66 81 10.2 29 5.63 126 39.8 153 41.0 153 32.8 154 14.3 42 17.4 22 9.30 41 29.6 95 33.6 95 12.2 33 19.5 102 27.6 103 7.00 95 31.4 142 48.0 142 10.3 104 23.0 78 28.9 79 8.34 93
AugFNG_ROB [144]88.0 12.1 134 19.0 134 1.94 85 8.44 72 10.6 63 5.42 114 17.3 67 23.9 116 10.8 38 26.2 151 34.7 150 11.5 87 31.6 145 35.9 145 15.3 109 18.7 72 26.4 73 6.38 63 24.9 59 38.2 62 7.91 53 19.8 8 24.8 8 8.36 104
EPMNet [133]88.7 16.1 151 24.7 151 2.22 116 9.04 101 11.7 136 4.55 67 17.3 67 19.2 72 13.0 65 26.7 153 36.3 154 12.1 106 28.0 52 31.8 53 12.3 45 18.4 57 26.0 60 6.25 50 26.7 93 40.8 96 8.04 59 22.6 61 28.4 62 8.35 101
IAOF2 [51]88.7 10.7 114 16.6 113 2.36 122 9.40 123 11.6 129 5.33 109 17.4 72 18.0 52 12.4 56 14.1 36 18.2 35 9.32 42 30.3 116 34.4 116 14.0 75 20.5 129 29.1 129 8.20 131 25.2 64 38.3 64 8.49 73 23.1 82 29.1 85 8.24 62
SVFilterOh [111]89.1 10.5 109 16.4 111 1.97 91 7.65 17 9.98 21 3.05 19 28.0 135 30.4 144 25.4 142 15.6 85 21.2 97 14.7 134 28.9 72 32.7 67 15.4 133 20.1 115 28.4 116 6.61 78 25.8 79 39.5 83 7.84 49 22.5 58 28.3 60 8.49 123
Heeger++ [104]89.5 14.5 146 21.7 145 4.63 145 9.50 130 11.0 89 5.73 130 25.4 132 23.5 115 14.4 77 15.5 82 18.8 46 12.4 116 28.7 65 32.5 63 15.2 105 15.8 9 22.2 9 6.73 81 27.6 103 39.8 87 9.28 85 22.1 48 27.6 48 8.34 93
Classic+NL [31]90.0 8.97 37 13.9 33 1.79 56 8.11 52 10.5 58 4.01 47 20.8 106 28.3 137 19.9 118 14.3 42 19.3 56 11.5 87 30.7 133 34.8 132 14.6 84 20.3 121 28.8 122 7.40 108 28.3 115 43.4 116 11.9 119 23.5 106 29.6 109 8.25 67
Dynamic MRF [7]90.1 10.1 98 15.9 103 1.81 64 8.42 71 10.8 77 4.73 77 19.5 95 19.1 68 12.2 52 15.6 85 19.3 56 12.8 122 27.2 35 30.8 34 15.2 105 18.6 66 26.3 69 7.28 104 28.8 125 44.1 126 12.4 122 24.6 136 30.7 135 9.73 138
Adaptive [20]90.2 11.0 117 17.3 120 1.89 75 9.41 125 11.6 129 5.19 99 14.8 28 17.1 40 11.1 43 15.7 92 21.1 94 12.1 106 31.1 142 35.3 142 12.0 18 18.8 76 26.6 77 8.00 127 27.8 106 42.3 104 8.01 56 22.6 61 28.4 62 8.63 125
TV-L1-improved [17]90.3 9.53 72 14.9 75 1.99 93 9.46 127 11.7 136 5.17 97 22.6 114 14.8 13 20.1 121 13.4 20 17.8 28 8.05 22 30.2 112 34.3 113 11.9 13 19.6 105 27.7 105 8.09 129 29.9 134 45.8 134 9.73 95 23.4 96 29.3 97 8.42 117
IAOF [50]90.4 11.1 119 16.6 113 5.32 149 10.6 150 12.3 149 5.87 132 23.3 121 24.2 117 19.4 114 15.4 78 19.7 70 12.0 104 28.9 72 32.8 73 12.1 21 18.8 76 26.6 77 7.26 103 25.6 74 39.1 77 7.35 12 22.1 48 27.8 50 8.26 70
Steered-L1 [118]90.7 8.76 20 13.7 23 1.82 67 8.00 42 10.3 39 4.72 73 31.9 144 33.2 150 29.2 149 17.4 121 22.8 120 14.1 128 29.5 94 33.5 94 14.2 77 19.7 109 27.9 110 6.28 57 26.4 87 40.4 89 18.7 142 23.8 114 29.9 116 7.04 12
ROF-ND [107]90.8 9.00 40 13.9 33 1.62 5 9.53 133 10.8 77 10.7 152 16.4 49 22.9 108 12.7 61 18.3 128 24.1 132 11.9 99 29.4 93 33.3 93 15.2 105 18.6 66 26.2 65 7.60 118 24.5 49 37.3 49 13.2 126 24.4 133 30.5 132 9.33 134
TriangleFlow [30]91.2 9.59 75 14.8 72 2.06 102 9.07 106 11.4 116 5.47 118 19.2 91 20.2 85 13.9 74 13.6 24 18.2 35 8.31 24 30.0 108 34.1 110 9.31 6 17.8 39 25.2 42 7.56 114 30.8 140 47.2 140 13.9 128 25.5 148 31.9 149 11.3 144
FF++_ROB [146]91.5 10.6 111 16.7 115 1.70 21 8.19 59 10.6 63 3.67 30 21.6 111 22.6 103 17.0 95 19.0 133 25.5 136 14.9 138 28.9 72 32.8 73 15.4 133 19.3 96 27.4 98 6.24 47 25.7 75 39.5 83 11.8 118 23.5 106 29.5 106 8.27 74
FOLKI [16]91.6 10.6 111 16.5 112 2.43 125 9.94 144 11.2 103 6.70 140 19.6 97 21.6 94 19.9 118 18.3 128 19.4 60 17.3 146 28.0 52 31.7 51 13.6 65 19.1 88 27.1 94 10.9 151 24.2 42 36.9 44 17.3 139 21.3 31 26.7 30 8.10 34
LocallyOriented [52]92.8 10.1 98 15.7 99 1.79 56 9.46 127 11.6 129 5.28 102 23.1 119 24.2 117 20.9 127 19.3 137 23.2 126 7.35 12 30.4 118 34.6 122 12.6 55 18.9 80 26.8 84 6.27 53 25.7 75 38.6 69 7.89 51 23.6 109 29.6 109 8.19 53
SILK [79]93.7 9.72 86 15.1 83 2.69 129 10.2 148 11.4 116 7.82 145 39.2 152 32.9 149 28.5 148 14.6 57 18.4 38 9.73 48 29.0 82 32.9 82 10.4 9 21.2 138 30.0 139 7.00 95 24.5 49 37.5 52 8.03 58 23.1 82 28.9 79 8.31 85
S2D-Matching [84]95.2 9.57 74 14.9 75 1.76 44 8.37 69 10.8 77 4.36 61 20.1 103 24.9 125 18.2 106 15.7 92 21.3 101 15.7 140 28.8 67 32.7 67 14.5 81 21.5 143 30.4 143 11.0 152 25.7 75 39.3 79 11.5 115 23.1 82 29.1 85 8.75 129
BriefMatch [124]95.7 9.89 95 15.5 96 2.11 110 8.05 46 10.2 29 5.90 133 23.5 124 18.0 52 22.7 135 18.2 127 18.6 41 18.7 149 28.1 57 31.9 58 13.8 71 19.5 102 27.7 105 7.05 98 26.7 93 39.4 81 21.6 150 23.4 96 29.3 97 14.3 152
GraphCuts [14]96.1 11.7 128 17.8 126 2.02 96 8.15 55 10.5 58 4.65 70 25.3 131 15.2 16 19.4 114 14.9 64 19.6 65 11.9 99 29.8 102 33.8 101 17.8 147 19.6 105 27.8 108 6.50 72 28.6 121 43.9 122 11.1 112 24.0 125 30.2 126 8.15 43
RFlow [90]96.7 9.71 84 15.2 86 1.91 82 9.06 104 11.2 103 5.42 114 22.8 118 22.6 103 17.9 103 15.8 95 21.2 97 12.7 121 29.2 86 33.2 88 11.9 13 19.2 94 27.2 96 7.63 119 28.9 126 44.4 128 7.73 38 23.4 96 29.5 106 8.46 121
ContinualFlow_ROB [153]96.9 12.2 135 19.3 135 2.23 117 8.71 84 11.3 111 4.73 77 17.5 73 19.9 79 13.3 68 22.6 144 30.7 147 8.65 28 32.4 148 36.8 149 15.3 109 18.5 63 26.2 65 6.12 39 28.6 121 43.9 122 8.50 74 22.7 67 28.4 62 8.39 109
ComponentFusion [96]97.3 12.3 136 19.5 136 1.66 10 8.65 80 11.4 116 2.88 14 19.7 98 21.0 92 15.1 87 15.4 78 20.9 91 14.4 129 29.7 96 33.7 96 14.5 81 18.6 66 26.3 69 7.67 120 31.9 143 49.0 145 20.5 146 24.2 127 30.4 129 8.18 50
Learning Flow [11]97.8 8.99 39 14.1 47 1.85 73 9.10 110 11.3 111 4.99 87 40.2 154 42.5 154 31.6 153 14.9 64 17.2 20 12.2 110 30.8 135 35.0 136 15.1 97 18.7 72 26.4 73 7.58 117 25.1 63 38.3 64 11.4 114 25.5 148 31.7 145 8.24 62
FC-2Layers-FF [74]98.5 9.71 84 14.9 75 2.11 110 7.51 13 9.66 16 4.67 71 20.5 104 25.1 127 20.2 122 15.6 85 21.1 94 11.9 99 30.5 124 34.6 122 15.3 109 20.8 132 29.4 131 7.31 105 29.7 131 45.6 133 9.76 96 23.6 109 29.7 113 8.22 60
SLK [47]98.7 9.63 76 15.0 80 1.90 79 9.14 113 10.3 39 5.63 126 34.7 146 19.7 77 22.4 133 18.9 132 24.2 133 20.4 152 29.8 102 33.8 101 12.2 33 18.1 49 25.5 49 6.93 89 31.9 143 48.8 143 9.12 82 22.8 71 28.5 70 14.2 151
Adaptive flow [45]98.8 10.3 103 14.8 72 2.37 123 9.87 143 11.5 126 5.57 121 18.0 82 17.9 51 17.1 99 16.4 105 20.0 75 14.8 136 32.3 147 36.7 147 16.6 145 21.1 136 29.8 136 8.41 135 23.8 35 36.4 37 13.1 124 21.7 41 27.1 39 7.17 14
Shiralkar [42]98.9 12.0 133 18.8 133 1.72 27 9.11 111 11.1 98 5.14 96 21.2 109 16.6 33 13.7 71 19.2 136 24.3 134 10.6 60 29.7 96 33.7 96 12.8 59 18.0 47 25.4 47 7.19 102 29.4 129 44.9 129 10.4 106 25.1 144 31.5 144 9.03 133
EPPM w/o HM [88]100.0 10.4 106 16.2 107 2.97 134 8.62 79 11.3 111 2.76 9 29.0 139 27.4 133 22.2 132 16.8 112 22.6 115 10.8 64 25.8 22 29.2 21 12.1 21 20.2 119 28.6 119 6.49 71 29.8 133 45.8 134 18.0 140 24.0 125 30.2 126 8.72 128
UnFlow [129]102.2 13.4 142 21.2 144 2.71 131 8.81 88 10.7 73 6.35 138 18.7 88 18.9 63 14.8 82 14.6 57 19.6 65 7.77 17 31.8 146 36.1 146 15.0 92 22.2 149 31.4 149 7.79 123 24.2 42 37.0 45 7.49 19 28.1 153 33.7 154 11.6 147
Correlation Flow [75]102.4 9.75 88 15.3 93 1.84 71 9.28 120 11.6 129 5.17 97 17.5 73 18.9 63 15.2 88 16.1 101 21.7 107 11.3 75 30.2 112 34.3 113 12.5 52 21.2 138 29.9 137 8.24 132 31.3 141 47.8 141 9.82 98 24.9 142 31.3 143 6.61 3
LiteFlowNet [143]105.3 12.6 137 19.7 138 2.06 102 8.19 59 10.7 73 3.96 44 20.8 106 26.3 130 15.4 89 26.2 151 35.0 152 12.2 110 30.9 141 35.0 136 14.9 91 20.4 125 28.8 122 6.74 84 28.3 115 43.1 113 7.70 37 22.8 71 28.6 72 8.87 130
HBpMotionGpu [43]105.6 12.7 138 19.5 136 2.69 129 9.65 137 11.7 136 5.48 119 20.0 100 23.3 113 17.0 95 17.6 125 23.4 127 10.6 60 30.8 135 35.0 136 25.1 154 20.4 125 28.9 127 7.95 125 22.0 8 33.7 9 7.44 15 23.2 87 29.1 85 8.40 113
PGAM+LK [55]106.4 11.9 132 18.0 127 7.26 153 9.48 129 10.8 77 7.62 143 31.5 142 39.9 152 31.4 152 19.0 133 23.6 129 16.3 145 29.1 84 33.0 83 12.6 55 18.4 57 26.0 60 6.80 85 25.5 72 39.0 74 14.8 130 22.6 61 28.4 62 8.41 116
2bit-BM-tele [98]107.7 11.1 119 17.2 119 2.34 120 9.40 123 11.7 136 5.36 111 28.5 137 37.1 151 31.0 151 15.7 92 20.8 89 9.18 39 28.6 63 32.5 63 15.0 92 22.0 148 31.1 148 9.53 147 39.1 154 59.9 154 26.9 154 20.8 19 26.1 19 8.11 37
Rannacher [23]108.2 11.1 119 17.5 123 1.89 75 9.59 135 11.8 142 5.28 102 24.3 128 18.0 52 20.8 125 15.9 97 21.2 97 11.6 92 30.4 118 34.5 119 12.3 45 19.7 109 27.9 110 7.98 126 29.6 130 45.3 130 9.57 91 24.7 139 31.0 139 8.19 53
OFRF [134]108.6 13.6 143 21.1 143 2.23 117 9.25 118 11.4 116 5.60 123 19.2 91 19.5 76 14.1 76 16.9 114 22.8 120 14.6 130 30.8 135 34.9 134 14.4 80 19.6 105 27.7 105 6.07 33 28.3 115 43.4 116 7.78 43 24.7 139 31.1 141 8.34 93
StereoFlow [44]109.2 14.9 147 22.2 147 3.28 137 10.0 146 12.7 152 4.99 87 16.8 56 18.9 63 12.1 51 15.2 73 20.4 84 10.4 57 33.4 151 37.9 151 20.8 151 23.8 151 33.5 151 8.41 135 25.3 69 38.8 72 7.81 46 23.6 109 29.6 109 8.67 127
SimpleFlow [49]111.8 9.15 53 14.3 57 1.73 29 9.05 103 11.4 116 5.35 110 36.0 150 32.6 148 29.4 150 14.9 64 20.2 77 11.2 74 30.6 131 34.7 129 15.1 97 22.6 150 32.0 150 9.11 144 34.7 148 53.2 149 13.8 127 23.9 121 29.9 116 8.33 90
Aniso-Texture [82]112.6 10.6 111 16.7 115 1.74 36 9.83 141 12.4 150 5.36 111 17.6 77 19.9 79 13.1 67 22.9 145 27.8 141 19.7 151 30.3 116 34.4 116 15.4 133 20.8 132 29.4 131 8.86 142 26.8 96 41.0 98 8.38 71 24.6 136 30.9 137 8.26 70
SPSA-learn [13]112.9 15.1 148 22.9 148 1.93 83 9.08 108 11.0 89 5.42 114 33.0 145 24.8 123 23.8 138 17.6 125 22.4 112 12.1 106 29.3 91 33.2 88 15.1 97 17.8 39 25.1 40 6.73 81 37.7 150 57.7 151 25.5 153 25.4 145 31.8 146 8.33 90
SegOF [10]113.0 11.8 130 18.2 129 5.53 150 8.88 93 11.4 116 4.62 69 31.1 141 20.5 87 23.7 137 25.8 149 34.8 151 18.2 148 30.2 112 34.2 112 15.3 109 19.1 88 27.0 89 7.08 99 32.5 146 49.7 146 16.8 137 22.8 71 28.6 72 8.08 30
HCIC-L [99]118.6 14.3 145 20.9 142 2.86 133 11.2 151 13.3 153 7.62 143 23.9 125 31.6 146 25.6 143 21.0 141 28.2 143 14.8 136 25.6 19 29.0 19 12.2 33 24.0 152 33.9 152 10.5 150 30.5 139 46.8 139 18.5 141 23.3 94 29.2 92 7.37 15
IIOF-NLDP [131]120.3 10.2 102 15.8 101 2.10 109 9.06 104 11.2 103 5.60 123 20.6 105 24.8 123 16.9 94 16.7 111 22.6 115 14.6 130 30.4 118 34.5 119 20.8 151 20.4 125 28.8 122 8.81 141 37.7 150 57.6 150 16.8 137 24.9 142 31.2 142 8.26 70
WOLF_ROB [149]120.5 16.7 152 24.7 151 3.17 135 9.76 139 11.8 142 5.28 102 22.6 114 24.5 121 19.4 114 21.5 143 28.4 144 8.71 30 30.8 135 35.0 136 15.2 105 20.1 115 28.3 114 7.04 97 32.0 145 48.8 143 8.28 68 25.4 145 31.8 146 8.20 55
WRT [151]127.6 10.3 103 15.8 101 2.35 121 9.41 125 10.6 63 9.35 147 35.2 148 27.6 134 26.8 145 21.2 142 21.4 102 13.6 124 31.3 144 35.6 144 13.7 69 21.5 143 30.4 143 8.56 137 39.0 153 59.6 153 16.0 133 26.1 151 32.6 152 8.31 85
GroupFlow [9]127.8 15.6 149 23.3 149 3.31 138 9.20 115 11.4 116 6.26 136 30.9 140 22.4 100 18.9 111 25.4 148 30.0 146 21.2 153 32.4 148 36.7 147 15.3 109 20.9 134 29.4 131 7.71 121 29.9 134 45.5 131 9.50 89 23.3 94 29.1 85 10.6 142
Pyramid LK [2]135.1 14.0 144 21.7 145 4.34 142 13.7 152 11.5 126 9.94 150 37.6 151 26.8 132 24.6 141 25.9 150 29.3 145 18.7 149 35.0 152 39.7 152 13.3 64 19.9 113 24.8 34 9.57 148 33.3 147 51.1 147 10.7 108 26.0 150 32.4 151 13.0 149
Periodicity [78]150.8 18.1 153 27.0 153 6.22 151 17.4 153 12.4 150 10.2 151 35.2 148 30.7 145 27.8 146 24.1 147 31.6 149 17.5 147 37.6 154 42.6 154 18.8 148 27.7 153 39.3 153 11.2 153 38.6 152 58.9 152 22.9 151 27.3 152 33.2 153 14.3 152
AVG_FLOW_ROB [142]151.6 31.2 154 31.0 154 11.6 154 19.8 154 21.4 154 12.0 154 34.9 147 32.0 147 28.1 147 31.2 154 36.2 153 23.9 154 36.4 153 41.0 153 16.9 146 39.5 154 55.0 154 16.9 154 36.8 149 52.3 148 20.8 147 30.2 154 31.8 146 15.5 154
Move the mouse over the numbers in the table to see the corresponding images. Click to compare with the ground truth.

References

Methodtime*framescolor Reference and notes
[1] 2D-CLG 844 2 gray The 2D-CLG method by Bruhn et al. as implemented by Stefan Roth. [A. Bruhn, J. Weickert, and C. Schnörr. Lucas/Kanade meets Horn/Schunck: combining local and global optic flow methods. IJCV 63(3), 2005.] Parameters were set to match the published performance on Yosemite sequence, which may not be optimal for other sequences.
[2] Pyramid LK 12 2 color A modification of Bouguet's pyramidal implementation of Lucas-Kanade.
[3] Horn & Schunck 49 2 gray A modern Matlab implementation of the Horn & Schunck method by Deqing Sun. Parameters set to optimize AAE on all training data.
[4] Black & Anandan 328 2 gray A modern Matlab implementation of the Black & Anandan method by Deqing Sun.
[5] Brox et al. 18 2 color T. Brox, A. Bruhn, N. Papenberg, and J. Weickert. High accuracy optical flow estimation based on a theory for warping. ECCV 2004. (Improved using separate robust functions as proposed in A. Bruhn and J. Weickert, Towards ultimate motion estimation, ICCV 2005; improved by training on the training set.)
[6] Fusion 2,666 2 color V. Lempitsky, S. Roth, and C. Rother. Discrete-continuous optimization for optical flow estimation. CVPR 2008.
[7] Dynamic MRF 366 2 gray B. Glocker, N. Paragios, N. Komodakis, G. Tziritas, and N. Navab. Optical flow estimation with uncertainties through dynamic MRFs. CVPR 2008. (Method improved since publication.)
[8] Second-order prior 14 2 gray W. Trobin, T. Pock, D. Cremers, and H. Bischof. An unbiased second-order prior for high-accuracy motion estimation. DAGM 2008. (Method improved since publication; for details see W. Trobin, Ph.D. thesis, 2009.)
[9] GroupFlow 600 2 gray X. Ren. Local Grouping for Optical Flow. CVPR 2008.
[10] SegOF 60 2 color L. Xu, J. Chen, and J. Jia. Segmentation based variational model for accurate optical flow estimation. ECCV 2008. Code available.
[11] Learning Flow 825 2 gray D. Sun, S. Roth, J.P. Lewis, and M. Black. Learning optical flow (SRF-LFC). ECCV 2008.
[12] CBF 69 2 color W. Trobin, T. Pock, D. Cremers, and H. Bischof. Continuous energy minimization via repeated binary fusion. ECCV 2008. (Method improved since publication; for details see W. Trobin, Ph.D. thesis, 2009.)
[13] SPSA-learn 200 2 color Y. Li and D. Huttenlocher. Learning for optical flow using stochastic optimization. ECCV 2008.
[14] GraphCuts 1,200 2 color T. Cooke. Two applications of graph-cuts to image processing. DICTA 2008.
[15] F-TV-L1 8 2 gray A. Wedel, T. Pock, J. Braun, U. Franke, and D. Cremers. Duality TV-L1 flow with fundamental matrix prior. IVCNZ 2008.
[16] FOLKI 1.4 2 gray G. Le Besnerais and F. Champagnat. Dense optical flow by iterative local window registration. ICIP 2005.
[17] TV-L1-improved 2.9 2 gray A. Wedel, T. Pock, C. Zach, H. Bischof, and D. Cremers. An improved algorithm for TV-L1 optical flow computation. Proceedings of the Dagstuhl Visual Motion Analysis Workshop 2008. Code at GPU4Vision.
[18] DPOF 287 2 color C. Lei and Y.-H. Yang. Optical flow estimation on coarse-to-fine region-trees using discrete optimization. ICCV 2009. (Method improved since publication.)
[19] Filter Flow 34,000 2 color S. Seitz and S. Baker. Filter flow. ICCV 2009.
[20] Adaptive 9.2 2 gray A. Wedel, D. Cremers, T. Pock, and H. Bischof. Structure- and motion-adaptive regularization for high accuracy optic flow. ICCV 2009.
[21] Complementary OF 44 2 color H. Zimmer, A. Bruhn, J. Weickert, L. Valgaerts, A. Salgado, B. Rosenhahn, and H.-P. Seidel. Complementary optic flow. EMMCVPR 2009.
[22] Aniso. Huber-L1 2 2 gray M. Werlberger, W. Trobin, T. Pock, A. Wedel, D. Cremers, and H. Bischof. Anisotropic Huber-L1 optical flow. BMVC 2009. Code at GPU4Vision.
[23] Rannacher 0.12 2 gray J. Rannacher. Realtime 3D motion estimation on graphics hardware. Bachelor thesis, Heidelberg University, 2009.
[24] TI-DOFE 260 2 gray C. Cassisa, S. Simoens, and V. Prinet. Two-frame optical flow formulation in an unwarped multiresolution scheme. CIARP 2009.
[25] NL-TV-NCC 20 2 color M. Werlberger, T. Pock, and H. Bischof. Motion estimation with non-local total variation regularization. CVPR 2010.
[26] MDP-Flow 188 2 color L. Xu, J. Jia, and Y. Matsushita. Motion detail preserving optical flow estimation. CVPR 2010.
[27] ACK-Prior 5872 2 color K. Lee, D. Kwon, I. Yun, and S. Lee. Optical flow estimation with adaptive convolution kernel prior on discrete framework. CVPR 2010.
[28] LDOF 122 2 color T. Brox and J. Malik. Large displacement optical flow: descriptor matching in variational motion estimation. PAMI 33(3):500-513, 2011.
[29] p-harmonic 565 2 gray J. Gai and R. Stevenson. Optical flow estimation with p-harmonic regularization. ICIP 2010.
[30] TriangleFlow 4200 2 gray B. Glocker, H. Heibel, N. Navab, P. Kohli, and C. Rother. TriangleFlow: Optical flow with triangulation-based higher-order likelihoods. ECCV 2010.
[31] Classic+NL 972 2 color D. Sun, S. Roth, and M. Black. Secrets of optical flow estimation and their principles. CVPR 2010. Matlab code.
[32] Classic++ 486 2 gray A modern implementation of the classical formulation descended from Horn & Schunck and Black & Anandan; see D. Sun, S. Roth, and M. Black, Secrets of optical flow estimation and their principles, CVPR 2010.
[33] Nguyen 33 2 gray D. Nguyen. Tuning optical flow estimation with image-driven functions. ICRA 2011.
[34] Modified CLG 133 2 gray R. Fezzani, F. Champagnat, and G. Le Besnerais. Combined local global method for optic flow computation. EUSIPCO 2010.
[35] ComplOF-FED-GPU 0.97 2 color P. Gwosdek, H. Zimmer, S. Grewenig, A. Bruhn, and J. Weickert. A highly efficient GPU implementation for variational optic flow based on the Euler-Lagrange framework. CVGPU Workshop 2010.
[36] Ad-TV-NDC 35 2 gray M. Nawaz. Motion estimation with adaptive regularization and neighborhood dependent constraint. DICTA 2010.
[37] Layers++ 18206 2 color D. Sun, E. Sudderth, and M. Black. Layered image motion with explicit occlusions, temporal consistency, and depth ordering. NIPS 2010.
[38] OFH 620 3 color H. Zimmer, A. Bruhn, J. Weickert. Optic flow in harmony. IJCV 93(3) 2011.
[39] LSM 1615 2 color K. Jia, X. Wang, and X. Tang. Optical flow estimation using learned sparse model. ICCV 2011.
[40] CostFilter 55 2 color C. Rhemann, A. Hosni, M. Bleyer, C. Rother, and M. Gelautz. Fast cost-volume filtering for visual correspondence and beyond. CVPR 2011.
[41] Bartels 0.15 2 gray C. Bartels and G. de Haan. Smoothness constraints in recursive search motion estimation for picture rate conversion. IEEE TCSVT 2010. Version improved since publication: mapped on GPU.
[42] Shiralkar 600 2 gray M. Shiralkar and R. Schalkoff. A self organization-based optical flow estimator with GPU implementation. MVA 23(6):1229-1242.
[43] HBpMotionGpu 1000 5 gray S. Grauer-Gray and C. Kambhamettu. Hierarchical belief propagation to reduce search space using CUDA for stereo and motion estimation. WACV 2009. (Method improved since publication.)
[44] StereoFlow 7200 2 color G. Rosman, S. Shem-Tov, D. Bitton, T. Nir, G. Adiv, R. Kimmel, A. Feuer, and A. Bruckstein. Over-parameterized optical flow using a stereoscopic constraint. SSVM 2011:761-772.
[45] Adaptive flow 121 2 gray T. Arici. Energy minimization based motion estimation using adaptive smoothness priors. Submitted to IEEE TIP 2011.
[46] TC-Flow 2500 5 color S. Volz, A. Bruhn, L. Valgaerts, and H. Zimmer. Modeling temporal coherence for optical flow. ICCV 2011.
[47] SLK 300 2 gray T. Corpetti and E. Mémin. Stochastic uncertainty models for the luminance consistency assumption. IEEE TIP 2011.
[48] CLG-TV 29 2 gray M. Drulea. Total variation regularization of local-global optical flow. ITSC 2011. Matlab code.
[49] SimpleFlow 1.7 2 color M. Tao, J. Bai, P. Kohli, S. Paris. SimpleFlow: a non-iterative, sublinear optical flow algorithm. EUROGRAPHICS 2012.
[50] IAOF 57 2 gray D. Nguyen. Improving motion estimation using image-driven functions and hybrid scheme. PSIVT 2011.
[51] IAOF2 56 2 gray D. Nguyen. Enhancing the sharpness of flow field using image-driven functions with occlusion-aware filter. Submitted to TIP 2011.
[52] LocallyOriented 9541 2 gray Y.Niu, A. Dick, and M. Brooks. Locally oriented optical flow computation. To appear in TIP 2012.
[53] IROF-TV 261 2 color H. Rashwan, D. Puig, and M. Garcia. On improving the robustness of differential optical flow. ICCV 2011 Artemis workshop.
[54] Sparse Occlusion 2312 2 color A. Ayvaci, M. Raptis, and S. Soatto. Sparse occlusion detection with optical flow. Submitted to IJCV 2011.
[55] PGAM+LK 0.37 2 gray A. Alba, E. Arce-Santana, and M. Rivera. Optical flow estimation with prior models obtained from phase correlation. ISVC 2010.
[56] Sparse-NonSparse 713 2 color L. Chen, J. Wang, and Y. Wu. Decomposing and regularizing sparse/non-sparse components for motion field estimation. Submitted to PAMI 2013.
[57] nLayers 36150 4 color D. Sun, E. Sudderth, and M. Black. Layered segmentation and optical flow estimation over time. CVPR 2012.
[58] IROF++ 187 2 color H. Rashwan, D. Puig, and M. Garcia. Variational optical flow estimation based on stick tensor voting. IEEE TIP 2013.
[59] COFM 600 3 color M. Mozerov. Constrained optical flow estimation as a matching problem. IEEE TIP 2013.
[60] Efficient-NL 400 2 color P. Krähenbühl and V. Koltun. Efficient nonlocal regularization for optical flow. ECCV 2012.
[61] BlockOverlap 2 2 gray M. Santoro, G. AlRegib, and Y. Altunbasak. Motion estimation using block overlap minimization. Submitted to MMSP 2012.
[62] Ramp 1200 2 color A. Singh and N. Ahuja. Exploiting ramp structures for improving optical flow estimation. ICPR 2012.
[63] Occlusion-TV-L1 538 3 gray C. Ballester, L. Garrido, V. Lazcano, and V. Caselles. A TV-L1 optical flow method with occlusion detection. DAGM-OAGM 2012.
[64] TV-L1-MCT 90 2 color M. Mohamed and B. Mertsching. TV-L1 optical flow estimation with image details recovering based on modified census transform. ISVC 2012.
[65] Local-TV-L1 500 2 gray L. Raket. Local smoothness for global optical flow. ICIP 2012.
[66] ALD-Flow 61 2 color M. Stoll, A. Bruhn, and S. Volz. Adaptive integration of feature matches into variational optic flow methods. ACCV 2012.
[67] SIOF 234 2 color L. Xu, Z. Dai, and J. Jia. Scale invariant optical flow. ECCV 2012.
[68] MDP-Flow2 342 2 color L. Xu, J. Jia, and Y. Matsushita. Motion detail preserving optical flow estimation. PAMI 34(9):1744-1757, 2012. Code available.
[69] TCOF 1421 all gray J. Sanchez, A. Salgado, and N. Monzon. Optical flow estimation with consistent spatio-temporal coherence models. VISAPP 2013.
[70] LME 476 2 color W. Li, D. Cosker, M. Brown, and R. Tang. Optical flow estimation using Laplacian mesh energy. CVPR 2013.
[71] NN-field 362 2 color L. Chen, H. Jin, Z. Lin, S. Cohen, and Y. Wu. Large displacement optical flow from nearest neighbor fields. CVPR 2013.
[72] FESL 3310 2 color W. Dong, G. Shi, X. Hu, and Y. Ma. Nonlocal sparse and low-rank regularization for optical flow estimation. Submitted to IEEE TIP 2013.
[73] PMF 35 2 color J. Lu, H. Yang, D. Min, and M. Do. PatchMatch filter: efficient edge-aware filtering meets randomized search for fast correspondence field estimation. CVPR 2013.
[74] FC-2Layers-FF 2662 4 color D. Sun, J. Wulff, E. Sudderth, H. Pfister, and M. Black. A fully-connected layered model of foreground and background flow. CVPR 2013.
[75] Correlation Flow 290 2 color M. Drulea and S. Nedevschi. Motion estimation using the correlation transform. TIP 2013. Matlab code.
[76] TC/T-Flow 341 5 color M. Stoll, S. Volz, and A. Bruhn. Joint trilateral filtering for multiframe optical flow. ICIP 2013.
[77] OFLAF 1530 2 color T. Kim, H. Lee, and K. Lee. Optical flow via locally adaptive fusion of complementary data costs. ICCV 2013.
[78] Periodicity 8000 4 color G. Khachaturov, S. Gonzalez-Brambila, and J. Gonzalez-Trejo. Periodicity-based computation of optical flow. Submitted to Computacion y Sistemas (CyS) 2013.
[79] SILK 572 2 gray P. Zille, C. Xu, T. Corpetti, L. Shao. Observation models based on scale interactions for optical flow estimation. Submitted to IEEE TIP.
[80] CRTflow 13 3 color O. Demetz, D. Hafner, and J. Weickert. The complete rank transform: a tool for accurate and morphologically invariant matching of structures. BMVC 2013.
[81] SuperFlow 178 2 color Anonymous. Superpixel based optical flow estimation. ICCV 2013 submission 507.
[82] Aniso-Texture 300 2 color Anonymous. Texture information-based optical flow estimation using an incremental multi-resolution approach. ITC-CSCC 2013 submission 267.
[83] Classic+CPF 640 2 gray Z. Tu, R. Veltkamp, and N. van der Aa. A combined post-filtering method to improve accuracy of variational optical flow estimation. Submitted to Pattern Recognition 2013.
[84] S2D-Matching 1200 2 color Anonymous. Locally affine sparse-to-dense matching for motion and occlusion estimation. ICCV 2013 submission 1479.
[85] AGIF+OF 438 2 gray Z. Tu, R. Poppe, and R. Veltkamp. Adaptive guided image filter to warped interpolation image for variational optical flow computation. Submitted to Signal Processing 2015.
[86] DeepFlow 13 2 color P. Weinzaepfel, J. Revaud, Z. Harchaoui, and C. Schmid. DeepFlow: large displacement optical flow with deep matching. ICCV 2013.
[87] NNF-Local 673 2 color Z. Chen, H. Jin, Z. Lin, S. Cohen, and Y. Wu. Large displacement optical flow with nearest neighbor field. Submitted to PAMI 2014.
[88] EPPM w/o HM 2.5 2 color L. Bao, Q. Yang, and H. Jin. Fast edge-preserving PatchMatch for large displacement optical flow. CVPR 2014.
[89] MLDP_OF 165 2 gray M. Mohamed, H. Rashwan, B. Mertsching, M. Garcia, and D. Puig. Illumination-robust optical flow approach using local directional pattern. IEEE TCSVT 24(9):1499-1508, 2014.
[90] RFlow 20 2 gray S. Ali, C. Daul, and W. Blondel. Robust and accurate optical flow estimation for weak texture and varying illumination condition: Application to cystoscopy. IPTA 2014.
[91] SRR-TVOF-NL 32 all color P. Pohl, M. Sirotenko, E. Tolstaya, and V. Bucha. Edge preserving motion estimation with occlusions correction for assisted 2D to 3D conversion. IS&T/SPIE Electronic Imaging 2014.
[92] 2DHMM-SAS 157 2 color M.-C. Shih, R. Shenoy, and K. Rose. A two-dimensional hidden Markov model with spatially-adaptive states with application of optical flow. ICIP 2014 submission.
[93] WLIF-Flow 700 2 color Z. Tu, R. Veltkamp, N. van der Aa, and C. Van Gemeren. Weighted local intensity fusion method for variational optical flow estimation. Submitted to TIP 2014.
[94] FMOF 215 2 color N. Jith, A. Ramakanth, and V. Babu. Optical flow estimation using approximate nearest neighbor field fusion. ICASSP 2014.
[95] TriFlow 150 2 color TriFlow. Optical flow with geometric occlusion estimation and fusion of multiple frames. ECCV 2014 submission 914.
[96] ComponentFusion 6.5 2 color Anonymous. Fast optical flow by component fusion. ECCV 2014 submission 941.
[97] AggregFlow 1642 2 color D. Fortun, P. Bouthemy, and C. Kervrann. Aggregation of local parametric candidates and exemplar-based occlusion handling for optical flow. Preprint arXiv:1407.5759.
[98] 2bit-BM-tele 124 2 gray R. Xu and D. Taubman. Robust dense block-based motion estimation using a two-bit transform on a Laplacian pyramid. ICIP 2013.
[99] HCIC-L 330 2 color Anonymous. Globally-optimal image correspondence using a hierarchical graphical model. NIPS 2014 submission 114.
[100] TF+OM 600 2 color R. Kennedy and C. Taylor. Optical flow with geometric occlusion estimation and fusion of multiple frames. EMMCVPR 2015.
[101] PH-Flow 800 2 color J. Yang and H. Li. Dense, accurate optical flow estimation with piecewise parametric model. CVPR 2015.
[102] EpicFlow 16 2 color J. Revaud, P. Weinzaepfel, Z. Harchaoui, and C. Schmid. EpicFlow: edge-preserving interpolation of correspondences for optical flow. CVPR 2015.
[103] NNF-EAC 380 2 color Anonymous. Variational method for joint optical flow estimation and edge-aware image restoration. CVPR 2015 submission 2336.
[104] Heeger++ 6600 5 gray Anonymous. A context aware biologically inspired algorithm for optical flow (updated results). CVPR 2015 submission 2238.
[105] HBM-GC 330 2 color A. Zheng and Y. Yuan. Motion estimation via hierarchical block matching and graph cut. Submitted to ICIP 2015.
[106] FFV1MT 358 5 gray F. Solari, M. Chessa, N. Medathati, and P. Kornprobst. What can we expect from a V1-MT feedforward architecture for optical flow estimation? Submitted to Signal Processing: Image Communication 2015.
[107] ROF-ND 4 2 color S. Ali, C. Daul, E. Galbrun, and W. Blondel. Illumination invariant large displacement optical flow using robust neighbourhood descriptors. Submitted to CVIU 2015.
[108] DeepFlow2 16 2 color J. Revaud, P. Weinzaepfel, Z. Harchaoui, and C. Schmid. Deep convolutional matching. Submitted to IJCV, 2015.
[109] HAST 2667 2 color Anonymous. Highly accurate optical flow estimation on superpixel tree. ICCV 2015 submission 2221.
[110] FlowFields 15 2 color C. Bailer, B. Taetz, and D. Stricker. Flow Fields: Dense unregularized correspondence fields for highly accurate large displacement optical flow estimation. ICCV 2015.
[111] SVFilterOh 1.56 2 color Anonymous. Fast estimation of large displacement optical flow using PatchMatch and dominant motion patterns. CVPR 2016 submission 1788.
[112] FlowNetS+ft+v 0.5 2 color Anonymous. Learning optical flow with convolutional neural networks. ICCV 2015 submission 235.
[113] CombBMOF 51 2 color M. Brüggemann, R. Kays, P. Springer, and O. Erdler. Combined block-matching and adaptive differential motion estimation in a hierarchical multi-scale framework. ICGIP 2014. (Method improved since publication.)
[114] PMMST 182 2 color F. Zhang, S. Xu, and X. Zhang. High accuracy correspondence field estimation via MST based patch matching. Submitted to TIP 2015.
[115] DF-Auto 70 2 color N. Monzon, A. Salgado, and J. Sanchez. Regularization strategies for discontinuity-preserving optical flow methods. Submitted to TIP 2015.
[116] CPM-Flow 3 2 color Anonymous. Efficient coarse-to-fine PatchMatch for large displacement optical flow. CVPR 2016 submission 241.
[117] CNN-flow-warp+ref 1.4 3 color D. Teney and M. Hebert. Learning to extract motion from videos in convolutional neural networks. ArXiv 1601.07532, 2016.
[118] Steered-L1 804 2 color Anonymous. Optical flow estimation via steered-L1 norm. Submitted to WSCG 2016.
[119] StereoOF-V1MT 343 2 gray Anonymous. Visual features for action-oriented tasks: a cortical-like model for disparity and optic flow computation. BMVC 2016 submission 132.
[120] PGM-C 5 2 color Y. Li. Pyramidal gradient matching for optical flow estimation. Submitted to PAMI 2016.
[121] RNLOD-Flow 1040 2 gray C. Zhang, Z. Chen, M. Wang, M. Li, and S. Jiang. Robust non-local TV-L1 optical flow estimation with occlusion detection. IEEE TIP 26(8):4055-4067, 2017.
[122] FlowNet2 0.091 2 color Anonymous. FlowNet 2.0: Evolution of optical flow estimation with deep networks. CVPR 2017 submission 900.
[123] S2F-IF 20 2 color Anonymous. S2F-IF: Slow-to-fast interpolator flow. CVPR 2017 submission 765.
[124] BriefMatch 0.068 2 gray G. Eilertsen, P.-E. Forssen, and J. Unger. Dense binary feature matching for real-time optical flow estimation. SCIA 2017 submission 62.
[125] OAR-Flow 60 2 color Anonymous. Order-adaptive regularisation for variational optical flow: global, local and in between. SSVM 2017 submission 20.
[126] AdaConv-v1 2.8 2 color S. Niklaus, L. Mai, and F. Liu. (Interpolation results only.) Video frame interpolation via adaptive convolution. CVPR 2017.
[127] SepConv-v1 0.2 2 color S. Niklaus, L. Mai, and F. Liu. (Interpolation results only.) Video frame interpolation via adaptive separable convolution. ICCV 2017.
[128] ProbFlowFields 37 2 color A. Wannenwetsch, M. Keuper, and S. Roth. ProbFlow: joint optical flow and uncertainty estimation. ICCV 2017.
[129] UnFlow 0.12 2 color Anonymous. UnFlow: Unsupervised learning of optical flow with a bidirectional census loss. Submitted to AAAI 2018.
[130] FlowFields+ 10.5 2 color C. Bailer, B. Taetz, and D. Stricker. Flow fields: Dense correspondence fields for highly accurate large displacement optical flow estimation. Submitted to PAMI 2017.
[131] IIOF-NLDP 150 2 color D.-H. Trinh, W. Blondel, and C. Daul. A general form of illumination-invariant descriptors in variational optical flow estimation. ICIP 2017.
[132] SuperSlomo 0.5 2 color Anonymous. (Interpolation results only.) Super SloMo: High quality estimation of multiple intermediate frames for video interpolation. CVPR 2018 submission 325.
[133] EPMNet 0.061 2 color Anonymous. EPM-convolution multilayer-network for optical flow estimation. ICME 2018 submission 1119.
[134] OFRF 90 2 color T. Mai, M. Gouiffes, and S. Bouchafa. Optical Flow refinement using iterative propagation under color, proximity and flow reliability constraints. Submitted to Signal, Image and Video Processing 2017.
[135] 3DFlow 328 2 color J. Chen, Z. Cai, J. Lai, and X. Xie. A filtering based framework for optical flow estimation. IEEE TCSVT 2018.
[136] FGIK 0.18 2 color Anonymous. (Interpolation results only.) Learning flow-guided interpolation kernels for video frame synthesis. ECCV 2018 submission 433.
[137] CtxSyn 0.07 2 color S. Niklaus and F. Liu. (Interpolation results only.) Context-aware synthesis for video frame interpolation. CVPR 2018.
[138] H+S_ROB 5 2 color ROB 2018 baseline submission, based on: E. Meinhardt-Llopis, J. Sanchez, and D. Kondermann. Horn-Schunck optical flow with a multi-scale strategy. Image Processing On Line 3:151–172, 2013.
[139] TVL1_ROB 1 2 color ROB 2018 baseline submission, based on: J. Sanchez, E. Meinhardt-Llopis, and G. Facciolo. TV-L1 optical flow estimation. Image Processing On Line 3:137-150, 2013.
[140] DMF_ROB 10 2 color ROB 2018 baseline submission, based on: P. Weinzaepfel, J. Revaud, Z. Harchaoui, and C. Schmid. DeepFlow: large displacement optical flow with deep matching. ICCV 2013.
[141] JOF 657 2 gray C. Zhang, L. Ge, Z. Chen, M. Li, W. Liu, and H. Chen. Refined TV-L1 optical flow estimation using joint filtering. Submitted to IEEE TMM, 2018.
[142] AVG_FLOW_ROB N/A 2 N/A Average flow field of ROB 2018 training set.
[143] LiteFlowNet 0.06 2 color T.-W. Hui, X. Tang, and C. C. Loy. LiteFlowNet: A lightweight convolutional neural network for optical flow estimation. CVPR 2018.
[144] AugFNG_ROB 0.10 all color Anonymous. FusionNet and AugmentedFlowNet: Selective proxy ground truth for training on unlabeled images. ECCV 2018 submission 2834.
[145] ResPWCR_ROB 0.2 2 color Anonymous. Learning optical flow with residual connections. ROB 2018 submission.
[146] FF++_ROB 17.43 2 color R. Schuster, C. Bailer, O. Wasenmueller, D. Stricker. FlowFields++: Accurate optical flow correspondences meet robust interpolation. ICIP 2018. Submitted to ROB 2018.
[147] ProFlow_ROB 76 3 color Anonymous. ProFlow: Learning to predict optical flow. BMVC 2018 submission 277.
[148] PWC-Net_ROB 0.069 2 color D. Sun, X. Yang, M.-Y. Liu, and J. Kautz. PWC-Net: CNNs for optical flow using pyramid, warping, and cost volume. CVPR 2018.
[149] WOLF_ROB 0.02 2 color Anonymous. Reversed deep neural network for optical flow. ROB 2018 submission.
[150] LFNet_ROB 0.068 2 color Anonymous. Learning a flow network. ROB 2018 submission.
[151] WRT 9 2 color L. Mei, J. Lai, X. Xie, J. Zhu, and J. Chen. Illumination-invariance optical flow estimation using weighted regularization transform. Submitted to IEEE TCSVT 2018.
[152] EAI-Flow 2.1 2 color Anonymous. Hierarchical coherency sensitive hashing and interpolation with RANSAC for large displacement optical flow. CVIU 2018 submission 17-678.
[153] ContinualFlow_ROB 0.5 all color Anonymous. Continual Flow. ACCV 2018 submission 273.
[154] CyclicGen 0.088 2 color Anonymous. (Interpolation results only.) Deep video frame interpolation using cyclic frame generation. AAAI 2019 submission 323.
* The "time" column lists the reported runtime in seconds on the "Urban" sequence. Note that these runtimes are not normalized by processor speed or type.