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        
SD
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
PyrWarp [159]3.4 0.73 1 0.99 1 0.64 1 1.30 2 1.69 4 0.91 1 1.77 1 1.79 4 1.20 1 1.15 1 1.14 1 1.25 1 1.04 1 1.07 1 1.03 1 1.34 2 1.84 2 0.93 1 2.70 4 4.16 4 1.17 3 3.27 16 4.13 16 0.87 12
CyclicGen [153]9.8 0.95 3 1.12 3 0.92 79 1.34 4 1.60 1 1.32 60 2.85 4 1.91 5 3.24 11 1.56 5 1.62 4 1.63 22 1.16 6 1.20 5 1.16 6 1.22 1 1.56 1 1.03 4 2.52 3 3.88 3 0.94 1 1.41 1 1.78 1 0.67 1
DAIN [158]10.2 0.96 4 1.39 5 0.76 4 1.41 10 1.84 10 0.96 7 3.89 6 2.27 7 3.44 12 1.52 4 1.89 6 1.64 23 1.23 8 1.32 9 1.07 3 1.48 4 2.07 5 1.02 3 3.52 15 5.44 15 1.19 4 4.64 31 5.86 33 0.90 16
MPRN [157]16.3 1.50 39 2.37 40 0.75 3 1.61 25 2.11 30 1.18 38 2.31 2 2.63 9 1.35 2 1.77 8 2.06 7 1.55 15 1.23 8 1.29 8 1.19 9 2.00 11 2.83 11 1.29 10 2.50 2 3.85 2 1.41 6 5.37 50 6.78 52 0.80 5
CtxSyn [136]16.6 0.96 4 1.26 4 0.89 63 1.23 1 1.60 1 0.91 1 9.66 108 1.50 1 12.6 115 1.29 2 1.43 2 1.36 2 1.13 4 1.15 4 1.18 8 1.49 5 2.05 4 1.07 7 4.20 22 6.49 22 1.46 10 2.48 2 3.13 2 0.77 4
CFRF [156]17.7 1.20 17 1.69 13 1.07 127 1.64 31 2.06 25 1.12 29 4.03 8 1.71 3 3.98 18 1.37 3 1.55 3 1.58 20 1.09 3 1.11 3 1.12 5 1.43 3 1.96 3 1.06 6 2.90 6 4.47 6 0.98 2 5.05 39 6.38 41 0.84 10
InterpCNN [160]20.0 1.69 52 2.65 52 0.95 94 1.47 11 1.90 13 0.99 12 4.35 15 2.06 6 5.49 34 1.85 10 2.42 13 2.28 62 1.13 4 1.20 5 1.05 2 1.67 6 2.36 6 0.98 2 2.70 4 4.16 4 1.46 10 3.73 21 4.72 22 0.94 21
SuperSlomo [132]20.1 1.40 29 2.16 28 0.87 46 1.67 36 2.06 25 1.32 60 2.53 3 3.23 11 2.15 4 2.12 18 2.35 11 2.21 55 1.18 7 1.22 7 1.16 6 1.98 10 2.79 10 1.27 9 2.34 1 3.59 1 1.44 7 4.21 28 5.32 30 1.08 41
OFRI [161]21.9 0.86 2 1.10 2 0.83 16 2.39 134 1.89 12 3.28 154 3.95 7 1.52 2 4.10 20 1.70 6 2.09 8 2.14 50 1.07 2 1.09 2 1.08 4 2.28 14 3.24 14 1.05 5 2.99 7 4.60 7 1.29 5 3.52 18 4.45 18 0.90 16
MEMC-Net+ [155]31.1 1.13 12 1.73 16 0.77 5 1.63 28 1.87 11 1.83 124 8.23 89 3.22 10 10.1 88 1.75 7 2.14 9 2.15 51 2.86 42 3.25 42 2.66 63 1.72 7 2.42 7 1.40 13 4.87 35 7.54 35 1.54 13 3.62 19 4.57 19 0.73 2
SepConv-v1 [127]36.2 2.15 111 3.44 113 0.69 2 1.70 39 2.16 37 1.17 35 8.00 85 3.78 13 7.22 55 2.71 72 2.44 14 2.23 56 1.63 16 1.81 17 1.22 11 1.80 8 2.53 8 1.09 8 3.88 18 6.01 18 1.90 24 5.43 52 6.86 54 0.76 3
TOF-M [154]38.6 1.31 21 2.02 23 1.05 122 1.84 66 2.33 67 1.23 46 5.50 39 2.54 8 7.05 54 2.29 32 1.76 5 1.84 35 1.72 19 1.93 19 1.21 10 1.83 9 2.56 9 1.42 14 5.37 53 8.30 53 1.44 7 8.23 105 10.4 104 0.81 6
NN-field [71]47.1 1.97 74 3.14 81 0.84 27 1.34 4 1.70 5 0.97 8 6.72 54 10.3 97 8.20 71 2.55 60 3.41 83 2.65 96 1.76 20 1.96 20 1.22 11 3.40 57 4.85 60 1.79 34 4.57 31 7.07 31 3.12 55 5.05 39 6.38 41 1.21 71
PMMST [114]47.8 1.96 73 3.12 74 0.81 9 1.60 24 2.09 29 1.03 21 6.35 48 9.32 84 7.35 57 2.93 90 3.99 117 2.65 96 1.64 18 1.82 18 1.23 14 3.08 37 4.39 37 1.53 18 4.97 42 7.68 42 2.63 36 5.04 36 6.37 39 1.34 89
SuperFlow [81]49.5 1.88 66 2.97 66 0.89 63 1.91 77 2.39 71 1.45 79 4.52 17 5.09 21 4.02 19 3.56 127 2.61 24 2.27 59 1.46 11 1.59 11 1.28 34 3.30 48 4.70 48 2.24 72 4.55 30 7.03 30 2.84 46 7.19 80 9.09 81 0.83 9
DeepFlow [86]50.2 2.03 96 3.23 97 0.83 16 1.95 81 2.44 82 1.58 96 4.26 13 5.34 28 2.24 6 2.77 79 2.57 21 1.98 40 3.30 51 3.75 51 1.27 29 2.92 33 4.16 33 1.69 27 3.34 11 5.16 11 1.58 17 8.21 103 10.4 104 1.27 79
ALD-Flow [66]50.7 1.35 25 2.01 22 0.88 52 1.79 62 2.34 68 1.25 50 4.95 23 5.90 37 3.11 9 1.97 14 2.56 20 1.48 8 3.77 67 4.29 68 6.03 95 3.99 94 5.70 95 4.20 143 3.92 19 6.07 19 1.44 7 7.78 91 9.82 92 1.05 37
ComplOF-FED-GPU [35]52.2 1.49 37 2.33 38 0.88 52 1.78 60 2.32 66 1.26 52 9.54 107 4.00 14 12.2 111 1.86 11 2.44 14 1.48 8 4.22 79 4.80 79 5.82 89 2.62 19 3.73 20 1.97 48 4.52 29 6.91 28 3.50 73 8.01 98 10.1 97 0.95 23
OAR-Flow [125]52.7 1.61 48 2.39 42 0.83 16 1.84 66 2.37 70 1.34 64 5.20 29 6.66 53 3.10 8 1.83 9 2.39 12 1.48 8 4.50 118 5.12 118 6.58 131 3.38 55 4.81 56 2.73 100 3.73 17 5.70 17 2.47 30 6.18 71 7.80 72 1.15 54
NNF-Local [87]53.9 1.08 9 1.64 12 0.79 6 1.30 2 1.64 3 0.98 10 8.95 100 13.9 121 11.2 102 2.52 54 3.37 82 2.49 76 1.45 10 1.58 10 1.22 11 4.26 114 6.08 115 2.22 70 5.38 54 8.33 54 5.57 133 4.99 33 6.31 35 1.26 78
IROF++ [58]55.4 1.07 7 1.58 7 0.81 9 1.67 36 2.16 37 1.08 26 6.49 51 7.78 66 6.90 51 1.93 12 2.55 19 2.04 42 4.40 93 5.00 93 6.59 142 3.77 80 5.38 82 1.84 38 3.98 20 6.16 20 2.66 39 10.5 152 13.3 153 1.15 54
TC-Flow [46]55.8 1.12 10 1.62 9 0.85 37 1.80 64 2.34 68 1.25 50 3.55 5 5.28 25 1.46 3 2.07 17 2.71 30 2.05 43 4.38 90 4.98 89 6.27 103 4.03 99 5.75 102 2.60 92 4.81 34 7.44 34 2.55 33 8.05 100 10.2 101 1.44 101
ProFlow_ROB [146]58.5 1.44 32 2.26 35 0.82 12 1.72 44 2.26 56 1.28 54 5.71 41 5.11 22 5.72 39 2.28 31 3.06 51 2.65 96 4.51 121 5.13 121 6.58 131 2.48 16 3.52 16 1.47 15 5.97 74 9.24 74 2.58 34 7.38 86 9.32 87 1.63 117
CLG-TV [48]58.7 1.83 63 2.89 64 0.97 101 2.20 113 2.74 124 1.65 106 5.14 28 6.68 54 5.81 42 2.23 27 2.69 27 2.60 90 4.17 78 4.74 78 1.25 20 2.65 22 3.77 22 1.47 15 3.24 10 5.01 10 1.66 18 9.56 136 12.1 136 0.96 25
Second-order prior [8]59.1 1.12 10 1.62 9 0.99 107 2.11 101 2.55 94 1.57 93 5.29 31 8.01 69 5.87 43 2.15 21 2.65 25 1.71 29 4.23 81 4.81 81 1.25 20 2.88 30 4.09 31 1.81 36 5.02 44 7.77 44 1.51 12 9.26 128 11.7 127 2.10 153
DeepFlow2 [108]59.6 2.32 125 3.69 127 0.87 46 1.90 73 2.43 78 1.30 57 4.25 10 5.73 33 2.18 5 3.00 96 3.42 86 1.92 37 3.44 54 3.91 55 2.60 62 2.24 13 3.17 13 1.48 17 3.41 12 5.27 13 2.92 50 8.45 109 10.7 110 2.03 149
FMOF [94]59.8 1.71 55 2.65 52 0.94 88 1.53 17 1.96 15 1.07 24 9.48 105 13.0 117 10.7 96 2.56 61 3.00 45 3.12 128 2.35 30 2.66 30 1.24 17 3.28 45 4.68 45 1.59 21 4.90 38 7.58 38 2.28 29 10.4 150 13.2 150 1.06 38
TC/T-Flow [76]60.3 1.54 42 2.36 39 1.06 124 1.85 68 2.41 76 1.44 77 5.40 34 7.43 62 5.66 37 2.57 63 2.47 16 1.48 8 4.45 104 5.06 104 6.55 120 3.37 54 4.80 55 1.36 11 4.27 23 6.49 22 4.17 96 7.94 95 10.0 96 0.94 21
LME [70]60.5 1.93 69 3.07 70 0.81 9 1.63 28 2.14 34 1.13 30 5.40 34 7.66 64 5.53 36 2.44 43 3.25 67 2.60 90 4.48 112 5.10 113 6.48 109 4.83 137 6.89 137 1.65 22 4.28 24 6.62 24 3.05 54 5.03 35 6.35 37 1.23 75
SIOF [67]60.8 1.33 22 2.02 23 0.92 79 2.30 126 2.84 136 1.72 113 6.84 60 9.18 82 6.21 44 2.57 63 3.17 62 2.81 107 1.61 15 1.78 15 1.27 29 3.79 82 5.40 83 1.54 19 4.49 28 6.94 29 3.42 68 5.85 65 7.39 66 1.09 43
MDP-Flow2 [68]60.8 1.94 71 3.09 73 0.79 6 1.62 26 2.13 33 0.95 5 9.47 104 14.8 126 11.9 108 2.53 57 3.42 86 3.00 122 1.63 16 1.80 16 1.95 56 4.25 113 6.07 114 1.81 36 5.50 58 8.52 58 2.71 40 5.02 34 6.34 36 1.19 66
MLDP_OF [89]61.2 1.43 31 2.24 34 0.93 85 1.96 85 2.48 86 1.38 71 5.03 27 6.04 38 3.03 7 2.80 81 3.36 80 2.91 113 3.46 56 3.94 56 4.42 79 2.77 26 3.94 26 2.43 84 4.88 36 7.54 35 4.19 97 5.83 64 7.37 65 1.47 106
WLIF-Flow [93]61.8 1.03 6 1.50 6 0.82 12 1.73 48 2.21 44 1.17 35 8.68 95 12.4 113 10.2 90 2.85 84 3.53 92 3.31 133 2.45 31 2.77 31 3.18 69 3.83 87 5.46 88 3.13 115 5.25 50 8.12 50 2.79 44 5.06 42 6.40 45 1.22 72
CBF [12]61.8 1.34 23 2.09 26 0.88 52 2.14 107 2.64 105 1.69 109 5.42 36 7.67 65 5.70 38 2.25 28 2.57 21 2.56 84 1.49 12 1.63 12 1.32 47 2.58 18 3.67 18 1.85 40 7.15 112 11.1 113 4.00 90 8.03 99 10.1 97 1.76 131
OFLAF [77]62.7 1.97 74 3.13 76 0.83 16 1.37 9 1.78 9 1.01 17 5.30 32 6.60 52 4.28 21 2.27 29 3.06 51 1.56 16 4.60 128 5.23 128 6.56 122 3.41 59 4.85 60 2.63 94 6.96 104 10.8 106 3.83 84 5.52 56 6.96 56 1.47 106
OFH [38]62.9 1.41 30 2.17 30 0.91 77 1.96 85 2.47 85 1.32 60 7.12 68 9.37 86 6.61 48 2.03 16 2.69 27 1.47 5 4.26 83 4.84 83 5.89 90 3.02 36 4.29 36 2.26 73 6.29 84 9.37 77 6.74 141 6.88 79 8.69 80 0.98 30
IROF-TV [53]63.1 1.93 69 3.06 69 0.94 88 1.75 51 2.24 52 1.20 39 4.66 19 6.05 39 3.72 14 2.13 20 2.85 34 1.53 14 4.41 95 5.02 95 6.59 142 3.61 66 5.14 68 2.31 76 6.73 94 10.4 96 3.37 66 6.77 77 8.56 78 1.15 54
p-harmonic [29]63.4 1.81 61 2.87 62 0.84 27 2.26 123 2.86 137 2.12 139 4.57 18 5.80 34 3.76 15 2.95 91 2.80 33 2.74 102 3.65 63 4.15 63 1.26 26 3.01 35 4.28 35 2.08 57 4.42 27 6.84 27 4.12 93 8.75 114 11.1 115 0.96 25
Aniso. Huber-L1 [22]65.7 1.44 32 2.23 32 0.90 74 2.23 116 2.73 120 1.50 86 5.43 37 6.83 56 5.46 33 2.49 49 2.94 41 3.01 123 4.09 76 4.65 76 1.25 20 4.29 115 6.12 116 1.38 12 4.08 21 6.31 21 1.72 19 10.2 146 12.9 147 0.82 8
PGM-C [120]65.8 1.60 47 2.51 48 0.83 16 1.56 22 2.04 22 0.99 12 6.27 47 6.36 45 5.76 40 2.67 66 3.46 88 2.74 102 4.44 100 5.05 100 6.56 122 2.53 17 3.58 17 1.87 41 9.06 139 14.0 138 4.55 109 7.97 96 10.1 97 1.12 48
MDP-Flow [26]66.2 1.07 7 1.61 8 0.85 37 1.64 31 2.16 37 1.07 24 8.65 94 5.50 29 11.0 100 2.77 79 3.46 88 2.60 90 4.48 112 5.10 113 6.56 122 4.21 110 6.00 111 3.35 122 6.09 78 9.42 80 3.48 72 3.01 9 3.79 9 0.97 28
Modified CLG [34]66.7 1.53 41 2.41 45 0.86 41 2.31 130 2.77 127 2.09 137 8.71 96 5.58 30 11.3 103 2.49 49 2.77 31 2.86 109 2.49 34 2.82 34 1.25 20 3.46 61 4.92 62 2.05 52 3.05 8 4.70 8 1.76 20 11.1 159 14.1 159 1.11 45
FlowNet2 [122]67.0 2.26 121 3.38 112 0.95 94 1.76 55 2.22 47 1.21 40 7.48 79 11.5 109 8.91 79 2.43 40 3.27 70 2.58 86 2.80 40 3.18 40 2.97 65 3.28 45 4.68 45 2.06 54 5.65 59 8.74 59 4.55 109 3.42 17 4.31 17 1.70 125
CostFilter [40]67.3 1.26 19 1.96 20 0.89 63 1.47 11 1.93 14 0.94 3 12.8 137 18.8 148 15.4 139 2.20 26 2.95 42 1.51 13 3.13 45 3.56 46 4.90 81 3.74 76 5.32 75 1.69 27 7.46 121 11.5 120 5.27 131 6.10 69 7.71 70 1.65 120
COFM [59]67.8 1.36 26 1.97 21 0.88 52 1.59 23 2.08 27 1.23 46 9.49 106 15.4 132 12.1 110 3.08 104 4.19 129 1.60 21 2.21 28 2.50 28 2.01 57 3.92 91 5.58 92 2.13 61 6.58 93 10.2 95 5.93 138 2.85 7 3.60 7 1.77 132
PH-Flow [101]68.0 1.97 74 3.13 76 0.83 16 1.55 20 2.05 23 1.22 44 8.79 98 13.5 119 10.6 94 2.33 34 3.14 59 2.10 48 1.78 21 1.99 21 2.14 58 5.44 150 7.77 150 4.46 147 5.15 47 7.97 47 3.91 87 5.25 45 6.63 47 1.47 106
CombBMOF [113]69.0 2.14 110 3.23 97 1.81 155 1.72 44 2.27 58 1.00 14 6.89 63 10.3 97 7.99 69 3.11 107 4.11 123 3.41 138 2.47 32 2.80 33 1.50 55 3.58 65 5.11 67 2.05 52 5.17 48 8.00 48 3.64 76 3.18 15 4.01 15 1.24 76
NL-TV-NCC [25]70.5 1.44 32 2.20 31 0.99 107 2.05 96 2.61 101 1.54 90 5.02 25 6.98 59 4.96 28 3.01 98 4.07 120 2.41 69 2.62 36 2.96 36 3.97 74 4.71 132 6.72 132 3.40 124 3.62 16 5.58 16 2.25 28 8.25 106 10.4 104 1.01 32
TCOF [69]71.0 1.34 23 2.05 25 0.84 27 2.47 143 3.10 158 2.36 148 6.61 52 8.86 75 6.94 52 2.52 54 3.35 79 2.27 59 3.95 71 4.49 71 1.29 39 3.12 39 4.45 39 1.96 46 7.32 117 11.3 118 3.45 69 5.79 61 7.31 62 1.25 77
TV-L1-MCT [64]71.1 1.97 74 3.12 74 0.87 46 1.95 81 2.39 71 1.54 90 7.42 78 11.1 104 8.52 75 2.46 46 3.14 59 2.51 80 4.86 142 5.52 142 6.08 97 3.29 47 4.69 47 2.17 65 5.14 46 7.95 46 2.65 38 6.33 72 8.00 73 0.88 13
AdaConv-v1 [126]71.4 2.21 116 3.48 117 1.50 151 2.62 151 2.27 58 3.48 157 7.18 70 5.05 19 8.08 70 3.52 126 2.96 43 4.59 147 2.79 39 3.17 39 1.25 20 2.08 12 2.95 12 1.58 20 6.09 78 9.43 81 3.03 53 5.26 46 6.64 48 1.08 41
nLayers [57]72.5 1.97 74 3.14 81 0.84 27 1.53 17 1.99 20 1.13 30 15.8 149 22.3 153 18.5 150 2.70 69 3.66 98 1.96 39 4.43 98 5.04 98 6.34 107 3.84 88 5.48 89 2.21 69 5.26 51 8.14 51 1.78 21 2.78 4 3.51 4 2.09 152
CRTflow [80]72.9 1.83 63 2.87 62 1.01 113 2.30 126 2.90 142 2.26 142 5.96 43 6.95 57 5.40 32 2.45 45 3.05 50 2.28 62 4.40 93 5.01 94 6.50 112 3.19 42 4.55 43 1.67 24 6.35 86 9.83 86 2.60 35 7.70 89 9.73 90 0.91 18
PWC-Net_ROB [147]73.2 1.81 61 2.86 60 0.99 107 1.64 31 2.15 35 0.95 5 4.71 21 6.26 42 4.28 21 2.74 73 3.72 102 1.69 26 4.49 117 5.11 117 6.58 131 3.68 72 5.24 73 1.84 38 7.57 124 11.7 124 4.12 93 6.12 70 7.73 71 1.93 142
CPM-Flow [116]73.7 1.80 59 2.86 60 0.83 16 1.51 16 1.98 19 1.02 20 5.28 30 5.59 31 4.93 27 2.52 54 3.26 69 2.60 90 4.43 98 5.04 98 6.56 122 4.06 102 5.79 104 3.55 129 6.83 99 10.6 101 4.21 98 8.75 114 11.1 115 1.42 98
NNF-EAC [103]73.9 2.01 91 3.16 87 1.02 114 1.75 51 2.29 63 1.17 35 11.4 123 18.5 144 14.6 132 4.57 148 6.10 151 2.97 118 3.85 69 4.38 69 1.23 14 2.78 27 3.95 27 1.65 22 5.19 49 8.03 49 2.23 27 5.05 39 6.38 41 1.29 83
PMF [73]73.9 1.69 52 2.67 55 0.83 16 1.53 17 2.00 21 0.97 8 13.5 138 20.0 149 16.5 145 2.88 87 3.78 108 2.97 118 2.16 27 2.44 27 1.27 29 3.76 78 5.36 79 1.70 29 7.19 115 11.1 113 4.74 122 5.29 48 6.68 50 1.95 143
JOF [140]74.5 2.08 103 3.31 105 0.89 63 1.62 26 2.08 27 1.21 40 7.10 67 9.32 84 7.84 64 2.56 61 3.03 49 2.75 104 4.68 132 5.32 132 6.58 131 5.29 147 7.55 147 2.15 64 4.95 40 7.66 41 1.78 21 2.72 3 3.43 3 1.80 134
Classic++ [32]74.8 1.39 27 2.14 27 0.88 52 2.12 104 2.71 116 1.76 119 4.40 16 5.20 23 3.44 12 3.04 99 3.10 53 2.94 115 3.49 57 3.97 57 2.88 64 4.14 106 5.90 107 2.33 77 6.85 100 10.6 101 3.78 82 8.54 113 10.8 113 1.15 54
FlowFields [110]75.5 2.27 122 3.62 122 1.09 131 1.50 13 1.97 16 0.94 3 12.4 135 17.3 138 14.9 134 2.49 49 3.36 80 1.65 24 3.73 66 4.24 66 6.03 95 3.39 56 4.83 58 1.90 44 7.03 110 10.9 109 1.87 23 6.86 78 8.67 79 1.18 62
HAST [109]76.5 1.54 42 2.41 45 0.83 16 1.50 13 1.97 16 1.01 17 14.2 142 20.2 150 16.8 146 2.16 23 2.90 37 1.46 4 4.04 74 4.60 74 1.26 26 4.22 111 6.02 112 3.98 137 9.18 140 14.2 141 4.53 108 5.99 67 7.57 68 1.75 128
Ad-TV-NDC [36]76.7 2.61 140 4.13 142 1.16 139 2.45 140 2.78 129 2.34 147 4.25 10 6.07 40 3.78 16 3.58 129 4.45 138 2.51 80 1.87 22 2.10 22 1.33 51 3.36 53 4.79 54 2.42 82 3.41 12 5.27 13 1.54 13 9.04 120 11.4 120 0.97 28
Bartels [41]77.0 2.18 112 3.46 115 1.03 117 2.07 98 2.71 116 1.72 113 5.71 41 6.16 41 5.76 40 2.80 81 3.33 75 2.60 90 1.58 13 1.72 13 1.38 52 4.88 139 6.96 139 2.47 87 4.66 32 7.21 32 3.81 83 7.43 87 9.39 88 1.10 44
EAI-Flow [151]77.5 2.24 119 3.53 119 1.03 117 1.71 42 2.21 44 1.28 54 7.17 69 10.0 92 7.76 63 2.43 40 3.27 70 1.48 8 3.95 71 4.49 71 6.10 98 2.42 15 3.43 15 1.80 35 9.22 141 14.3 142 7.18 143 9.27 129 11.7 127 1.03 36
EPMNet [133]77.6 2.84 152 4.50 152 1.08 129 1.75 51 2.23 50 1.13 30 7.48 79 11.5 109 8.91 79 5.47 154 7.48 154 2.49 76 2.80 40 3.18 40 2.97 65 2.70 24 3.84 24 1.96 46 5.65 59 8.74 59 4.55 109 4.80 32 6.07 34 1.62 116
AugFNG_ROB [143]77.8 2.07 100 3.28 102 0.94 88 1.89 70 2.40 73 1.30 57 6.76 59 8.58 74 6.94 52 5.27 153 7.01 153 1.74 30 4.41 95 5.02 95 6.53 115 3.30 48 4.71 49 2.43 84 5.68 62 8.79 62 3.12 55 5.33 49 6.73 51 1.39 91
2DHMM-SAS [92]78.0 1.13 12 1.63 11 0.89 63 2.18 110 2.69 113 1.88 126 7.19 71 9.66 87 7.61 62 2.67 66 3.53 92 2.25 58 4.96 146 5.64 146 6.29 104 2.83 29 4.03 29 1.78 33 4.96 41 7.68 42 3.98 89 9.87 141 12.5 141 1.17 60
F-TV-L1 [15]78.5 2.41 130 3.82 131 0.97 101 2.33 131 2.89 140 1.91 129 4.72 22 6.38 46 4.58 25 2.49 49 2.90 37 2.98 121 2.25 29 2.54 29 1.30 41 2.73 25 3.88 25 2.40 81 6.37 88 9.85 88 3.69 79 11.1 159 14.1 159 0.92 20
Sparse Occlusion [54]79.0 2.21 116 3.52 118 0.93 85 2.12 104 2.73 120 1.46 81 4.67 20 6.57 50 4.98 30 2.16 23 2.85 34 1.75 31 5.06 149 5.76 149 6.58 131 2.64 21 3.74 21 2.23 71 6.21 83 9.61 84 3.46 71 9.82 140 12.4 140 0.95 23
EPPM w/o HM [88]79.1 1.72 56 2.70 57 1.27 146 1.65 34 2.18 41 1.30 57 14.6 144 15.2 129 13.1 118 2.76 76 3.69 100 2.07 45 1.98 23 2.22 23 1.28 34 3.63 67 5.17 69 2.42 82 8.88 137 13.7 137 8.97 147 5.63 58 7.11 59 1.17 60
ACK-Prior [27]79.2 1.52 40 2.37 40 1.10 132 1.90 73 2.51 89 1.04 22 12.1 132 8.87 76 14.9 134 2.53 57 3.02 46 2.32 66 4.80 138 5.46 138 6.63 150 4.03 99 5.74 100 2.68 97 5.80 65 8.97 65 4.01 91 3.02 10 3.80 10 1.01 32
Complementary OF [21]79.3 2.67 145 4.25 146 0.82 12 1.89 70 2.49 88 1.47 84 14.5 143 11.4 108 15.7 141 2.33 34 3.13 57 1.56 16 4.37 89 4.98 89 6.31 105 3.19 42 4.54 42 2.26 73 6.11 80 9.40 79 5.87 135 3.12 13 3.94 13 1.43 99
LSM [39]79.3 1.15 14 1.70 15 0.89 63 1.72 44 2.11 30 1.37 68 7.40 77 11.1 104 8.52 75 1.96 13 2.60 23 1.79 33 4.95 145 5.63 145 6.24 101 4.37 118 6.23 119 2.08 57 6.95 103 10.7 104 5.20 129 8.77 116 11.1 115 1.40 92
ComponentFusion [96]79.5 1.44 32 2.27 36 0.80 8 1.68 38 2.23 50 1.00 14 11.0 117 14.9 127 11.7 106 2.43 40 3.28 73 1.47 5 3.77 67 4.27 67 4.63 80 4.18 108 5.96 109 3.42 125 10.2 152 15.8 152 11.7 155 6.07 68 7.66 69 1.51 111
CNN-flow-warp+ref [117]81.0 1.79 58 2.83 58 0.87 46 2.11 101 2.69 113 1.69 109 5.64 40 6.46 49 6.31 45 3.11 107 2.91 40 2.85 108 4.52 123 5.14 123 6.58 131 3.34 52 4.75 52 2.76 103 6.52 92 10.1 91 2.88 48 8.42 107 10.6 108 1.07 39
SVFilterOh [111]81.0 2.18 112 3.45 114 0.92 79 1.66 35 2.19 43 1.22 44 15.7 148 22.4 154 18.5 150 2.48 48 3.32 74 2.35 67 4.46 108 5.08 109 6.58 131 3.66 69 5.21 71 2.37 80 5.40 56 8.36 56 1.54 13 3.87 23 4.89 24 1.83 136
FlowNetS+ft+v [112]81.6 2.75 150 4.38 150 0.92 79 3.05 156 2.83 134 3.29 155 4.97 24 3.66 12 4.97 29 1.97 14 2.34 10 2.24 57 4.78 136 5.44 136 6.58 131 2.93 34 4.17 34 1.74 32 6.15 81 9.52 82 2.63 36 9.27 129 11.7 127 0.98 30
S2F-IF [123]82.1 2.28 123 3.63 123 0.95 94 1.55 20 2.05 23 1.09 28 8.10 87 12.9 115 10.1 88 2.33 34 3.13 57 1.93 38 4.33 88 4.93 88 6.57 128 2.66 23 3.77 22 2.10 60 6.50 90 10.1 91 4.58 114 10.3 149 13.0 149 1.91 138
DPOF [18]82.6 2.79 151 4.46 151 1.92 157 1.36 8 1.77 8 1.01 17 10.5 114 7.45 63 13.5 120 3.60 132 4.88 145 2.92 114 3.40 53 3.86 53 1.23 14 3.10 38 4.42 38 2.06 54 5.73 64 8.87 64 2.76 41 10.2 146 12.8 146 1.40 92
LDOF [28]82.8 2.50 136 3.97 136 0.94 88 3.47 158 2.44 82 3.90 159 6.75 58 5.87 35 7.37 58 2.12 18 2.70 29 2.27 59 2.09 25 2.35 25 1.30 41 3.15 40 4.48 40 2.14 62 11.8 156 18.3 156 11.2 153 5.67 59 7.16 60 2.25 155
Layers++ [37]83.1 1.97 74 3.13 76 0.84 27 1.34 4 1.72 6 0.98 10 10.9 116 17.8 141 13.8 124 2.74 73 3.71 101 2.06 44 6.74 160 7.67 160 8.88 158 5.04 142 7.20 142 4.08 139 5.81 66 8.98 66 1.54 13 5.06 42 6.39 44 1.19 66
EpicFlow [102]83.9 2.07 100 3.30 103 0.83 16 1.75 51 2.29 63 1.23 46 5.47 38 4.72 18 4.44 24 5.04 152 6.87 152 3.09 125 4.46 108 5.08 109 6.56 122 2.81 28 4.00 28 2.00 49 9.31 143 14.4 144 5.04 125 8.21 103 10.4 104 1.18 62
3DFlow [135]83.9 1.21 18 1.82 18 0.88 52 1.70 39 2.21 44 1.37 68 8.30 90 13.6 120 10.7 96 3.22 115 4.38 137 2.71 101 3.25 50 3.69 49 4.37 78 6.56 158 9.36 158 6.71 158 4.79 33 7.41 33 2.22 26 10.2 146 12.9 147 1.27 79
Horn & Schunck [3]83.9 2.06 98 3.26 100 0.89 63 2.63 152 2.90 142 2.77 152 11.3 122 6.80 55 13.6 121 3.18 112 3.23 66 2.95 116 3.59 61 4.08 61 1.30 41 2.91 32 4.14 32 1.70 29 4.91 39 7.59 39 2.79 44 10.1 145 12.7 144 1.12 48
Brox et al. [5]84.4 2.86 153 4.57 153 0.85 37 1.89 70 2.48 86 1.41 73 5.02 25 5.06 20 5.06 31 2.74 73 2.48 17 2.20 53 2.57 35 2.91 35 1.28 34 4.77 135 6.81 135 4.16 140 17.2 161 26.6 161 21.6 161 8.51 112 10.8 113 0.88 13
ContinualFlow_ROB [152]84.5 2.11 106 3.35 109 1.05 122 1.73 48 2.24 52 1.21 40 7.58 81 11.3 107 9.11 81 3.18 112 4.33 135 1.75 31 4.45 104 5.06 104 6.54 118 3.76 78 5.36 79 1.67 24 7.32 117 11.3 118 3.22 60 5.04 36 6.37 39 1.75 128
Nguyen [33]85.0 2.62 142 4.16 143 0.94 88 2.35 133 2.74 124 1.78 121 5.37 33 5.59 31 4.85 26 2.76 76 2.97 44 2.96 117 3.62 62 4.12 62 1.25 20 4.16 107 5.94 108 3.86 133 6.94 102 10.7 104 3.64 76 7.20 81 9.10 82 0.96 25
Sparse-NonSparse [56]85.5 1.98 82 3.14 81 0.85 37 1.72 44 2.11 30 1.36 67 8.49 92 13.4 118 10.4 92 2.15 21 2.87 36 1.47 5 4.86 142 5.52 142 6.00 93 4.42 122 6.31 123 2.81 107 6.82 98 10.5 98 4.64 119 8.92 119 11.3 119 1.19 66
FESL [72]85.8 1.16 15 1.69 13 0.87 46 1.71 42 2.18 41 1.21 40 6.73 55 8.98 77 6.68 50 2.81 83 3.41 83 3.29 132 5.02 148 5.71 148 6.58 131 4.39 120 6.26 121 2.29 75 6.36 87 9.84 87 4.42 105 9.75 137 12.3 137 1.31 85
Local-TV-L1 [65]86.9 1.56 45 2.40 43 0.98 105 2.43 138 2.91 145 2.04 135 4.29 14 5.30 26 3.22 10 2.27 29 2.65 25 2.08 46 5.34 150 6.08 151 6.63 150 3.66 69 5.02 66 2.69 98 7.43 120 11.5 120 4.10 92 7.36 84 9.29 85 1.92 139
Fusion [6]87.3 2.25 120 3.58 121 1.33 147 1.95 81 2.53 91 1.23 46 8.48 91 5.32 27 10.5 93 2.89 88 3.75 104 1.67 25 4.65 131 5.29 131 3.65 72 5.17 145 7.37 145 4.89 150 5.87 70 9.09 70 2.49 31 3.09 12 3.89 12 1.40 92
SRR-TVOF-NL [91]87.3 1.49 37 2.29 37 0.95 94 1.85 68 2.40 73 1.41 73 14.1 141 20.2 150 16.3 143 2.54 59 3.41 83 1.81 34 4.52 123 5.14 123 6.60 148 4.49 125 6.40 126 1.88 42 5.35 52 8.28 52 3.57 75 5.81 63 7.34 64 1.50 110
DF-Auto [115]87.8 2.69 147 4.30 149 1.12 135 2.13 106 2.69 113 1.58 96 6.20 45 8.02 70 5.50 35 2.68 68 3.34 76 2.70 100 1.59 14 1.76 14 1.28 34 3.52 63 5.01 64 1.68 26 8.36 132 12.9 132 10.1 148 8.45 109 10.7 110 1.67 122
ResPWCR_ROB [144]88.0 1.84 65 2.91 65 0.89 63 2.01 90 2.64 105 1.41 73 6.21 46 9.00 78 7.24 56 3.64 134 4.96 147 1.70 28 4.48 112 5.10 113 6.59 142 3.65 68 5.20 70 1.71 31 7.79 128 12.1 129 3.88 85 7.25 82 9.15 83 1.65 120
TI-DOFE [24]88.4 2.32 125 3.67 124 1.10 132 2.56 150 2.92 146 2.11 138 4.25 10 4.63 17 4.31 23 3.59 130 3.76 107 2.86 109 3.24 49 3.69 49 1.32 47 4.55 126 6.50 128 3.36 123 4.41 26 6.81 26 2.76 41 9.18 125 11.6 125 1.11 45
DMF_ROB [139]89.1 2.10 105 3.33 108 0.89 63 2.10 99 2.72 119 1.75 118 6.74 57 7.22 60 6.53 47 2.98 95 3.27 70 2.28 62 4.44 100 5.05 100 6.58 131 4.76 134 6.80 134 3.43 126 5.72 63 8.85 63 2.52 32 7.74 90 9.78 91 1.22 72
TriangleFlow [30]89.6 1.80 59 2.83 58 0.95 94 2.18 110 2.77 127 1.61 101 7.36 75 9.16 80 8.83 78 2.70 69 3.56 94 2.88 112 3.17 47 3.60 47 1.27 29 5.17 145 7.38 146 5.34 153 7.58 125 11.7 124 5.87 135 5.12 44 6.45 46 1.14 53
Black & Anandan [4]89.7 2.44 131 3.86 132 1.02 114 2.49 146 2.93 147 1.98 131 13.5 138 7.92 67 14.3 127 3.14 110 3.16 61 2.55 83 3.13 45 3.55 45 1.27 29 3.51 62 5.01 64 2.20 67 5.13 45 7.94 45 3.20 59 7.26 83 9.17 84 1.89 137
Ramp [62]90.2 1.99 86 3.16 87 0.84 27 1.76 55 2.17 40 1.33 63 12.3 133 18.7 147 15.2 137 2.41 39 3.25 67 2.12 49 4.79 137 5.45 137 6.02 94 4.46 123 6.36 124 2.80 105 5.82 67 9.00 67 3.35 65 8.47 111 10.7 110 1.41 95
StereoOF-V1MT [119]90.4 1.70 54 2.65 52 0.87 46 2.10 99 2.64 105 1.74 117 9.86 110 4.53 16 12.2 111 4.12 143 4.14 126 4.98 150 4.38 90 4.98 89 6.21 99 3.56 64 4.72 50 3.00 114 7.15 112 11.1 113 4.77 123 6.63 75 8.37 76 1.02 35
AGIF+OF [85]90.5 1.61 48 2.52 49 0.82 12 1.79 62 2.25 54 1.40 72 7.38 76 11.2 106 9.13 82 2.97 92 3.75 104 3.12 128 4.51 121 5.13 121 6.54 118 3.89 90 5.54 91 2.34 79 5.85 68 9.05 68 4.40 103 10.8 157 13.6 157 1.55 113
Occlusion-TV-L1 [63]90.6 2.12 107 3.37 111 1.25 145 2.30 126 2.90 142 1.91 129 4.07 9 5.27 24 3.82 17 3.60 132 4.67 142 3.14 130 3.03 43 3.45 43 1.24 17 3.73 74 5.32 75 2.95 112 6.99 107 9.92 89 4.22 99 9.25 127 11.7 127 1.12 48
Shiralkar [42]90.9 2.18 112 3.46 115 0.88 52 2.25 119 2.71 116 1.88 126 7.65 82 6.40 47 9.36 86 3.87 141 4.79 143 2.75 104 3.52 59 4.00 59 3.60 70 3.75 77 5.34 78 2.74 101 6.85 100 10.6 101 2.98 52 7.97 96 10.1 97 1.12 48
Filter Flow [19]91.1 2.67 145 4.25 146 0.94 88 2.40 135 2.89 140 1.70 112 7.19 71 10.2 96 8.54 77 3.40 122 3.60 95 3.38 135 2.09 25 2.36 26 1.32 47 3.73 74 5.32 75 2.69 98 5.67 61 8.77 61 4.39 102 7.91 94 9.99 95 1.19 66
H+S_ROB [137]91.1 1.63 50 2.54 50 0.88 52 2.11 101 2.55 94 1.59 98 21.2 161 27.9 160 23.2 160 3.30 118 3.34 76 3.39 136 3.88 70 4.41 70 1.48 54 4.02 98 5.74 100 2.48 88 9.48 146 14.7 146 3.14 58 5.26 46 6.64 48 0.81 6
2D-CLG [1]92.2 2.03 96 3.22 96 0.89 63 2.24 117 2.67 109 1.80 122 7.75 83 4.01 15 9.13 82 3.05 100 2.78 32 3.46 139 6.30 159 7.17 159 9.00 159 2.62 19 3.71 19 2.33 77 6.41 89 9.92 89 3.52 74 9.35 131 11.8 131 1.13 52
Classic+CPF [83]92.2 1.98 82 3.14 81 0.86 41 1.82 65 2.26 56 1.47 84 7.22 73 10.0 92 7.55 61 2.31 33 3.11 56 1.44 3 4.80 138 5.46 138 6.56 122 4.55 126 6.48 127 2.85 110 6.73 94 10.4 96 4.62 118 10.4 150 13.2 150 1.63 117
BlockOverlap [61]92.3 2.06 98 3.26 100 1.03 117 2.22 114 2.67 109 1.85 125 8.84 99 6.32 43 11.0 100 4.26 146 3.93 115 5.84 151 4.23 81 4.81 81 1.31 46 4.01 97 5.71 98 3.22 117 4.98 43 7.64 40 4.44 106 4.57 30 5.77 32 1.72 127
Adaptive [20]93.0 2.51 137 3.99 137 0.93 85 2.42 136 3.03 152 2.24 141 6.42 49 9.95 91 7.88 65 2.89 88 3.63 96 3.26 131 4.30 85 4.89 85 1.26 26 4.07 104 5.81 105 3.25 118 6.07 77 9.39 78 2.94 51 7.57 88 9.57 89 0.91 18
OFRF [134]93.1 2.13 109 3.31 105 1.06 124 2.25 119 2.73 120 1.77 120 6.95 66 10.5 101 8.46 73 2.85 84 3.85 110 1.57 18 4.08 75 4.64 75 5.33 85 2.88 30 4.08 30 2.01 50 8.13 130 12.6 130 4.58 114 9.40 134 11.9 132 1.44 101
ROF-ND [107]93.5 1.28 20 1.94 19 0.83 16 3.07 157 2.65 108 4.13 160 7.93 84 11.9 112 9.17 84 3.22 115 4.28 133 2.58 86 4.29 84 4.88 84 6.50 112 4.98 141 7.10 141 4.47 148 6.01 75 9.30 76 3.24 61 6.57 74 8.29 75 1.27 79
Steered-L1 [118]93.6 1.89 67 3.01 68 0.89 63 1.93 79 2.51 89 1.50 86 17.3 153 23.1 156 20.6 156 3.65 136 4.27 132 3.39 136 3.51 58 3.99 58 3.99 75 3.32 50 4.73 51 1.95 45 7.20 116 11.1 113 4.57 112 8.42 107 10.6 108 1.01 32
PGAM+LK [55]93.8 2.12 107 3.32 107 1.16 139 2.33 131 2.79 130 2.01 133 18.8 157 29.3 161 23.5 161 4.83 150 3.93 115 6.42 153 2.76 38 3.13 38 1.47 53 3.40 57 4.83 58 2.49 89 3.41 12 5.25 12 2.90 49 6.38 73 8.05 74 1.15 54
Efficient-NL [60]94.1 1.57 46 2.46 47 0.84 27 1.97 87 2.43 78 1.46 81 11.2 120 7.97 68 14.4 128 2.61 65 3.50 91 2.08 46 4.84 141 5.51 141 6.23 100 4.00 95 5.70 95 2.09 59 7.55 123 11.7 124 4.15 95 10.5 152 13.2 150 1.43 99
TriFlow [95]94.2 2.63 144 4.19 145 1.07 127 2.05 96 2.61 101 1.57 93 6.62 53 10.1 94 7.49 60 2.70 69 3.10 53 2.86 109 4.46 108 5.07 108 6.53 115 3.72 73 5.30 74 1.88 42 5.90 71 9.13 71 4.57 112 9.77 139 12.3 137 1.19 66
Heeger++ [104]94.8 3.50 159 5.53 159 2.53 158 2.00 89 2.45 84 1.37 68 8.20 88 5.88 36 9.23 85 3.50 124 3.20 64 3.83 143 4.31 86 4.90 86 6.32 106 3.45 60 4.82 57 2.97 113 7.84 129 10.1 91 6.00 139 3.91 25 4.93 26 1.44 101
FFV1MT [106]95.0 2.33 127 3.68 125 1.04 121 2.15 108 2.53 91 1.51 88 8.99 101 9.89 89 10.6 94 3.50 124 3.20 64 3.83 143 3.19 48 3.62 48 3.04 68 4.61 129 6.56 129 3.72 132 6.78 96 10.5 98 5.04 125 2.81 5 3.54 5 1.67 122
RNLOD-Flow [121]95.2 1.97 74 3.13 76 0.84 27 1.90 73 2.43 78 1.46 81 7.30 74 10.8 103 8.46 73 2.51 53 3.02 46 3.06 124 4.50 118 5.12 118 6.48 109 5.34 148 7.62 148 4.17 141 6.16 82 9.53 83 3.27 62 10.6 154 13.4 154 1.31 85
RFlow [90]95.5 1.17 16 1.77 17 0.88 52 2.16 109 2.68 111 1.68 108 10.8 115 15.8 134 12.8 116 3.06 102 3.81 109 3.10 126 4.14 77 4.71 77 1.30 41 4.00 95 5.70 95 2.75 102 6.80 97 10.5 98 3.39 67 10.0 144 12.7 144 1.92 139
LFNet_ROB [149]96.0 2.48 135 3.95 135 1.40 149 1.76 55 2.27 58 1.15 33 6.90 64 9.74 88 7.46 59 2.18 25 2.90 37 1.57 18 4.44 100 5.05 100 6.59 142 4.30 116 6.13 117 3.67 131 9.25 142 14.3 142 4.87 124 7.88 93 9.96 94 2.01 146
FlowFields+ [130]96.3 2.34 128 3.73 128 1.81 155 1.50 13 1.97 16 1.00 14 12.7 136 17.9 143 15.4 139 3.08 104 4.14 126 2.41 69 4.44 100 5.05 100 6.58 131 4.11 105 5.87 106 2.06 54 8.38 133 13.0 133 5.69 134 5.43 52 6.86 54 1.07 39
IAOF2 [51]96.5 2.09 104 3.25 99 1.11 134 2.25 119 2.81 132 1.69 109 6.45 50 8.52 72 6.67 49 2.44 43 3.02 46 2.49 76 4.62 130 5.25 130 5.71 87 3.94 93 5.61 94 3.18 116 5.86 69 9.06 69 4.66 120 9.21 126 11.6 125 1.69 124
Classic+NL [31]97.3 1.97 74 3.13 76 0.88 52 1.78 60 2.22 47 1.44 77 11.8 129 17.8 141 14.5 131 2.47 47 3.34 76 2.43 71 4.81 140 5.47 140 5.90 91 4.41 121 6.29 122 2.58 91 7.03 110 10.9 109 5.12 128 9.04 120 11.4 120 1.18 62
LiteFlowNet [142]97.3 2.00 89 3.18 90 1.08 129 1.70 39 2.22 47 1.35 66 9.21 102 13.9 121 10.9 99 6.29 155 8.38 155 8.10 156 4.45 104 5.06 104 6.59 142 3.80 84 5.41 84 2.44 86 6.50 90 10.1 91 2.76 41 5.55 57 7.01 58 2.01 146
SLK [47]97.9 1.66 51 2.59 51 0.99 107 2.25 119 2.58 99 1.73 115 13.8 140 9.17 81 15.3 138 4.21 144 5.03 149 4.67 148 4.31 86 4.90 86 4.35 77 3.88 89 5.52 90 2.78 104 9.91 149 15.3 149 3.29 63 4.40 29 5.55 31 1.15 54
BriefMatch [124]98.0 1.45 36 2.23 32 1.06 124 2.03 92 2.62 104 2.01 133 11.4 123 6.57 50 14.0 125 4.56 147 3.92 114 6.24 152 3.37 52 3.80 52 2.51 60 4.47 124 6.37 125 2.80 105 6.03 76 9.28 75 7.81 145 8.90 118 5.08 27 13.3 161
LocallyOriented [52]98.2 1.99 86 3.14 81 0.92 79 2.22 114 2.68 111 1.60 99 12.0 131 15.6 133 14.9 134 4.67 149 5.61 150 2.20 53 4.22 79 4.80 79 2.99 67 3.66 69 5.22 72 2.20 67 5.95 73 9.21 73 3.13 57 10.7 156 13.5 156 1.38 90
FC-2Layers-FF [74]99.1 1.91 68 2.98 67 0.95 94 1.34 4 1.73 7 1.04 22 11.8 129 17.6 140 14.4 128 3.36 120 4.58 140 1.84 35 5.00 147 5.69 147 6.49 111 3.93 92 5.60 93 2.61 93 7.47 122 11.6 123 4.60 117 9.38 132 11.9 132 1.58 115
FF++_ROB [145]99.2 2.31 124 3.68 125 0.84 27 1.77 59 2.27 58 1.16 34 11.6 126 15.2 129 12.4 113 3.39 121 4.56 139 2.52 82 4.45 104 5.06 104 6.57 128 3.16 41 4.50 41 2.17 65 7.41 119 11.5 120 7.10 142 9.04 120 11.4 120 1.92 139
S2D-Matching [84]99.7 1.98 82 3.14 81 0.90 74 2.02 91 2.55 94 1.63 104 10.3 113 15.0 128 11.9 108 2.76 76 3.74 103 2.15 51 4.02 73 4.57 73 5.30 84 5.99 154 8.55 154 5.16 151 5.45 57 8.43 57 4.59 116 9.75 137 12.3 137 1.41 95
HBM-GC [105]99.9 1.99 86 3.16 87 0.97 101 1.98 88 2.60 100 1.26 52 9.23 103 8.07 71 11.4 104 3.08 104 4.11 123 2.59 88 4.61 129 5.24 129 6.64 152 4.79 136 6.83 136 3.97 136 6.97 106 10.8 106 4.41 104 2.84 6 3.58 6 1.98 145
ProbFlowFields [128]100.0 2.07 100 3.30 103 0.86 41 1.63 28 2.15 35 1.08 26 15.1 146 21.3 152 17.8 148 2.85 84 3.87 112 2.56 84 4.41 95 5.02 95 6.55 120 4.37 118 6.23 119 3.86 133 8.64 134 13.4 134 11.3 154 5.51 55 6.96 56 1.75 128
AggregFlow [97]100.2 2.69 147 4.28 148 0.86 41 1.76 55 2.31 65 1.34 64 11.2 120 16.4 135 13.7 122 3.22 115 4.29 134 2.65 96 2.08 24 2.34 24 2.34 59 3.26 44 4.65 44 2.14 62 12.4 157 19.2 157 15.8 159 9.95 142 12.6 142 2.03 149
TV-L1-improved [17]100.3 1.54 42 2.40 43 1.00 111 2.46 142 3.09 156 2.33 145 11.4 123 6.97 58 14.4 128 2.35 37 2.50 18 2.44 74 4.48 112 5.09 111 1.32 47 4.23 112 6.03 113 3.29 120 7.02 108 10.9 109 3.93 88 9.39 133 11.9 132 2.01 146
TVL1_ROB [138]105.0 3.04 155 4.84 156 0.98 105 2.47 143 2.94 148 1.82 123 8.00 85 11.8 111 9.68 87 3.34 119 4.01 118 2.49 76 2.71 37 3.08 37 1.30 41 4.30 116 6.14 118 2.83 109 9.33 144 14.0 138 12.4 156 7.85 92 9.91 93 0.88 13
Aniso-Texture [82]105.5 1.39 27 2.16 28 0.84 27 2.48 145 3.09 156 2.36 148 6.94 65 10.3 97 8.32 72 8.54 159 10.4 159 11.9 159 4.54 125 5.16 125 6.60 148 5.09 143 7.25 143 4.60 149 6.96 104 10.8 106 3.68 78 5.80 62 7.32 63 1.11 45
Adaptive flow [45]105.8 2.69 147 4.00 139 1.20 142 2.49 146 2.95 151 1.99 132 8.50 93 9.02 79 10.8 98 3.93 142 4.14 126 4.77 149 5.90 155 6.71 155 5.77 88 4.85 138 6.91 138 4.19 142 4.88 36 7.54 35 3.70 80 2.88 8 3.64 8 0.86 11
UnFlow [129]105.9 2.61 140 4.16 143 1.12 135 1.95 81 2.40 73 1.57 93 6.73 55 9.90 90 7.88 65 2.38 38 3.19 63 1.69 26 4.55 127 5.17 127 6.37 108 5.77 153 8.23 153 5.32 152 7.15 112 11.1 113 2.87 47 11.3 161 14.3 161 1.70 125
SILK [79]106.4 1.72 56 2.69 56 1.00 111 2.80 154 2.87 138 3.18 153 19.4 158 18.6 146 18.8 154 3.49 123 4.07 120 3.78 142 3.44 54 3.90 54 2.57 61 4.69 131 6.68 131 2.82 108 3.07 9 4.72 9 3.31 64 10.9 158 13.8 158 1.45 105
Rannacher [23]106.9 2.22 118 3.54 120 0.92 79 2.45 140 3.08 155 2.39 151 11.7 127 9.26 83 14.6 132 2.97 92 3.88 113 2.44 74 3.71 65 4.22 65 1.29 39 4.68 130 6.67 130 3.29 120 7.02 108 10.9 109 3.72 81 8.05 100 10.2 101 1.78 133
Correlation Flow [75]107.7 2.01 91 3.18 90 0.91 77 2.18 110 2.73 120 1.60 99 8.77 97 12.9 115 10.3 91 3.00 96 4.07 120 2.00 41 3.52 59 4.00 59 4.02 76 6.31 156 9.00 156 6.21 156 7.72 127 11.9 128 5.07 127 8.86 117 11.2 118 2.61 158
GraphCuts [14]108.0 2.47 134 3.90 133 1.03 117 1.91 77 2.43 78 1.56 92 11.0 117 6.33 44 13.7 122 2.97 92 3.47 90 3.11 127 5.47 152 6.22 152 7.89 155 3.33 51 4.75 52 2.03 51 8.66 135 13.4 134 6.07 140 9.96 143 12.6 142 1.18 62
TF+OM [100]108.5 1.95 72 3.07 70 1.54 152 1.73 48 2.25 54 1.28 54 6.86 62 10.4 100 7.88 65 3.72 140 4.88 145 2.97 118 5.54 153 6.30 153 8.26 157 4.18 108 5.97 110 2.90 111 7.63 126 11.8 127 5.25 130 8.05 100 10.2 101 2.03 149
Dynamic MRF [7]108.6 2.00 89 3.18 90 0.88 52 2.26 123 2.87 138 2.30 144 6.85 61 6.45 48 7.97 68 3.65 136 4.22 130 4.17 146 4.47 111 5.09 111 6.53 115 4.03 99 5.73 99 3.48 127 8.15 131 12.6 130 4.25 100 9.08 123 11.5 123 1.53 112
FOLKI [16]108.8 1.98 82 3.08 72 1.24 144 2.52 149 2.83 134 2.33 145 10.1 111 8.57 73 12.8 116 4.91 151 4.36 136 6.54 154 3.10 44 3.52 44 3.73 73 8.35 159 11.9 159 9.26 160 4.40 25 6.80 25 5.91 137 9.08 123 11.5 123 1.22 72
HCIC-L [99]110.5 2.19 115 3.36 110 1.38 148 1.90 73 2.28 62 1.62 103 15.4 147 22.7 155 18.7 152 3.66 138 4.96 147 2.28 62 2.47 32 2.79 32 1.28 34 5.47 151 7.81 151 5.56 154 10.4 153 16.2 153 10.4 150 5.04 36 6.36 38 2.40 157
Learning Flow [11]111.7 2.38 129 3.78 130 0.95 94 2.24 117 2.80 131 1.63 104 20.4 160 24.8 158 20.8 157 3.05 100 3.10 53 2.35 67 4.76 135 5.41 135 5.57 86 3.79 82 5.41 84 3.26 119 5.92 72 9.16 72 3.90 86 10.6 154 13.4 154 1.44 101
IAOF [50]111.7 3.34 157 5.18 157 3.47 160 2.75 153 3.20 160 2.07 136 11.7 127 16.6 136 14.1 126 3.56 127 3.67 99 3.65 140 3.65 63 4.16 64 1.24 17 3.82 85 5.45 86 2.66 96 6.34 85 9.81 85 3.45 69 9.42 135 11.9 132 1.31 85
SegOF [10]112.7 2.46 133 3.92 134 0.97 101 1.93 79 2.42 77 1.45 79 14.8 145 12.5 114 16.0 142 6.84 156 9.29 156 6.71 155 4.69 133 5.34 133 6.57 128 4.06 102 5.78 103 2.56 90 9.69 148 15.0 148 7.78 144 3.03 11 3.82 11 1.29 83
SimpleFlow [49]113.1 2.01 91 3.19 93 0.86 41 2.04 93 2.57 98 1.51 88 19.8 159 24.0 157 21.5 158 3.07 103 3.85 110 3.31 133 4.93 144 5.61 144 6.26 102 5.48 152 7.81 151 3.88 135 9.39 145 14.5 145 10.2 149 3.62 19 4.57 19 1.31 85
WRT [150]117.7 2.02 94 3.19 93 1.02 114 2.82 155 2.56 97 3.75 158 17.1 151 17.2 137 18.0 149 4.24 145 4.11 123 2.43 71 4.39 92 4.99 92 5.22 83 4.93 140 7.03 140 4.39 145 10.1 151 15.7 151 8.96 146 4.13 26 5.21 28 1.95 143
StereoFlow [44]118.4 2.98 154 4.65 154 1.15 137 2.26 123 2.74 124 1.61 101 6.11 44 7.37 61 6.48 46 3.16 111 4.04 119 2.76 106 6.10 158 6.94 158 8.09 156 5.13 144 7.31 144 4.06 138 9.04 138 14.0 138 4.66 120 6.69 76 8.45 77 1.56 114
IIOF-NLDP [131]118.8 2.02 94 3.20 95 0.90 74 2.04 93 2.61 101 1.43 76 10.1 111 14.7 125 11.7 106 3.13 109 4.25 131 2.43 71 6.02 157 6.85 157 9.58 160 6.28 155 8.96 155 6.18 155 10.9 154 16.8 154 10.6 152 7.37 85 9.31 86 1.41 95
HBpMotionGpu [43]119.4 2.51 137 3.99 137 1.40 149 2.50 148 3.10 158 2.21 140 9.83 109 14.5 124 11.5 105 3.59 130 4.62 141 2.59 88 7.38 161 8.39 161 11.8 161 5.43 149 7.75 149 4.40 146 5.39 55 8.33 54 2.06 25 5.92 66 7.48 67 1.47 106
WOLF_ROB [148]119.5 3.29 156 4.72 155 1.64 154 2.43 138 2.94 148 1.88 126 11.1 119 15.3 131 13.4 119 3.64 134 4.86 144 2.64 95 4.48 112 5.10 113 6.51 114 3.77 80 5.37 81 2.63 94 9.92 150 15.3 149 5.33 132 5.40 51 6.82 53 1.64 119
GroupFlow [9]121.4 2.44 131 3.75 129 1.61 153 2.04 93 2.53 91 1.67 107 12.3 133 10.6 102 12.5 114 9.06 160 10.6 160 12.0 160 4.69 133 5.34 133 6.59 142 4.72 133 6.72 132 4.23 144 8.80 136 13.6 136 4.48 107 3.88 24 4.90 25 1.81 135
Pyramid LK [2]122.6 2.62 142 4.11 140 1.23 143 3.77 159 2.94 148 2.29 143 17.2 152 10.1 94 16.4 144 7.25 157 9.29 156 8.44 157 5.34 150 6.07 150 3.60 70 4.58 128 4.97 63 3.49 128 9.56 147 14.8 147 4.29 101 3.73 21 4.70 21 1.28 82
SPSA-learn [13]122.8 3.49 158 5.31 158 1.15 137 2.30 126 2.81 132 1.73 115 16.0 150 14.3 123 16.8 146 3.70 139 3.65 97 4.08 145 4.54 125 5.16 125 5.09 82 3.82 85 5.45 86 3.56 130 16.4 160 25.3 160 19.6 160 4.18 27 5.27 29 2.10 153
2bit-BM-tele [98]131.2 2.60 139 4.11 140 1.19 141 2.42 136 3.07 154 2.37 150 18.2 156 27.8 159 23.0 159 3.19 114 3.75 104 3.71 141 4.50 118 5.12 118 5.98 92 6.47 157 9.23 157 6.38 157 14.3 158 22.1 158 15.2 158 3.15 14 3.97 14 2.39 156
AVG_FLOW_ROB [141]147.6 8.35 161 7.59 160 5.23 161 5.53 161 5.85 161 5.12 161 18.0 155 18.5 144 18.9 155 9.93 161 13.2 161 12.4 161 5.79 154 6.59 154 6.79 153 14.6 161 20.8 161 12.4 161 11.3 155 17.1 155 10.5 151 5.49 54 4.72 22 5.67 160
Periodicity [78]148.6 5.15 160 7.86 161 3.24 159 5.39 160 3.06 153 3.39 156 17.5 154 17.3 138 18.7 152 7.38 158 9.91 158 8.96 158 5.97 156 6.80 156 6.94 154 8.36 160 11.9 159 8.85 159 15.0 159 23.2 159 14.1 157 5.71 60 7.19 61 4.90 159
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.