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        
R10.0
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
PMMST [114]12.2 1.35 15 4.72 16 0.01 4 2.31 21 6.03 14 0.07 5 1.11 4 3.99 9 0.12 4 3.07 3 6.79 6 0.40 15 13.6 10 22.1 7 1.00 19 4.53 9 18.6 9 0.11 12 4.33 11 24.1 12 0.07 15 8.94 31 22.6 30 0.19 12
CyclicGen [154]15.2 0.55 1 1.82 1 0.01 4 1.48 2 3.28 1 0.45 126 0.94 3 2.65 4 0.17 29 3.61 40 6.73 4 1.04 139 9.11 1 14.9 1 0.60 1 1.58 1 6.55 1 0.03 1 2.59 1 14.4 1 0.04 1 3.82 1 9.70 1 0.11 1
CtxSyn [137]15.7 0.83 2 2.64 2 0.01 4 1.46 1 3.91 2 0.12 31 0.89 2 1.58 1 0.21 62 1.91 1 4.91 1 0.47 62 13.3 5 21.6 5 0.88 6 4.13 5 16.8 5 0.05 3 5.12 123 24.8 34 0.06 5 7.83 4 19.5 4 0.17 6
MDP-Flow2 [68]16.0 1.29 8 4.46 10 0.01 4 2.18 12 6.04 15 0.06 2 1.11 4 4.13 13 0.14 11 3.08 4 6.86 8 0.36 5 13.5 6 22.1 7 1.01 29 4.96 41 20.4 40 0.18 42 4.31 9 24.0 10 0.07 15 8.94 31 22.6 30 0.20 29
SuperSlomo [132]16.0 0.91 3 2.94 3 0.01 4 2.72 45 6.28 18 0.48 130 0.71 1 1.99 2 0.12 4 3.19 6 6.15 2 0.81 128 12.7 3 20.8 3 0.74 4 3.92 3 16.2 3 0.04 2 4.20 5 22.7 5 0.05 2 7.68 2 19.4 2 0.15 5
NN-field [71]27.1 1.45 34 5.41 50 0.01 4 1.87 6 5.01 6 0.06 2 1.51 76 3.94 7 0.16 23 3.78 63 8.92 86 0.41 24 13.6 10 22.2 12 1.00 19 5.05 56 20.7 53 0.20 51 4.27 6 23.7 6 0.07 15 8.85 16 22.4 14 0.19 12
NNF-Local [87]27.8 1.41 22 5.16 34 0.01 4 1.85 4 5.00 5 0.07 5 1.12 6 4.05 10 0.14 11 3.68 50 8.61 73 0.42 29 13.6 10 22.3 14 1.00 19 5.21 76 21.4 77 0.24 72 4.35 13 24.1 12 0.11 80 8.85 16 22.4 14 0.19 12
DeepFlow [86]29.8 1.35 15 4.67 15 0.00 1 3.01 67 8.09 65 0.21 63 1.26 17 5.00 45 0.12 4 3.97 87 8.00 46 0.44 45 13.7 17 22.4 20 1.03 52 4.46 7 18.3 7 0.08 6 4.40 19 24.5 23 0.06 5 8.74 12 22.1 12 0.21 66
SepConv-v1 [127]31.5 0.93 4 3.75 4 0.02 72 2.74 47 6.52 27 0.46 128 1.13 7 2.38 3 0.42 120 3.64 49 7.10 13 0.94 136 13.6 10 22.1 7 0.82 5 4.01 4 16.5 4 0.05 3 4.02 3 22.3 3 0.12 94 8.05 5 20.3 5 0.12 2
SuperFlow [81]31.8 1.30 9 4.40 8 0.01 4 3.28 84 8.24 73 0.30 89 1.46 64 4.37 16 0.23 72 3.62 43 7.44 23 0.46 55 13.7 17 22.4 20 0.99 14 4.56 10 18.7 10 0.08 6 4.44 27 24.7 30 0.09 54 8.83 14 22.4 14 0.17 6
DeepFlow2 [108]31.9 1.40 21 4.87 20 0.01 4 3.01 67 8.13 67 0.20 59 1.25 16 5.01 48 0.11 2 3.83 73 8.21 57 0.43 38 13.7 17 22.4 20 1.03 52 4.49 8 18.5 8 0.08 6 4.49 37 24.8 34 0.06 5 8.87 19 22.5 21 0.21 66
DF-Auto [115]32.5 1.26 6 3.92 5 0.00 1 3.13 75 7.98 60 0.29 84 1.14 8 4.08 12 0.16 23 3.57 34 7.48 26 0.46 55 13.5 6 22.0 6 1.01 29 4.66 11 19.2 12 0.15 25 4.49 37 24.8 34 0.09 54 9.09 54 23.0 56 0.21 66
IROF++ [58]34.2 1.54 63 5.77 77 0.01 4 2.35 24 6.39 20 0.10 22 1.50 73 5.04 50 0.25 83 3.10 5 6.75 5 0.39 10 13.7 17 22.4 20 1.16 93 4.66 11 19.2 12 0.11 12 4.55 58 25.1 53 0.10 67 8.84 15 22.4 14 0.19 12
NNF-EAC [103]34.3 1.48 39 4.99 26 0.03 128 2.49 34 6.76 35 0.08 11 1.56 82 4.39 17 0.24 79 3.28 10 7.08 11 0.39 10 13.7 17 22.3 14 1.01 29 4.66 11 19.1 11 0.12 16 4.41 22 24.5 23 0.11 80 9.02 44 22.8 45 0.20 29
PH-Flow [101]35.6 1.54 63 5.76 76 0.01 4 1.93 7 5.29 7 0.08 11 1.17 9 4.41 18 0.17 29 3.05 2 6.71 3 0.39 10 13.6 10 22.3 14 1.01 29 5.35 94 21.9 94 0.34 115 4.36 16 24.3 16 0.10 67 8.92 29 22.6 30 0.22 101
FGIK [136]35.9 1.16 5 4.53 11 0.01 4 3.09 72 7.12 43 0.48 130 1.54 81 2.89 5 0.29 104 4.86 125 9.21 102 0.97 138 9.84 2 16.2 2 0.73 3 3.20 2 13.2 2 0.05 3 3.37 2 18.8 2 0.07 15 7.68 2 19.4 2 0.14 4
Local-TV-L1 [65]37.4 1.34 13 4.39 7 0.02 72 4.19 115 9.42 103 0.37 106 1.34 32 4.34 15 0.15 17 3.61 40 7.78 40 0.44 45 13.8 32 22.5 30 1.04 56 4.71 17 19.5 19 0.16 33 4.40 19 24.5 23 0.07 15 8.70 11 22.0 9 0.20 29
nLayers [57]38.5 1.51 50 5.46 53 0.01 4 2.17 11 5.91 12 0.08 11 1.24 15 3.88 6 0.18 38 3.47 30 7.71 38 0.40 15 13.9 52 22.7 64 1.18 101 5.25 78 21.6 83 0.31 98 4.35 13 23.8 7 0.09 54 8.99 40 22.7 39 0.19 12
Brox et al. [5]38.7 1.43 29 4.82 18 0.01 4 2.95 58 7.80 55 0.18 50 1.40 50 5.17 62 0.16 23 3.73 55 7.70 37 0.45 49 13.7 17 22.4 20 1.00 19 5.04 54 20.6 50 0.27 84 4.52 42 25.1 53 0.09 54 8.87 19 22.4 14 0.19 12
Layers++ [37]38.9 1.44 32 5.17 35 0.01 4 1.83 3 4.87 3 0.05 1 1.32 30 4.86 34 0.22 67 3.34 17 7.36 19 0.43 38 13.8 32 22.6 46 1.13 86 5.37 100 22.0 98 0.24 72 4.39 18 24.3 16 0.05 2 9.00 41 22.7 39 0.22 101
ALD-Flow [66]40.5 1.53 60 5.62 63 0.01 4 2.86 55 7.94 57 0.16 39 1.29 21 5.15 59 0.12 4 3.34 17 7.68 35 0.38 8 14.0 87 22.8 85 1.15 91 4.71 17 19.2 12 0.11 12 4.41 22 24.4 20 0.06 5 9.30 85 23.5 86 0.20 29
LME [70]41.7 1.39 19 5.10 30 0.01 4 2.49 34 6.95 40 0.10 22 1.35 36 5.64 95 0.15 17 3.29 12 7.55 29 0.39 10 14.1 111 23.0 115 1.26 149 5.13 66 21.1 68 0.19 48 4.37 17 24.2 15 0.06 5 8.89 25 22.5 21 0.19 12
WLIF-Flow [93]41.9 1.44 32 5.19 37 0.01 4 2.52 37 6.81 37 0.16 39 1.38 43 4.67 22 0.20 54 3.21 7 6.98 9 0.42 29 13.7 17 22.4 20 1.07 68 5.37 100 22.1 103 0.28 89 4.43 25 24.4 20 0.08 38 9.09 54 23.0 56 0.21 66
CBF [12]42.0 1.28 7 4.40 8 0.01 4 3.24 82 8.20 71 0.25 72 1.59 89 4.63 21 0.18 38 3.60 39 7.61 32 0.49 69 13.8 32 22.5 30 0.99 14 4.89 30 20.2 33 0.18 42 4.56 61 25.3 70 0.07 15 9.21 70 23.3 72 0.18 8
Aniso. Huber-L1 [22]42.7 1.43 29 5.00 27 0.01 4 4.12 111 9.46 106 0.36 101 1.60 91 4.77 25 0.17 29 3.69 51 7.99 44 0.43 38 13.7 17 22.3 14 1.00 19 4.90 33 20.1 28 0.12 16 4.62 72 25.2 60 0.07 15 8.95 36 22.6 30 0.20 29
JOF [141]43.1 1.58 83 5.80 79 0.01 4 2.18 12 5.88 11 0.10 22 1.28 19 4.68 23 0.19 46 3.37 23 7.21 16 0.42 29 13.9 52 22.7 64 1.21 114 5.35 94 21.9 94 0.20 51 4.33 11 24.0 10 0.07 15 9.18 67 23.2 67 0.20 29
ComponentFusion [96]44.0 1.54 63 6.05 95 0.01 4 2.33 23 6.57 29 0.07 5 1.31 29 4.82 29 0.16 23 3.35 21 7.66 34 0.37 7 13.9 52 22.6 46 1.13 86 4.93 35 20.3 37 0.20 51 4.61 70 25.7 88 0.13 101 9.06 50 22.9 50 0.20 29
TV-L1-MCT [64]44.5 1.70 112 6.35 111 0.02 72 2.90 56 7.98 60 0.17 44 1.39 47 5.19 66 0.20 54 3.32 16 7.10 13 0.45 49 13.9 52 22.6 46 1.17 99 4.67 14 19.3 16 0.15 25 4.44 27 24.4 20 0.08 38 8.67 9 22.0 9 0.19 12
CLG-TV [48]45.5 1.35 15 4.56 12 0.02 72 3.84 99 9.35 98 0.29 84 1.38 43 5.11 54 0.18 38 3.71 53 7.93 42 0.50 78 13.8 32 22.4 20 1.00 19 4.74 20 19.5 19 0.13 20 4.59 68 25.2 60 0.07 15 9.08 51 22.9 50 0.20 29
COFM [59]46.8 1.49 45 5.57 58 0.01 4 2.37 25 6.48 25 0.11 30 1.30 24 4.84 32 0.23 72 3.29 12 7.33 18 0.40 15 13.8 32 22.5 30 1.02 43 5.52 114 22.7 118 0.39 130 4.04 4 22.5 4 0.11 80 9.33 89 23.6 91 0.20 29
CombBMOF [113]47.2 1.59 85 5.46 53 0.02 72 2.30 20 6.40 21 0.10 22 1.39 47 4.83 31 0.21 62 3.90 81 8.60 71 0.46 55 13.8 32 22.5 30 1.01 29 4.92 34 20.1 28 0.11 12 5.35 135 25.9 95 0.10 67 8.89 25 22.4 14 0.19 12
Sparse-NonSparse [56]48.4 1.54 63 5.67 69 0.02 72 2.38 28 6.47 24 0.12 31 1.42 56 5.13 56 0.17 29 3.42 27 7.36 19 0.42 29 13.8 32 22.5 30 1.18 101 5.29 84 21.7 88 0.23 65 4.43 25 24.5 23 0.10 67 9.11 58 23.0 56 0.20 29
FlowFields [110]48.9 1.52 57 5.99 90 0.02 72 2.31 21 6.55 28 0.09 17 1.30 24 4.97 42 0.20 54 3.76 59 9.03 91 0.40 15 13.9 52 22.7 64 1.16 93 5.16 68 21.3 73 0.28 89 4.40 19 24.5 23 0.07 15 8.88 21 22.5 21 0.21 66
ProbFlowFields [128]48.9 1.46 36 5.63 64 0.02 72 2.23 17 6.35 19 0.09 17 1.19 10 4.55 19 0.20 54 3.51 32 8.02 48 0.45 49 13.9 52 22.7 64 1.24 138 5.28 83 21.6 83 0.39 130 4.32 10 24.1 12 0.11 80 8.68 10 22.0 9 0.21 66
TF+OM [100]49.1 1.41 22 5.19 37 0.01 4 2.38 28 6.62 30 0.12 31 1.42 56 5.51 89 0.15 17 3.87 76 8.66 78 0.49 69 13.9 52 22.6 46 1.05 61 4.96 41 20.3 37 0.16 33 4.54 52 25.3 70 0.10 67 9.12 60 23.0 56 0.21 66
IROF-TV [53]49.1 1.51 50 5.64 65 0.02 72 2.60 38 6.84 38 0.16 39 1.34 32 5.38 81 0.18 38 3.29 12 7.49 27 0.42 29 14.0 87 22.8 85 1.21 114 5.05 56 20.8 56 0.20 51 4.52 42 25.2 60 0.07 15 8.82 13 22.3 13 0.21 66
PGM-C [120]49.8 1.50 47 5.75 74 0.01 4 2.37 25 6.68 32 0.10 22 1.48 69 5.32 73 0.15 17 3.77 61 9.13 98 0.50 78 13.9 52 22.6 46 1.21 114 4.97 46 20.4 40 0.23 65 4.54 52 25.1 53 0.08 38 8.94 31 22.6 30 0.20 29
SIOF [67]50.0 1.53 60 5.38 48 0.01 4 4.17 112 10.1 123 0.31 91 1.40 50 5.38 81 0.14 11 3.62 43 7.99 44 0.61 105 13.5 6 22.1 7 0.98 11 4.87 26 20.1 28 0.13 20 4.49 37 24.9 43 0.08 38 9.34 91 23.6 91 0.20 29
CPM-Flow [116]50.7 1.50 47 5.73 73 0.01 4 2.37 25 6.69 33 0.10 22 1.38 43 5.06 52 0.12 4 4.02 94 9.68 113 0.51 84 13.9 52 22.6 46 1.21 114 4.83 24 19.9 23 0.15 25 4.63 75 25.6 82 0.08 38 8.88 21 22.5 21 0.22 101
HAST [109]50.8 1.48 39 5.42 51 0.01 4 2.14 10 5.79 10 0.06 2 1.52 78 5.24 68 0.23 72 3.27 9 7.14 15 0.33 2 14.0 87 22.8 85 0.99 14 5.52 114 22.7 118 0.31 98 4.35 13 24.3 16 0.07 15 9.68 116 24.4 118 0.21 66
FMOF [94]51.3 1.67 109 5.98 88 0.03 128 2.22 15 6.04 15 0.08 11 1.56 82 5.22 67 0.28 100 3.81 70 8.34 61 0.49 69 13.8 32 22.5 30 1.01 29 5.01 51 20.5 46 0.15 25 4.30 8 23.9 9 0.07 15 9.21 70 23.3 72 0.20 29
Ramp [62]52.2 1.57 81 5.75 74 0.01 4 2.38 28 6.51 26 0.19 52 1.41 53 5.11 54 0.17 29 3.26 8 7.09 12 0.41 24 13.9 52 22.6 46 1.15 91 5.51 112 22.5 112 0.32 107 4.48 33 24.7 30 0.07 15 9.33 89 23.6 91 0.20 29
ProFlow_ROB [147]52.3 1.48 39 5.67 69 0.01 4 2.74 47 7.77 54 0.14 36 1.36 38 4.82 29 0.15 17 3.58 36 8.56 68 0.35 3 14.0 87 22.8 85 1.22 127 4.69 16 19.3 16 0.09 9 4.75 94 25.9 95 0.08 38 9.34 91 23.6 91 0.21 66
BlockOverlap [61]52.5 1.34 13 4.33 6 0.02 72 4.04 108 9.04 92 0.48 130 1.36 38 4.31 14 0.32 109 3.43 29 6.81 7 0.71 118 14.0 87 22.8 85 1.04 56 4.89 30 20.0 26 0.22 60 4.44 27 24.7 30 0.11 80 8.64 8 21.8 7 0.20 29
FlowFields+ [130]53.6 1.52 57 5.97 87 0.02 72 2.26 18 6.42 23 0.09 17 1.29 21 5.05 51 0.19 46 3.69 51 8.96 88 0.44 45 14.0 87 22.8 85 1.21 114 5.26 79 21.6 83 0.33 112 4.41 22 24.5 23 0.08 38 8.86 18 22.5 21 0.20 29
2DHMM-SAS [92]53.8 1.65 103 6.25 107 0.02 72 3.41 87 8.58 80 0.22 65 1.36 38 4.84 32 0.19 46 3.28 10 7.02 10 0.43 38 13.8 32 22.5 30 1.19 106 4.94 38 20.1 28 0.09 9 4.52 42 24.8 34 0.12 94 9.25 80 23.4 80 0.20 29
F-TV-L1 [15]54.2 1.54 63 5.24 42 0.02 72 4.11 110 9.73 117 0.32 94 1.50 73 5.44 85 0.23 72 3.76 59 8.03 49 0.47 62 13.5 6 22.1 7 0.94 7 4.71 17 19.4 18 0.17 37 4.61 70 25.3 70 0.13 101 8.92 29 22.6 30 0.19 12
Classic+NL [31]55.0 1.62 93 5.98 88 0.02 72 2.48 31 6.65 31 0.17 44 1.41 53 5.13 56 0.19 46 3.37 23 7.27 17 0.44 45 13.9 52 22.6 46 1.12 85 5.34 91 21.8 91 0.22 60 4.48 33 24.8 34 0.09 54 9.27 83 23.4 80 0.19 12
Classic++ [32]55.1 1.46 36 5.27 43 0.01 4 3.38 86 8.81 87 0.25 72 1.47 68 5.14 58 0.17 29 3.93 86 8.15 54 0.46 55 13.8 32 22.6 46 1.00 19 5.16 68 21.2 72 0.25 79 4.59 68 25.2 60 0.08 38 9.17 66 23.2 67 0.20 29
MDP-Flow [26]55.6 1.35 15 4.97 25 0.02 72 2.22 15 6.25 17 0.09 17 1.20 11 4.07 11 0.14 11 3.89 79 8.34 61 0.48 65 13.8 32 22.5 30 1.24 138 5.74 128 23.6 133 0.50 148 4.63 75 25.6 82 0.11 80 8.97 38 22.7 39 0.19 12
Second-order prior [8]55.8 1.39 19 4.88 21 0.02 72 3.85 101 9.45 105 0.27 78 1.83 106 5.99 108 0.26 89 3.83 73 8.56 68 0.49 69 13.6 10 22.3 14 1.01 29 4.82 23 19.9 23 0.18 42 4.67 82 25.6 82 0.06 5 9.03 46 22.8 45 0.20 29
LDOF [28]57.0 1.45 34 4.81 17 0.02 72 3.10 73 7.33 44 0.56 145 1.58 87 5.39 84 0.22 67 3.92 85 8.47 65 0.63 107 13.8 32 22.5 30 1.02 43 4.68 15 19.2 12 0.13 20 4.48 33 25.0 48 0.10 67 9.01 42 22.8 45 0.22 101
OAR-Flow [125]57.8 1.56 77 5.61 62 0.01 4 2.99 66 8.09 65 0.22 65 1.29 21 4.89 36 0.10 1 3.29 12 7.60 31 0.38 8 14.0 87 22.8 85 1.23 131 5.11 64 21.0 61 0.29 92 4.78 99 26.1 101 0.09 54 9.19 68 23.2 67 0.20 29
LSM [39]57.9 1.64 99 6.31 110 0.01 4 2.48 31 6.79 36 0.12 31 1.51 76 5.55 91 0.17 29 3.59 38 7.98 43 0.41 24 13.9 52 22.6 46 1.19 106 5.36 97 21.9 94 0.25 79 4.47 31 24.7 30 0.10 67 9.23 75 23.3 72 0.20 29
AggregFlow [97]58.4 1.89 131 7.48 131 0.01 4 2.95 58 8.13 67 0.16 39 1.22 13 4.86 34 0.14 11 4.18 101 9.69 114 0.45 49 13.9 52 22.6 46 1.04 56 4.94 38 20.2 33 0.15 25 4.51 40 25.0 48 0.12 94 9.24 79 23.3 72 0.21 66
DMF_ROB [140]59.1 1.54 63 5.65 66 0.01 4 3.31 85 8.76 85 0.28 82 2.11 121 6.42 120 0.44 122 4.01 92 8.89 85 0.46 55 13.7 17 22.4 20 1.20 109 4.87 26 20.1 28 0.22 60 4.54 52 24.8 34 0.07 15 8.90 27 22.5 21 0.20 29
ComplOF-FED-GPU [35]59.1 1.56 77 5.92 83 0.02 72 2.71 44 7.63 49 0.17 44 1.95 113 5.09 53 0.39 119 3.63 45 8.64 75 0.40 15 13.7 17 22.5 30 1.13 86 4.98 47 20.5 46 0.17 37 4.70 86 25.7 88 0.07 15 9.31 86 23.4 80 0.19 12
CRTflow [80]59.2 1.50 47 5.48 55 0.02 72 3.87 102 9.40 102 0.36 101 1.62 94 6.22 115 0.24 79 3.55 33 7.76 39 0.43 38 13.9 52 22.7 64 1.21 114 4.77 22 19.6 21 0.14 23 4.53 47 25.2 60 0.07 15 9.04 47 22.9 50 0.20 29
DPOF [18]61.5 1.64 99 6.52 119 0.04 146 1.98 8 5.37 8 0.07 5 1.83 106 4.75 24 0.34 112 3.75 58 8.80 83 0.48 65 13.7 17 22.3 14 1.01 29 5.16 68 21.0 61 0.14 23 4.67 82 25.3 70 0.06 5 9.31 86 23.5 86 0.22 101
S2F-IF [123]61.5 1.55 72 6.13 99 0.01 4 2.26 18 6.41 22 0.09 17 1.30 24 5.17 62 0.17 29 3.72 54 9.01 90 0.45 49 14.0 87 22.9 105 1.24 138 5.27 80 21.6 83 0.32 107 4.54 52 25.2 60 0.09 54 8.88 21 22.5 21 0.23 127
TC-Flow [46]61.7 1.51 50 5.70 72 0.01 4 2.97 61 8.31 75 0.21 63 1.46 64 5.35 75 0.11 2 3.63 45 8.10 51 0.58 102 14.0 87 22.8 85 1.21 114 5.10 61 21.0 61 0.29 92 4.53 47 25.0 48 0.07 15 9.19 68 23.3 72 0.21 66
p-harmonic [29]62.2 1.42 25 4.96 24 0.01 4 4.00 105 9.50 107 0.39 111 1.38 43 5.68 97 0.19 46 4.20 103 8.58 70 0.49 69 13.9 52 22.6 46 1.01 29 4.93 35 20.3 37 0.21 56 4.81 102 26.1 101 0.11 80 9.12 60 23.1 63 0.20 29
OFLAF [77]63.6 1.48 39 5.49 57 0.01 4 2.00 9 5.50 9 0.07 5 1.30 24 4.97 42 0.15 17 3.34 17 7.40 22 0.40 15 14.0 87 22.8 85 1.21 114 5.56 119 22.8 122 0.41 133 4.83 104 26.3 108 0.17 122 9.73 121 24.5 123 0.20 29
SVFilterOh [111]63.8 1.51 50 5.48 55 0.02 72 2.19 14 5.94 13 0.10 22 1.50 73 5.00 45 0.25 83 3.78 63 8.01 47 0.40 15 14.3 130 23.2 128 1.22 127 5.35 94 22.0 98 0.23 65 4.27 6 23.8 7 0.06 5 9.55 108 24.1 111 0.22 101
EAI-Flow [152]63.8 1.65 103 5.96 86 0.03 128 2.84 52 7.64 51 0.25 72 1.56 82 5.73 100 0.19 46 3.82 71 8.93 87 0.35 3 13.9 52 22.6 46 1.18 101 4.87 26 20.0 26 0.22 60 4.75 94 26.0 98 0.14 105 8.63 7 21.8 7 0.20 29
FC-2Layers-FF [74]64.8 1.56 77 5.94 85 0.02 72 1.86 5 4.90 4 0.08 11 1.39 47 5.29 72 0.20 54 3.34 17 7.46 25 0.40 15 13.9 52 22.8 85 1.20 109 5.56 119 22.9 123 0.37 126 4.53 47 24.9 43 0.11 80 9.35 94 23.6 91 0.22 101
Occlusion-TV-L1 [63]66.9 1.43 29 5.20 40 0.01 4 4.18 114 10.3 127 0.37 106 1.34 32 5.35 75 0.27 96 4.19 102 9.14 99 0.56 97 13.7 17 22.4 20 0.97 8 4.99 49 20.6 50 0.33 112 5.12 123 25.4 76 0.27 137 9.01 42 22.7 39 0.19 12
S2D-Matching [84]67.2 1.65 103 6.07 97 0.02 72 3.21 78 8.58 80 0.23 70 1.34 32 5.00 45 0.23 72 3.35 21 7.36 19 0.42 29 13.9 52 22.7 64 1.07 68 5.53 117 22.6 115 0.35 120 4.57 64 24.8 34 0.07 15 9.22 74 23.3 72 0.22 101
TC/T-Flow [76]68.7 1.69 111 6.24 105 0.02 72 2.98 63 8.17 69 0.19 52 1.27 18 4.78 26 0.13 9 3.58 36 8.26 58 0.36 5 14.1 111 23.0 115 1.23 131 5.03 53 20.6 50 0.12 16 4.85 106 26.3 108 0.16 119 9.40 100 23.8 103 0.19 12
RFlow [90]69.0 1.41 22 5.18 36 0.02 72 3.96 104 9.67 114 0.35 98 1.41 53 5.37 78 0.26 89 3.91 84 8.76 79 0.52 85 13.7 17 22.5 30 1.01 29 4.96 41 20.5 46 0.21 56 4.58 65 25.5 78 0.09 54 9.37 98 23.7 100 0.23 127
EpicFlow [102]69.6 1.51 50 5.82 81 0.01 4 2.90 56 8.08 63 0.18 50 1.43 59 5.25 69 0.16 23 3.88 78 9.42 108 0.54 91 13.9 52 22.7 64 1.21 114 5.13 66 21.1 68 0.31 98 4.71 88 25.8 92 0.14 105 9.13 62 23.1 63 0.21 66
AGIF+OF [85]69.9 1.66 108 6.04 94 0.02 72 2.48 31 6.73 34 0.19 52 1.44 60 4.96 41 0.26 89 3.41 26 7.54 28 0.45 49 14.1 111 23.1 124 1.21 114 5.47 108 22.3 107 0.27 84 4.52 42 24.3 16 0.07 15 9.41 103 23.8 103 0.21 66
MLDP_OF [89]69.9 1.55 72 6.03 93 0.02 72 3.12 74 8.40 78 0.20 59 1.22 13 4.97 42 0.13 9 3.77 61 7.62 33 0.73 120 13.9 52 22.7 64 1.01 29 5.60 123 23.0 127 0.31 98 4.63 75 25.3 70 0.14 105 9.21 70 23.3 72 0.21 66
HBM-GC [105]70.4 1.54 63 5.66 67 0.01 4 2.98 63 8.21 72 0.17 44 1.20 11 3.96 8 0.18 38 3.63 45 7.79 41 0.43 38 14.4 135 23.5 136 1.29 152 5.96 140 24.4 142 0.48 144 4.46 30 24.6 29 0.05 2 9.35 94 23.6 91 0.22 101
OFH [38]71.8 1.56 77 5.79 78 0.01 4 3.55 91 8.78 86 0.30 89 1.62 94 6.44 121 0.16 23 3.57 34 8.52 67 0.39 10 13.8 32 22.6 46 1.16 93 5.18 72 21.3 73 0.31 98 4.94 112 26.6 111 0.15 112 9.36 96 23.6 91 0.19 12
RNLOD-Flow [121]71.9 1.52 57 5.81 80 0.01 4 3.05 71 8.32 77 0.19 52 1.58 87 5.84 107 0.32 109 3.47 30 7.69 36 0.43 38 13.9 52 22.7 64 1.19 106 5.36 97 22.0 98 0.26 83 4.51 40 24.8 34 0.14 105 9.66 114 24.4 118 0.21 66
Ad-TV-NDC [36]72.4 1.63 96 4.88 21 0.03 128 5.06 139 10.3 127 0.36 101 1.45 62 5.55 91 0.20 54 4.53 116 9.15 100 0.57 99 14.1 111 22.9 105 0.99 14 4.74 20 19.6 21 0.16 33 4.80 101 25.8 92 0.06 5 9.02 44 22.8 45 0.19 12
PMF [73]72.7 1.59 85 6.16 103 0.01 4 2.73 46 7.62 48 0.07 5 1.65 98 6.90 130 0.28 100 3.74 56 8.38 63 0.40 15 14.1 111 23.0 115 1.02 43 5.10 61 20.9 59 0.19 48 4.56 61 25.4 76 0.09 54 9.84 128 24.9 135 0.22 101
Sparse Occlusion [54]73.5 1.51 50 5.58 59 0.02 72 3.51 90 9.43 104 0.19 52 1.37 42 4.95 40 0.18 38 3.80 67 8.33 59 0.49 69 13.9 52 22.7 64 1.20 109 5.58 121 22.9 123 0.37 126 4.73 91 25.8 92 0.07 15 9.38 99 23.7 100 0.20 29
TCOF [69]74.6 1.54 63 5.59 60 0.01 4 4.46 124 10.4 130 0.43 122 1.28 19 5.15 59 0.14 11 3.63 45 8.04 50 0.42 29 13.9 52 22.7 64 0.98 11 5.41 103 22.3 107 0.24 72 5.00 116 26.7 113 0.09 54 9.76 122 24.6 126 0.24 134
Modified CLG [34]75.8 1.31 10 4.60 13 0.01 4 4.56 129 9.63 112 0.50 138 1.63 96 6.45 123 0.33 111 4.14 100 9.05 93 0.62 106 13.9 52 22.6 46 1.02 43 5.08 60 20.8 56 0.31 98 4.65 79 25.9 95 0.09 54 9.08 51 22.9 50 0.22 101
HBpMotionGpu [43]75.9 1.64 99 5.67 69 0.02 72 5.07 140 11.0 143 0.48 130 1.33 31 4.89 36 0.22 67 4.40 107 9.95 121 0.52 85 13.8 32 22.6 46 1.17 99 5.30 86 21.4 77 0.29 92 4.48 33 24.9 43 0.06 5 9.23 75 23.2 67 0.21 66
CostFilter [40]76.5 1.77 120 7.36 129 0.01 4 2.66 41 7.51 47 0.08 11 1.82 105 7.88 143 0.29 104 4.00 90 9.50 111 0.31 1 14.2 126 23.1 124 1.07 68 4.98 47 20.4 40 0.17 37 4.62 72 25.6 82 0.08 38 9.65 112 24.4 118 0.21 66
Adaptive [20]76.8 1.48 39 5.33 46 0.02 72 4.48 125 10.6 137 0.43 122 1.53 80 5.50 88 0.18 38 3.84 75 8.33 59 0.54 91 13.9 52 22.7 64 1.00 19 5.20 75 21.4 77 0.28 89 4.89 107 26.1 101 0.07 15 9.41 103 23.8 103 0.21 66
AdaConv-v1 [126]77.8 2.34 143 9.17 143 0.04 146 4.08 109 8.31 75 0.75 149 2.48 133 6.07 112 0.62 134 7.79 147 14.5 150 2.10 152 12.8 4 20.9 4 0.69 2 4.24 6 17.6 6 0.09 9 4.52 42 25.1 53 0.22 135 8.39 6 21.2 6 0.12 2
FF++_ROB [146]78.0 1.55 72 6.13 99 0.01 4 2.70 42 7.41 45 0.14 36 1.46 64 5.35 75 0.25 83 3.98 88 9.53 112 0.53 88 14.1 111 22.9 105 1.25 146 5.34 91 21.9 94 0.32 107 4.55 58 25.2 60 0.12 94 8.94 31 22.6 30 0.25 138
FlowNetS+ft+v [112]78.3 1.48 39 5.14 33 0.02 72 4.36 120 9.68 116 0.78 150 1.57 85 5.37 78 0.26 89 3.90 81 8.47 65 0.81 128 13.9 52 22.7 64 1.23 131 4.83 24 19.9 23 0.24 72 4.70 86 26.1 101 0.12 94 9.09 54 23.0 56 0.21 66
PWC-Net_ROB [148]78.4 1.82 124 7.81 135 0.02 72 3.01 67 8.61 82 0.12 31 1.46 64 6.05 110 0.19 46 3.79 65 9.29 107 0.41 24 14.1 111 23.1 124 1.24 138 5.30 86 21.3 73 0.20 51 4.54 52 25.0 48 0.10 67 8.98 39 22.7 39 0.23 127
Bartels [41]79.0 1.59 85 6.24 105 0.03 128 3.20 77 8.92 90 0.31 91 1.40 50 5.17 62 0.25 83 4.09 96 9.05 93 0.86 133 14.1 111 22.9 105 0.97 8 5.43 106 22.2 106 0.25 79 4.47 31 24.8 34 0.10 67 9.15 64 23.1 63 0.20 29
TriFlow [95]79.8 1.65 103 6.58 120 0.01 4 3.78 98 9.53 109 0.29 84 1.45 62 5.99 108 0.18 38 4.07 95 9.25 105 0.49 69 14.0 87 22.9 105 1.18 101 5.34 91 21.4 77 0.15 25 4.64 78 25.2 60 0.08 38 9.40 100 23.6 91 0.21 66
Efficient-NL [60]80.8 1.53 60 5.60 61 0.01 4 2.98 63 7.94 57 0.16 39 2.05 115 5.45 87 0.56 130 3.82 71 8.12 53 0.42 29 13.8 32 22.5 30 1.18 101 5.69 126 23.2 129 0.32 107 4.75 94 26.1 101 0.11 80 9.94 137 24.8 132 0.22 101
EPPM w/o HM [88]81.4 1.68 110 7.02 125 0.02 72 2.84 52 8.08 63 0.10 22 2.13 124 7.82 142 0.36 114 3.87 76 9.12 96 0.49 69 13.9 52 22.7 64 1.04 56 5.27 80 21.6 83 0.17 37 4.56 61 25.2 60 0.15 112 9.34 91 23.6 91 0.22 101
TVL1_ROB [139]81.5 1.55 72 4.95 23 0.01 4 5.64 145 11.1 144 0.48 130 1.49 70 5.79 105 0.31 108 4.84 123 9.28 106 0.70 117 13.9 52 22.7 64 1.03 52 5.06 58 20.9 59 0.23 65 4.93 111 26.9 117 0.18 125 9.05 49 22.9 50 0.18 8
FESL [72]82.6 1.63 96 5.93 84 0.02 72 2.50 36 6.86 39 0.14 36 1.49 70 5.44 85 0.26 89 3.80 67 8.16 55 0.50 78 14.1 111 22.9 105 1.21 114 5.60 123 22.9 123 0.40 132 4.58 65 25.0 48 0.08 38 9.50 105 24.0 110 0.22 101
Filter Flow [19]82.8 1.54 63 5.28 44 0.01 4 4.52 128 9.97 122 0.35 98 1.61 93 5.53 90 0.20 54 4.55 117 8.61 73 0.46 55 14.3 130 23.2 128 1.08 74 5.10 61 21.0 61 0.21 56 4.76 98 26.1 101 0.11 80 9.65 112 24.3 117 0.20 29
Nguyen [33]82.8 1.55 72 5.02 28 0.00 1 5.69 146 10.8 139 0.48 130 1.63 96 6.70 127 0.28 100 5.35 133 10.5 127 0.77 125 13.8 32 22.5 30 1.01 29 4.99 49 20.7 53 0.18 42 5.44 137 28.2 131 0.21 131 9.08 51 22.9 50 0.20 29
Classic+CPF [83]83.3 1.65 103 6.18 104 0.02 72 2.63 40 7.07 42 0.17 44 1.44 60 5.38 81 0.23 72 3.40 25 7.45 24 0.42 29 14.3 130 23.3 132 1.21 114 5.68 125 23.2 129 0.31 98 4.71 88 25.3 70 0.10 67 9.77 123 24.6 126 0.22 101
2D-CLG [1]83.4 1.42 25 5.09 29 0.01 4 4.91 135 9.82 121 0.48 130 2.21 126 5.62 93 0.57 131 5.05 127 9.90 120 0.80 127 13.8 32 22.5 30 1.09 78 5.07 59 21.0 61 0.42 135 4.89 107 26.7 113 0.13 101 9.11 58 22.6 30 0.20 29
GraphCuts [14]83.5 1.90 132 6.51 118 0.02 72 2.96 60 7.63 49 0.22 65 3.79 149 5.16 61 0.64 136 4.49 114 9.24 104 0.55 94 13.9 52 22.7 64 0.99 14 5.04 54 20.8 56 0.21 56 4.53 47 25.2 60 0.15 112 9.88 133 24.9 135 0.21 66
IAOF [50]83.7 1.79 121 5.85 82 0.02 72 6.44 151 12.4 152 0.55 143 1.84 109 5.77 103 0.36 114 4.78 121 9.12 96 0.75 124 13.7 17 22.4 20 1.01 29 5.01 51 20.7 53 0.15 25 4.73 91 25.7 88 0.09 54 9.29 84 23.4 80 0.20 29
SRR-TVOF-NL [91]84.0 1.79 121 6.72 122 0.02 72 3.21 78 8.72 84 0.29 84 1.42 56 5.34 74 0.20 54 4.29 104 9.00 89 0.53 88 13.9 52 22.8 85 1.20 109 5.24 77 21.5 82 0.22 60 4.68 84 25.1 53 0.07 15 9.93 136 25.0 137 0.22 101
Complementary OF [21]84.1 1.61 90 6.47 115 0.01 4 2.74 47 7.75 53 0.19 52 2.67 140 5.71 98 0.89 151 3.74 56 8.76 79 0.41 24 13.9 52 22.7 64 1.14 90 5.19 73 21.4 77 0.31 98 4.98 115 26.9 117 0.13 101 9.78 125 24.8 132 0.21 66
Steered-L1 [118]84.2 1.33 12 4.83 19 0.02 72 2.85 54 7.86 56 0.32 94 2.08 117 5.18 65 0.64 136 4.36 106 8.60 71 1.06 140 14.1 111 23.0 115 0.97 8 5.12 65 21.1 68 0.34 115 4.69 85 26.1 101 0.12 94 9.50 105 24.1 111 0.22 101
Black & Anandan [4]84.8 1.63 96 5.12 31 0.01 4 5.17 142 10.5 134 0.41 115 2.30 129 6.36 117 0.47 126 5.20 131 9.84 119 0.49 69 14.0 87 22.9 105 1.02 43 4.88 29 20.2 33 0.18 42 5.10 122 27.0 119 0.08 38 9.23 75 23.1 63 0.21 66
TV-L1-improved [17]84.8 1.42 25 5.19 37 0.01 4 4.41 122 10.4 130 0.40 112 2.10 118 5.27 71 0.50 127 3.89 79 8.46 64 0.52 85 14.0 87 22.8 85 1.00 19 5.33 90 22.0 98 0.25 79 4.96 114 27.5 127 0.23 136 9.26 81 23.4 80 0.21 66
LFNet_ROB [150]85.8 1.82 124 7.57 133 0.03 128 3.04 70 8.27 74 0.22 65 1.49 70 6.12 114 0.21 62 3.99 89 9.80 117 0.57 99 13.8 32 22.6 46 1.26 149 5.51 112 22.5 112 0.34 115 4.54 52 25.1 53 0.09 54 8.94 31 22.5 21 0.25 138
Fusion [6]87.6 1.46 36 5.40 49 0.02 72 2.70 42 6.97 41 0.17 44 1.30 24 4.55 19 0.29 104 4.32 105 8.77 81 0.46 55 14.3 130 23.5 136 1.02 43 5.93 139 24.4 142 0.47 141 4.83 104 26.7 113 0.12 94 10.4 143 26.1 145 0.22 101
BriefMatch [124]88.6 1.61 90 6.15 102 0.03 128 2.97 61 7.73 52 0.61 147 2.12 122 4.79 27 0.58 132 4.84 123 9.05 93 1.68 149 13.9 52 22.7 64 1.08 74 5.30 86 21.8 91 0.24 72 4.53 47 24.9 43 0.14 105 9.13 62 23.0 56 0.28 149
ResPWCR_ROB [145]88.6 1.70 112 6.74 123 0.02 72 3.24 82 8.87 88 0.27 78 2.22 127 6.11 113 0.23 72 4.42 108 10.6 128 0.69 115 13.6 10 22.2 12 1.23 131 5.36 97 21.7 88 0.16 33 4.79 100 25.7 88 0.15 112 9.31 86 23.5 86 0.21 66
3DFlow [135]88.9 1.71 114 6.28 109 0.02 72 2.62 39 7.42 46 0.20 59 1.91 112 4.91 39 0.25 83 3.61 40 8.10 51 0.55 94 13.9 52 22.7 64 1.02 43 6.30 150 25.3 151 0.47 141 5.26 131 27.2 124 0.15 112 9.71 118 24.5 123 0.21 66
CNN-flow-warp+ref [117]89.2 1.31 10 4.62 14 0.02 72 3.65 97 9.05 93 0.47 129 2.07 116 6.56 126 0.43 121 5.51 135 9.78 116 1.12 143 13.9 52 22.7 64 1.23 131 4.96 41 20.5 46 0.35 120 4.90 109 27.1 122 0.18 125 9.04 47 22.8 45 0.21 66
Rannacher [23]91.0 1.49 45 5.66 67 0.01 4 4.49 126 10.6 137 0.40 112 2.19 125 5.77 103 0.52 129 3.79 65 8.65 77 0.53 88 14.0 87 22.8 85 1.02 43 5.29 84 21.8 91 0.27 84 4.95 113 27.4 125 0.21 131 9.26 81 23.4 80 0.22 101
FlowNet2 [122]91.1 2.65 147 10.2 146 0.02 72 3.21 78 8.40 78 0.22 65 1.75 100 6.40 118 0.27 96 4.43 111 11.4 133 0.67 113 14.1 111 23.0 115 1.11 82 5.17 71 21.0 61 0.19 48 4.65 79 25.5 78 0.08 38 9.10 57 23.0 56 0.24 134
ContinualFlow_ROB [153]91.4 1.99 135 8.31 141 0.03 128 3.22 81 8.92 90 0.25 72 1.78 102 6.97 132 0.26 89 4.00 90 9.83 118 0.48 65 14.0 87 22.9 105 1.24 138 4.93 35 20.4 40 0.18 42 4.62 72 25.1 53 0.08 38 9.40 100 23.8 103 0.25 138
AugFNG_ROB [144]91.8 1.87 129 7.84 137 0.02 72 3.57 93 9.08 94 0.35 98 1.85 110 8.05 146 0.22 67 4.52 115 11.3 131 0.55 94 14.2 126 23.2 128 1.25 146 4.89 30 20.2 33 0.12 16 4.92 110 26.0 98 0.10 67 8.88 21 22.4 14 0.23 127
SimpleFlow [49]96.0 1.60 89 6.00 91 0.01 4 3.50 89 8.63 83 0.32 94 2.53 134 6.06 111 0.79 141 3.42 27 7.56 30 0.48 65 14.0 87 22.8 85 1.20 109 5.77 130 23.7 135 0.43 138 5.01 117 28.0 129 0.42 147 9.68 116 24.5 123 0.20 29
Horn & Schunck [3]97.0 1.62 93 5.42 51 0.01 4 5.40 144 10.9 140 0.44 125 2.27 128 6.97 132 0.45 123 6.35 143 11.5 135 0.65 110 14.1 111 23.0 115 1.06 65 4.95 40 20.4 40 0.17 37 5.51 138 28.2 131 0.14 105 9.57 109 23.7 100 0.18 8
LocallyOriented [52]97.0 1.61 90 6.13 99 0.01 4 4.64 130 10.9 140 0.40 112 1.83 106 6.44 121 0.26 89 4.45 113 10.2 125 0.47 62 14.0 87 22.8 85 1.01 29 5.58 121 22.6 115 0.27 84 5.18 127 26.8 116 0.15 112 9.66 114 24.4 118 0.20 29
TriangleFlow [30]97.4 1.73 116 6.50 116 0.02 72 3.84 99 9.51 108 0.31 91 1.78 102 5.71 98 0.27 96 4.43 111 10.1 123 0.57 99 13.7 17 22.5 30 0.98 11 5.69 126 22.9 123 0.23 65 5.16 125 28.3 134 0.21 131 10.1 139 25.4 139 0.21 66
2bit-BM-tele [98]99.5 1.51 50 5.13 32 0.04 146 4.17 112 10.2 124 0.41 115 1.52 78 4.90 38 0.46 124 4.01 92 8.64 75 0.60 104 14.4 135 23.3 132 1.05 61 5.89 134 24.1 137 0.45 140 5.77 144 32.3 151 0.60 150 8.90 27 22.5 21 0.21 66
EPMNet [133]100.0 2.65 147 10.8 149 0.03 128 3.16 76 8.19 70 0.25 72 1.75 100 6.40 118 0.27 96 5.13 128 13.4 147 0.65 110 14.1 111 23.0 115 1.11 82 5.39 102 22.1 103 0.23 65 4.65 79 25.5 78 0.08 38 9.23 75 23.3 72 0.25 138
Shiralkar [42]100.3 1.85 127 7.19 126 0.01 4 4.31 119 9.75 118 0.37 106 2.10 118 7.58 137 0.36 114 5.54 136 11.4 133 0.63 107 13.8 32 22.5 30 1.07 68 5.46 107 22.4 109 0.34 115 5.32 133 27.8 128 0.20 130 9.36 96 23.5 86 0.20 29
LiteFlowNet [143]100.3 1.90 132 8.16 139 0.03 128 2.81 50 8.01 62 0.20 59 1.60 91 6.95 131 0.24 79 4.75 120 11.8 138 0.86 133 13.9 52 22.7 64 1.25 146 5.50 111 22.1 103 0.32 107 5.07 121 26.6 111 0.19 127 8.95 36 22.6 30 0.25 138
Aniso-Texture [82]100.6 1.42 25 5.22 41 0.02 72 4.23 116 10.4 130 0.48 130 2.33 130 5.37 78 0.25 83 4.09 96 9.18 101 0.87 135 14.1 111 23.1 124 1.24 138 6.03 143 24.8 147 0.57 152 4.55 58 24.9 43 0.08 38 9.58 111 24.1 111 0.22 101
Correlation Flow [75]101.5 1.71 114 6.50 116 0.02 72 4.00 105 10.2 124 0.36 101 1.35 36 4.81 28 0.19 46 3.90 81 8.77 81 0.50 78 14.0 87 22.9 105 1.05 61 6.25 149 24.8 147 0.43 138 5.22 129 28.1 130 0.19 127 9.80 126 24.7 129 0.23 127
TI-DOFE [24]102.9 1.76 118 6.02 92 0.01 4 6.21 150 11.7 150 0.51 140 1.97 114 7.23 135 0.28 100 6.30 142 11.3 131 0.85 132 14.0 87 22.8 85 1.02 43 4.96 41 20.4 40 0.15 25 5.21 128 27.1 122 0.15 112 9.86 130 23.8 103 0.26 146
SPSA-learn [13]103.6 1.59 85 5.32 45 0.01 4 4.23 116 9.36 100 0.42 119 2.54 136 6.27 116 0.80 142 5.24 132 9.44 109 0.73 120 14.0 87 22.8 85 1.03 52 5.27 80 21.7 88 0.31 98 5.93 149 33.0 152 0.86 153 10.4 143 26.2 147 0.20 29
IIOF-NLDP [131]103.8 1.76 118 6.66 121 0.02 72 3.58 94 9.59 111 0.28 82 1.79 104 5.02 49 0.24 79 4.13 99 8.86 84 0.63 107 13.8 32 22.6 46 1.05 61 6.33 151 24.8 147 0.52 150 5.81 145 31.8 150 0.63 151 9.72 120 24.2 116 0.22 101
ROF-ND [107]104.0 1.73 116 5.36 47 0.01 4 3.43 88 9.15 95 0.27 78 1.59 89 5.63 94 0.21 62 5.50 134 12.2 144 0.77 125 14.0 87 22.8 85 1.21 114 5.96 140 24.2 140 0.42 135 5.51 138 28.7 137 0.11 80 9.90 135 24.7 129 0.22 101
IAOF2 [51]104.2 1.81 123 6.46 114 0.02 72 4.65 131 11.3 148 0.36 101 1.57 85 5.79 105 0.21 62 4.61 119 10.0 122 0.56 97 14.4 135 23.5 136 1.16 93 5.48 109 22.6 115 0.29 92 4.75 94 25.6 82 0.11 80 9.57 109 24.1 111 0.21 66
StereoFlow [44]106.2 4.05 153 12.8 153 0.02 72 5.34 143 12.0 151 0.29 84 1.36 38 5.65 96 0.22 67 3.80 67 8.17 56 0.50 78 16.7 152 27.3 152 1.13 86 7.27 153 29.7 153 0.42 135 4.58 65 25.5 78 0.10 67 10.3 140 26.1 145 0.21 66
SegOF [10]109.8 1.57 81 6.05 95 0.02 72 3.63 95 8.91 89 0.24 71 2.71 141 6.79 129 0.74 140 4.81 122 11.7 137 0.73 120 14.0 87 22.8 85 1.22 127 5.52 114 22.7 118 0.48 144 5.17 126 28.7 137 0.33 142 9.21 70 23.2 67 0.23 127
OFRF [134]111.1 2.03 138 7.52 132 0.03 128 4.40 121 10.2 124 0.41 115 1.67 99 6.55 124 0.17 29 4.10 98 9.44 109 0.50 78 14.2 126 23.2 128 1.16 93 5.75 129 23.2 129 0.24 72 5.02 118 27.0 119 0.11 80 10.0 138 25.4 139 0.22 101
ACK-Prior [27]113.9 1.64 99 6.39 113 0.02 72 2.81 50 7.97 59 0.19 52 2.53 134 5.74 101 0.63 135 4.56 118 10.1 123 1.09 142 14.7 145 24.0 146 1.27 151 6.04 144 24.5 145 0.29 92 5.04 120 27.4 125 0.10 67 11.0 149 27.7 150 0.22 101
SILK [79]114.0 1.86 128 7.35 128 0.01 4 5.84 147 11.2 145 0.60 146 3.00 143 7.69 138 0.83 145 5.68 138 10.6 128 0.69 115 14.1 111 23.0 115 1.00 19 5.31 89 21.3 73 0.36 123 5.02 118 27.0 119 0.28 138 9.50 105 23.5 86 0.24 134
Dynamic MRF [7]116.2 1.58 83 6.37 112 0.02 72 3.55 91 9.67 114 0.27 78 2.35 131 7.75 141 0.50 127 5.74 139 10.9 130 1.06 140 14.0 87 22.8 85 1.23 131 5.90 137 24.1 137 0.51 149 5.28 132 28.7 137 0.34 143 9.71 118 23.9 108 0.21 66
Adaptive flow [45]117.0 2.02 137 6.27 108 0.03 128 5.93 148 11.2 145 0.55 143 1.87 111 5.74 101 0.37 118 5.16 129 9.21 102 0.73 120 14.7 145 24.0 146 1.04 56 5.89 134 24.3 141 0.37 126 4.72 90 26.3 108 0.14 105 9.84 128 24.8 132 0.18 8
Learning Flow [11]118.6 1.62 93 6.12 98 0.02 72 4.43 123 10.4 130 0.33 97 2.86 142 7.92 144 0.82 144 5.19 130 9.72 115 0.59 103 14.6 142 23.8 144 1.10 80 5.42 104 22.4 109 0.27 84 5.24 130 28.3 134 0.17 122 10.3 140 25.4 139 0.23 127
H+S_ROB [138]119.1 2.01 136 7.83 136 0.01 4 4.66 132 9.24 97 0.42 119 3.34 146 8.79 147 0.86 149 8.48 151 12.0 141 0.84 130 14.4 135 23.4 134 1.07 68 5.48 109 22.4 109 0.34 115 5.84 146 30.0 147 0.35 144 9.82 127 23.9 108 0.20 29
StereoOF-V1MT [119]119.6 1.94 134 7.40 130 0.02 72 3.93 103 9.53 109 0.41 115 2.57 137 7.29 136 0.60 133 6.23 141 11.5 135 0.94 136 14.2 126 23.0 115 1.24 138 5.81 132 22.7 118 0.48 144 5.53 141 28.2 131 0.29 139 9.16 65 22.7 39 0.22 101
FOLKI [16]121.2 1.88 130 7.22 127 0.02 72 6.20 149 11.2 145 0.86 151 2.60 138 9.02 148 0.64 136 7.81 148 12.1 143 1.70 150 14.6 142 23.7 141 1.06 65 5.19 73 21.0 61 0.24 72 5.43 136 28.9 141 0.32 140 9.77 123 24.1 111 0.21 66
UnFlow [129]122.0 2.13 141 8.96 142 0.03 128 4.27 118 9.79 120 0.42 119 2.12 122 7.71 139 0.35 113 4.42 108 10.4 126 0.67 113 14.0 87 22.9 105 1.16 93 5.85 133 23.3 132 0.47 141 4.82 103 25.6 82 0.16 119 11.0 149 25.7 143 0.33 151
NL-TV-NCC [25]122.2 2.10 140 7.63 134 0.03 128 3.63 95 9.77 119 0.25 72 2.10 118 6.55 124 0.29 104 5.56 137 12.2 144 0.54 91 14.5 140 23.5 136 1.08 74 6.15 145 24.4 142 0.36 123 6.66 153 29.8 146 0.16 119 10.3 140 25.8 144 0.21 66
SLK [47]126.7 2.08 139 8.23 140 0.01 4 5.10 141 9.38 101 0.50 138 3.21 144 7.73 140 0.83 145 8.10 150 14.2 149 1.61 148 14.5 140 23.7 141 1.06 65 5.77 130 22.5 112 0.36 123 5.84 146 30.6 148 0.36 146 9.87 131 24.4 118 0.22 101
HCIC-L [99]127.0 3.37 152 10.9 150 0.07 152 4.98 136 10.3 127 0.38 110 2.44 132 8.00 145 0.36 114 7.09 144 13.0 146 0.71 118 14.9 148 24.2 148 1.07 68 5.99 142 24.1 137 0.23 65 4.74 93 26.0 98 0.11 80 12.3 153 30.4 153 0.25 138
WRT [151]127.9 1.84 126 6.75 124 0.03 128 4.02 107 9.17 96 0.37 106 3.21 144 5.25 69 0.84 147 4.42 108 9.03 91 0.84 130 14.3 130 23.4 134 1.08 74 6.54 152 26.5 152 0.54 151 6.27 151 34.8 153 0.84 152 10.9 148 27.5 149 0.27 148
WOLF_ROB [149]133.1 2.64 145 9.90 145 0.03 128 5.04 137 10.9 140 0.45 126 2.66 139 6.77 128 0.46 124 4.88 126 12.0 141 0.65 110 14.4 135 23.5 136 1.22 127 5.89 134 23.6 133 0.37 126 5.89 148 29.3 142 0.19 127 9.87 131 24.7 129 0.25 138
FFV1MT [106]133.2 2.64 145 10.4 148 0.03 128 4.88 134 9.35 98 0.52 142 4.45 150 13.1 152 0.71 139 7.42 145 11.9 139 1.22 145 14.6 142 23.7 141 1.09 78 5.53 117 21.1 68 0.33 112 6.16 150 29.7 145 0.35 144 10.5 146 25.6 142 0.26 146
PGAM+LK [55]138.0 2.35 144 9.74 144 0.05 151 5.04 137 10.5 134 0.66 148 3.38 147 9.68 149 0.84 147 7.99 149 14.6 151 1.24 147 14.7 145 23.8 144 1.10 80 5.92 138 23.9 136 0.35 120 5.33 134 28.5 136 0.17 122 9.88 133 24.6 126 0.32 150
Pyramid LK [2]139.5 2.13 141 8.10 138 0.04 146 7.17 152 11.5 149 0.99 153 6.22 152 6.97 132 1.21 152 13.9 153 24.7 153 2.97 153 15.7 151 25.7 151 1.11 82 5.42 104 22.0 98 0.30 97 5.55 142 29.5 144 0.52 148 11.9 151 29.7 152 0.54 153
Heeger++ [104]141.7 3.11 151 11.9 152 0.03 128 4.77 133 9.65 113 0.51 140 4.48 151 12.0 151 0.80 142 7.42 145 11.9 139 1.22 145 15.0 149 24.5 149 1.23 131 6.24 147 23.0 127 0.60 153 6.37 152 29.4 143 0.32 140 10.4 143 25.1 138 0.25 138
GroupFlow [9]142.8 2.80 150 11.2 151 0.04 146 4.49 126 10.5 134 0.43 122 3.45 148 9.80 150 0.88 150 5.90 140 13.7 148 1.21 144 15.1 150 24.6 150 1.24 138 6.24 147 25.2 150 0.49 147 5.51 138 28.7 137 0.21 131 10.6 147 26.4 148 0.24 134
Periodicity [78]149.9 2.65 147 10.3 147 0.09 153 9.86 153 13.0 153 0.95 152 7.07 153 15.7 153 2.07 153 9.47 152 22.6 152 1.94 151 16.9 153 27.6 153 1.35 153 6.22 146 24.5 145 0.41 133 5.73 143 30.9 149 0.58 149 12.2 152 29.3 151 0.40 152
AVG_FLOW_ROB [142]154.0 18.5 154 43.1 154 1.46 154 22.0 154 25.4 154 2.17 154 18.9 154 25.1 154 4.07 154 34.3 154 52.0 154 7.35 154 26.2 154 39.9 154 2.35 154 16.3 154 52.5 154 2.03 154 21.5 154 44.4 154 1.54 154 26.6 154 42.2 154 2.44 154
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.