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        
R5.0
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]16.6 4.93 7 13.9 6 0.13 6 8.97 31 17.1 16 0.43 10 6.00 14 13.4 8 0.27 1 17.6 4 26.2 9 5.24 23 43.0 19 57.7 13 5.17 25 10.3 9 39.1 12 0.87 19 9.75 20 41.0 20 0.44 24 21.5 37 51.9 45 0.47 20
MDP-Flow2 [68]17.7 4.89 6 14.4 8 0.12 5 8.58 13 16.9 14 0.39 3 5.95 11 13.6 10 0.28 3 17.7 5 26.7 17 5.32 40 42.9 11 57.6 9 5.13 20 10.6 26 40.1 32 0.92 28 9.75 20 41.0 20 0.43 14 21.6 50 51.9 45 0.46 14
CtxSyn [137]17.9 3.42 2 9.96 2 0.08 1 6.79 1 13.5 2 0.50 31 2.17 1 5.94 1 0.43 63 15.1 1 23.3 1 4.57 1 42.2 5 56.7 5 4.54 5 10.0 6 36.5 5 0.63 4 13.5 152 44.0 115 0.40 6 21.1 8 49.2 5 0.43 6
SepConv-v1 [127]18.9 3.41 1 11.0 4 0.08 1 8.39 9 16.7 13 1.04 114 2.81 2 7.63 2 0.74 118 18.0 19 25.2 3 5.82 113 42.9 11 57.4 6 4.74 6 9.03 3 34.1 3 0.60 3 9.34 4 38.6 3 0.42 8 20.1 3 48.6 3 0.35 2
SuperSlomo [132]19.4 3.75 4 10.1 3 0.19 67 8.96 30 16.5 9 1.31 130 3.32 3 8.42 3 0.29 7 17.7 5 24.1 2 5.32 40 41.4 3 55.9 3 4.24 3 9.50 4 35.3 4 0.67 5 10.8 119 40.3 7 0.37 3 20.4 4 48.7 4 0.42 4
CyclicGen [154]19.7 3.50 3 9.95 1 0.13 6 7.67 2 12.9 1 1.52 144 3.74 5 10.6 5 0.48 78 19.1 95 25.7 4 5.85 116 36.7 1 49.3 1 3.80 1 5.69 1 21.5 1 0.50 2 7.53 1 30.4 1 0.26 1 13.5 1 33.4 1 0.28 1
NNF-Local [87]22.5 5.11 20 15.7 32 0.11 4 8.18 4 15.8 6 0.39 3 6.01 17 13.5 9 0.27 1 18.3 36 28.3 59 5.29 32 43.0 19 57.6 9 5.11 18 10.8 50 40.9 65 1.01 55 9.67 12 40.7 13 0.46 40 21.2 10 51.2 13 0.46 14
NN-field [71]22.7 5.14 24 16.1 56 0.13 6 8.21 5 15.7 5 0.38 2 6.39 50 13.6 10 0.30 8 18.4 40 28.7 76 5.33 45 42.9 11 57.6 9 5.08 15 10.7 34 40.2 38 0.94 37 9.62 9 40.5 10 0.44 24 21.2 10 51.2 13 0.45 8
FGIK [136]26.1 4.55 5 14.6 10 0.09 3 10.5 87 19.3 48 1.09 118 3.51 4 9.66 4 0.67 115 20.6 130 28.7 76 5.10 3 37.5 2 51.1 2 4.02 2 7.84 2 30.4 2 0.49 1 8.54 2 35.1 2 0.32 2 19.5 2 47.3 2 0.37 3
Layers++ [37]26.8 5.25 38 15.9 41 0.17 46 8.27 6 15.5 4 0.37 1 6.16 27 14.3 17 0.38 43 18.0 19 26.9 22 5.32 40 43.1 30 57.9 26 5.24 44 10.7 34 40.6 58 0.97 49 9.70 13 40.7 13 0.39 4 21.3 17 51.3 16 0.48 34
PH-Flow [101]29.8 5.32 51 16.4 68 0.16 35 8.28 7 15.9 7 0.44 13 6.12 24 13.9 13 0.33 17 17.5 2 25.8 5 5.15 10 42.8 9 57.5 7 5.03 12 11.0 76 41.6 97 1.09 79 9.71 14 41.0 20 0.46 40 21.3 17 51.4 19 0.50 72
nLayers [57]31.9 5.26 41 15.8 38 0.16 35 8.54 11 16.6 11 0.45 16 5.89 8 13.1 6 0.30 8 18.1 25 27.1 27 5.35 51 43.3 61 58.0 38 5.36 81 10.8 50 40.9 65 1.11 84 9.65 11 40.1 5 0.48 65 21.2 10 51.1 11 0.45 8
COFM [59]31.9 5.08 17 15.1 17 0.19 67 8.86 22 17.4 21 0.48 25 6.37 46 14.2 16 0.40 51 17.7 5 26.2 9 5.11 4 42.9 11 57.8 17 5.02 11 10.9 62 41.6 97 1.11 84 9.24 3 38.8 4 0.50 81 21.5 37 51.9 45 0.46 14
Sparse-NonSparse [56]32.2 5.31 50 16.3 66 0.17 46 8.74 17 17.2 20 0.48 25 6.19 28 14.7 29 0.34 22 17.9 14 26.3 11 5.23 20 43.1 30 57.8 17 5.25 47 11.0 76 41.2 75 1.04 63 9.71 14 40.9 17 0.46 40 21.2 10 51.3 16 0.47 20
IROF++ [58]33.8 5.37 63 16.8 84 0.14 14 8.87 24 17.4 21 0.45 16 6.41 56 14.6 26 0.43 63 17.5 2 25.8 5 5.22 15 42.9 11 57.8 17 5.19 30 10.5 16 39.4 18 0.87 19 10.0 62 42.4 70 0.47 58 21.4 29 51.5 21 0.50 72
TV-L1-MCT [64]34.7 5.74 119 18.1 122 0.18 59 9.50 48 19.1 45 0.58 44 5.73 6 14.5 24 0.38 43 17.8 10 26.0 8 5.28 30 43.0 19 57.9 26 5.22 38 10.4 11 39.1 12 0.94 37 9.78 26 41.1 26 0.44 24 21.2 10 51.1 11 0.48 34
HAST [109]35.7 5.12 21 15.2 19 0.16 35 8.74 17 17.1 16 0.43 10 6.62 83 15.3 54 0.39 47 17.7 5 26.4 13 4.98 2 43.0 19 58.0 38 5.05 13 11.0 76 41.4 85 1.06 71 9.53 5 40.4 8 0.42 8 22.0 97 52.8 94 0.47 20
ComponentFusion [96]37.1 5.15 25 16.1 56 0.14 14 8.86 22 17.9 29 0.41 6 6.38 47 15.4 55 0.33 17 17.8 10 27.0 25 5.15 10 43.2 51 58.0 38 5.24 44 10.6 26 39.8 22 0.94 37 10.0 62 42.7 89 0.57 110 21.5 37 51.8 38 0.47 20
ProbFlowFields [128]37.8 5.03 12 15.6 29 0.17 46 8.55 12 17.1 16 0.41 6 6.00 14 14.4 20 0.32 14 18.1 25 27.1 27 5.38 56 43.3 61 58.1 56 5.49 127 10.9 62 41.2 75 1.20 102 9.61 8 40.7 13 0.47 58 21.0 7 50.8 8 0.49 52
FMOF [94]38.0 5.62 105 17.2 96 0.21 79 8.71 16 17.0 15 0.44 13 6.38 47 14.7 29 0.46 72 18.6 55 28.0 45 5.31 37 43.1 30 57.9 26 5.15 23 10.8 50 40.5 54 0.87 19 9.60 7 40.4 8 0.40 6 21.5 37 51.7 30 0.46 14
2DHMM-SAS [92]41.3 5.62 105 17.6 113 0.18 59 10.1 73 19.7 63 0.64 60 5.73 6 14.4 20 0.37 41 17.7 5 25.9 7 5.30 34 43.0 19 57.8 17 5.26 50 10.7 34 40.0 30 0.82 10 9.83 32 41.3 29 0.48 65 21.6 50 52.0 49 0.47 20
CombBMOF [113]41.9 5.46 78 16.2 64 0.22 91 8.89 25 18.0 31 0.45 16 6.29 35 14.7 29 0.40 51 18.5 49 28.0 45 5.24 23 43.0 19 57.7 13 5.08 15 10.8 50 40.2 38 0.82 10 11.7 142 42.9 97 0.47 58 21.2 10 50.9 9 0.45 8
LSM [39]42.4 5.49 83 17.4 106 0.18 59 8.93 27 17.7 26 0.48 25 6.32 39 15.4 55 0.35 31 18.1 25 27.1 27 5.22 15 43.1 30 57.9 26 5.28 60 11.0 76 41.3 81 1.03 61 9.72 16 40.9 17 0.46 40 21.4 29 51.7 30 0.48 34
Ramp [62]44.1 5.46 78 17.1 93 0.18 59 8.84 20 17.4 21 0.58 44 6.14 25 14.7 29 0.34 22 17.8 10 26.4 13 5.23 20 43.2 51 58.0 38 5.27 55 11.2 99 42.0 108 1.15 92 9.72 16 40.9 17 0.42 8 21.6 50 52.1 56 0.48 34
DeepFlow [86]44.8 5.06 16 14.6 10 0.19 67 9.80 63 19.5 53 0.75 77 6.45 59 16.6 91 0.35 31 18.7 65 27.6 38 5.41 65 43.4 78 58.0 38 5.37 84 10.3 9 38.3 8 0.99 51 9.83 32 41.8 47 0.43 14 21.3 17 51.6 28 0.48 34
NNF-EAC [103]45.0 5.52 87 15.7 32 0.34 133 9.27 41 18.1 33 0.48 25 6.53 66 13.8 12 0.40 51 18.2 32 27.0 25 5.71 102 43.0 19 57.7 13 5.11 18 10.4 11 39.1 12 0.83 13 9.89 41 41.6 41 0.52 95 21.7 63 52.2 64 0.49 52
DeepFlow2 [108]45.5 5.16 26 14.9 16 0.21 79 9.81 64 19.7 63 0.65 62 6.38 47 16.3 80 0.34 22 18.6 55 28.1 50 5.29 32 43.4 78 58.0 38 5.37 84 10.2 8 38.4 9 0.85 16 9.96 57 42.1 63 0.44 24 21.4 29 51.8 38 0.49 52
LME [70]45.8 5.13 23 15.8 38 0.14 14 9.15 38 18.4 42 0.51 32 6.32 39 15.7 63 0.34 22 17.9 14 27.1 27 5.34 47 43.8 131 58.8 129 5.79 145 10.8 50 41.2 75 0.93 32 9.86 37 41.3 29 0.43 14 21.3 17 51.5 21 0.47 20
SuperFlow [81]45.9 4.99 11 14.3 7 0.22 91 10.3 80 19.9 68 0.90 95 6.61 76 15.5 59 0.51 83 18.5 49 27.2 32 5.52 84 43.3 61 58.1 56 5.37 84 10.1 7 38.0 7 0.73 7 9.73 19 41.4 34 0.46 40 21.3 17 51.5 21 0.46 14
FlowFields+ [130]46.0 5.23 37 16.6 76 0.15 22 8.91 26 18.3 36 0.45 16 6.28 34 15.9 67 0.34 22 18.2 32 28.1 50 5.34 47 43.4 78 58.2 68 5.35 77 10.9 62 41.6 97 1.10 82 9.79 27 41.5 36 0.46 40 21.3 17 51.5 21 0.48 34
WLIF-Flow [93]46.2 5.25 38 16.0 49 0.15 22 9.14 37 18.1 33 0.59 50 6.29 35 14.3 17 0.34 22 17.9 14 26.3 11 5.65 98 43.1 30 57.9 26 5.26 50 11.2 99 41.9 107 1.22 108 9.82 31 41.3 29 0.44 24 21.7 63 52.2 64 0.49 52
PGM-C [120]46.3 5.18 29 16.0 49 0.15 22 8.97 31 18.2 35 0.46 22 6.51 62 16.4 86 0.33 17 18.4 40 28.5 67 5.36 53 43.4 78 58.1 56 5.40 100 10.7 34 40.5 54 0.96 46 9.92 46 41.9 51 0.45 32 21.4 29 51.8 38 0.48 34
EAI-Flow [152]46.7 5.33 57 15.9 41 0.17 46 9.73 61 19.6 59 0.71 70 6.61 76 16.3 80 0.36 36 18.3 36 28.1 50 5.11 4 43.1 30 57.9 26 5.31 63 10.7 34 39.9 26 0.98 50 10.1 72 42.7 89 0.51 90 21.1 8 50.9 9 0.44 7
FlowFields [110]47.0 5.22 35 16.5 71 0.16 35 8.95 28 18.3 36 0.42 8 6.29 35 15.9 67 0.35 31 18.4 40 28.5 67 5.41 65 43.4 78 58.1 56 5.33 67 10.9 62 41.3 81 1.08 75 9.79 27 41.5 36 0.45 32 21.3 17 51.6 28 0.49 52
Classic+NL [31]47.4 5.56 95 17.4 106 0.22 91 8.99 33 17.6 25 0.54 36 6.02 18 14.7 29 0.36 36 18.1 25 26.8 18 5.41 65 43.1 30 58.0 38 5.23 41 11.1 94 41.5 91 1.06 71 9.72 16 41.0 20 0.46 40 21.6 50 52.0 49 0.47 20
JOF [141]48.8 5.53 91 16.9 88 0.21 79 8.65 14 16.6 11 0.48 25 6.08 20 14.0 14 0.34 22 18.1 25 26.8 18 5.59 91 43.4 78 58.2 68 5.45 116 11.1 94 41.4 85 1.04 63 9.64 10 40.6 11 0.43 14 21.6 50 52.0 49 0.48 34
DF-Auto [115]48.9 5.03 12 13.8 5 0.17 46 10.2 74 19.3 48 0.79 81 6.09 21 14.4 20 0.34 22 18.7 65 28.1 50 5.24 23 43.2 51 57.9 26 5.31 63 10.4 11 39.3 15 0.93 32 10.1 72 42.3 67 0.49 71 21.9 88 52.9 101 0.53 110
S2F-IF [123]49.5 5.22 35 16.5 71 0.15 22 8.84 20 18.0 31 0.44 13 6.27 33 15.7 63 0.33 17 18.3 36 28.3 59 5.14 9 43.4 78 58.2 68 5.41 104 11.0 76 41.5 91 1.11 84 9.91 45 41.9 51 0.47 58 21.3 17 51.5 21 0.51 86
FC-2Layers-FF [74]49.7 5.40 68 17.0 91 0.17 46 8.15 3 15.3 3 0.42 8 6.14 25 14.9 37 0.35 31 18.1 25 27.2 32 5.31 37 43.3 61 58.2 68 5.36 81 11.2 99 42.2 112 1.20 102 9.75 20 41.0 20 0.49 71 21.7 63 52.1 56 0.48 34
AGIF+OF [85]49.8 5.60 101 17.4 106 0.15 22 8.95 28 17.7 26 0.59 50 6.20 30 14.5 24 0.43 63 17.9 14 26.6 16 5.22 15 43.4 78 58.3 92 5.38 91 11.1 94 42.0 108 1.01 55 9.87 40 40.7 13 0.42 8 21.5 37 52.0 49 0.48 34
OFLAF [77]50.6 5.16 26 15.9 41 0.14 14 8.28 7 16.1 8 0.40 5 6.34 44 14.9 37 0.30 8 18.0 19 27.3 34 5.11 4 43.3 61 58.1 56 5.39 93 11.2 99 42.4 114 1.21 105 10.1 72 42.4 70 0.60 118 21.9 88 52.6 83 0.45 8
MDP-Flow [26]51.8 5.03 12 15.4 21 0.14 14 8.68 15 17.4 21 0.47 23 5.97 12 14.3 17 0.32 14 18.9 82 28.5 67 5.50 81 43.2 51 58.0 38 5.39 93 11.2 99 42.6 117 1.31 119 10.3 92 43.1 104 0.49 71 21.4 29 51.7 30 0.47 20
S2D-Matching [84]52.4 5.56 95 17.3 100 0.18 59 9.96 69 19.9 68 0.66 64 5.99 13 14.7 29 0.41 56 17.9 14 26.4 13 5.40 62 43.2 51 58.0 38 5.17 25 11.2 99 42.0 108 1.17 97 9.93 49 41.1 26 0.43 14 21.5 37 51.8 38 0.48 34
TF+OM [100]53.8 4.98 9 14.6 10 0.20 73 9.03 35 17.9 29 0.55 38 6.29 35 16.2 76 0.39 47 18.5 49 28.0 45 5.50 81 43.3 61 58.1 56 5.47 122 10.6 26 39.8 22 1.03 61 9.86 37 42.0 57 0.51 90 21.7 63 52.3 69 0.52 100
Brox et al. [5]54.2 5.33 57 15.4 21 0.19 67 10.2 74 20.1 73 0.64 60 6.61 76 17.2 106 0.46 72 18.7 65 28.2 55 5.21 12 43.4 78 58.1 56 5.27 55 10.7 34 40.1 32 0.99 51 9.90 43 42.0 57 0.45 32 21.6 50 52.1 56 0.47 20
ProFlow_ROB [147]54.5 5.09 18 15.4 21 0.17 46 9.40 47 19.3 48 0.55 38 6.34 44 15.4 55 0.33 17 18.4 40 28.7 76 5.39 60 43.5 105 58.3 92 5.41 104 10.4 11 39.3 15 0.79 8 10.2 84 42.9 97 0.49 71 21.8 76 52.6 83 0.49 52
ALD-Flow [66]54.6 5.37 63 16.1 56 0.23 99 9.53 49 19.2 47 0.57 42 6.51 62 16.7 95 0.34 22 18.2 32 27.9 41 5.32 40 43.4 78 58.3 92 5.46 119 10.7 34 39.9 26 0.99 51 9.76 25 41.2 28 0.44 24 21.8 76 52.7 90 0.47 20
CPM-Flow [116]55.3 5.20 34 16.1 56 0.16 35 8.99 33 18.3 36 0.47 23 6.42 57 16.0 72 0.30 8 18.8 74 29.2 100 5.43 72 43.4 78 58.2 68 5.44 114 10.6 26 40.1 32 1.02 57 10.0 62 42.6 81 0.45 32 21.4 29 51.8 38 0.53 110
DMF_ROB [140]55.5 5.30 48 15.8 38 0.20 73 10.2 74 20.5 81 0.73 73 7.26 116 18.0 122 0.75 119 18.9 82 28.8 82 5.40 62 43.1 30 57.9 26 5.34 73 10.5 16 39.8 22 0.92 28 9.98 60 41.5 36 0.43 14 21.3 17 51.4 19 0.47 20
SVFilterOh [111]55.8 5.32 51 15.7 32 0.21 79 8.78 19 17.1 16 0.49 30 6.40 53 14.6 26 0.38 43 18.4 40 27.1 27 5.80 111 43.8 131 58.6 123 5.65 139 10.9 62 41.0 71 1.04 63 9.54 6 40.1 5 0.43 14 21.7 63 52.2 64 0.50 72
AggregFlow [97]56.0 5.64 108 17.2 96 0.22 91 9.81 64 19.5 53 0.59 50 6.11 23 14.4 20 0.28 3 18.9 82 29.0 91 5.30 34 43.4 78 58.2 68 5.33 67 10.7 34 40.2 38 0.96 46 9.89 41 41.7 44 0.50 81 21.4 29 51.7 30 0.50 72
RNLOD-Flow [121]56.9 5.32 51 16.6 76 0.16 35 9.70 58 19.6 59 0.60 54 6.57 70 15.5 59 0.51 83 18.2 32 27.4 35 5.22 15 43.1 30 58.0 38 5.28 60 11.0 76 41.4 85 1.08 75 9.85 35 41.3 29 0.50 81 21.9 88 52.7 90 0.49 52
Second-order prior [8]58.4 5.29 46 15.3 20 0.27 116 10.8 95 21.1 91 0.78 80 7.14 109 17.8 120 0.62 109 18.6 55 28.3 59 5.21 12 42.9 11 57.7 13 5.16 24 10.5 16 39.6 20 0.93 32 10.2 84 42.8 94 0.44 24 21.6 50 52.3 69 0.49 52
IROF-TV [53]59.2 5.35 62 16.6 76 0.21 79 9.10 36 17.8 28 0.57 42 6.61 76 16.8 97 0.44 67 17.8 10 26.9 22 5.37 55 43.5 105 58.4 107 5.50 129 10.5 16 40.1 32 0.90 25 9.98 60 42.2 65 0.46 40 21.6 50 52.1 56 0.51 86
DPOF [18]60.0 5.51 86 17.9 120 0.22 91 8.45 10 16.5 9 0.43 10 6.87 93 15.1 47 0.59 100 18.9 82 29.5 106 5.43 72 42.9 11 57.8 17 5.05 13 11.0 76 40.9 65 0.84 15 10.3 92 42.5 77 0.45 32 21.9 88 52.8 94 0.48 34
TC-Flow [46]61.8 5.19 30 15.9 41 0.21 79 9.57 50 19.6 59 0.63 57 6.78 91 17.0 103 0.36 36 18.1 25 27.4 35 5.61 94 43.3 61 58.2 68 5.46 119 11.0 76 41.6 97 1.18 98 9.93 49 41.7 44 0.45 32 21.5 37 52.0 49 0.49 52
Aniso. Huber-L1 [22]62.5 5.41 70 16.0 49 0.23 99 11.2 106 21.1 91 0.90 95 6.72 86 15.4 55 0.46 72 18.5 49 28.1 50 5.39 60 43.0 19 57.8 17 5.23 41 10.5 16 40.1 32 0.81 9 10.2 84 42.6 81 0.46 40 21.9 88 52.7 90 0.52 100
OAR-Flow [125]62.7 5.28 44 15.5 25 0.18 59 9.71 60 19.5 53 0.67 65 6.43 58 16.3 80 0.28 3 18.0 19 27.6 38 5.23 20 43.5 105 58.4 107 5.48 125 10.9 62 41.3 81 1.13 89 10.2 84 42.9 97 0.51 90 21.7 63 52.3 69 0.45 8
EpicFlow [102]63.0 5.19 30 16.1 56 0.15 22 9.60 51 19.8 67 0.58 44 6.40 53 16.4 86 0.35 31 18.6 55 29.1 98 5.47 78 43.4 78 58.2 68 5.42 108 10.8 50 41.2 75 1.08 75 10.1 72 42.5 77 0.54 101 21.5 37 52.0 49 0.49 52
ComplOF-FED-GPU [35]63.2 5.30 48 16.1 56 0.19 67 9.39 45 19.3 48 0.58 44 7.21 113 16.9 100 0.66 112 18.4 40 28.6 73 5.32 40 43.1 30 58.0 38 5.27 55 10.8 50 40.9 65 0.99 51 10.1 72 42.8 94 0.47 58 21.8 76 52.3 69 0.50 72
FF++_ROB [146]64.2 5.19 30 16.1 56 0.13 6 9.36 44 19.0 44 0.51 32 6.52 65 16.2 76 0.46 72 18.6 55 28.8 82 5.41 65 43.4 78 58.2 68 5.44 114 11.3 107 41.2 75 1.71 144 9.85 35 41.8 47 0.49 71 21.3 17 51.5 21 0.57 138
FESL [72]64.6 5.65 111 17.3 100 0.17 46 9.18 39 18.3 36 0.55 38 6.22 31 15.0 43 0.44 67 18.8 74 28.4 62 5.38 56 43.4 78 58.2 68 5.41 104 11.3 107 42.8 121 1.19 100 9.92 46 41.5 36 0.42 8 21.8 76 52.3 69 0.48 34
Classic+CPF [83]65.4 5.59 100 17.3 100 0.16 35 9.22 40 18.3 36 0.58 44 6.00 14 14.9 37 0.40 51 18.0 19 26.8 18 5.22 15 43.5 105 58.5 116 5.38 91 11.4 114 43.0 130 1.15 92 10.1 72 41.9 51 0.45 32 22.0 97 53.1 108 0.49 52
PMF [73]66.0 5.32 51 16.6 76 0.14 14 9.67 57 19.9 68 0.45 16 6.89 98 18.2 126 0.49 80 18.4 40 27.9 41 5.21 12 43.5 105 58.4 107 5.22 38 11.0 76 40.5 54 1.27 115 9.86 37 41.8 47 0.46 40 22.1 106 53.1 108 0.50 72
RFlow [90]67.8 5.19 30 16.1 56 0.23 99 10.8 95 21.2 95 0.85 90 6.59 74 16.0 72 0.51 83 18.8 74 28.8 82 5.47 78 43.1 30 58.0 38 5.21 36 10.5 16 40.0 30 0.93 32 10.0 62 42.6 81 0.49 71 22.1 106 53.2 112 0.51 86
Local-TV-L1 [65]67.9 5.29 46 14.6 10 0.35 135 11.5 114 21.1 91 1.23 125 6.39 50 14.9 37 0.37 41 19.0 89 27.9 41 6.64 134 43.3 61 58.3 92 5.33 67 10.9 62 39.0 10 1.58 143 9.79 27 41.6 41 0.48 65 21.3 17 51.5 21 0.53 110
PWC-Net_ROB [148]69.0 5.47 81 18.4 128 0.13 6 9.99 70 20.9 88 0.53 34 6.74 87 17.5 114 0.41 56 18.3 36 28.8 82 5.25 27 43.5 105 58.3 92 5.45 116 11.2 99 41.0 71 1.22 108 9.93 49 41.8 47 0.46 40 21.3 17 51.3 16 0.51 86
CLG-TV [48]69.4 5.32 51 15.7 32 0.26 113 11.0 102 21.2 95 0.83 88 6.75 89 16.6 91 0.56 94 18.9 82 28.4 62 5.50 81 43.3 61 58.1 56 5.25 47 10.5 16 39.8 22 0.87 19 10.1 72 42.5 77 0.44 24 22.0 97 53.1 108 0.51 86
EPPM w/o HM [88]69.7 5.34 60 17.3 100 0.13 6 9.73 61 20.1 73 0.53 34 7.33 123 18.7 132 0.63 110 18.5 49 29.1 98 5.33 45 43.1 30 58.0 38 5.20 33 11.0 76 41.4 85 0.96 46 10.3 92 42.3 67 0.56 107 21.8 76 52.4 79 0.49 52
TriFlow [95]69.7 5.42 71 17.0 91 0.24 105 10.9 98 21.2 95 0.91 97 6.61 76 16.8 97 0.36 36 18.9 82 29.0 91 5.28 30 43.2 51 58.2 68 5.37 84 11.0 76 40.9 65 0.95 41 9.96 57 41.7 44 0.49 71 21.7 63 52.2 64 0.47 20
Classic++ [32]69.9 5.33 57 16.0 49 0.28 117 10.2 74 20.3 77 0.69 68 6.87 93 16.6 91 0.50 81 18.7 65 27.7 40 5.64 96 43.2 51 58.0 38 5.26 50 11.0 76 40.7 61 1.34 122 9.93 49 41.9 51 0.47 58 21.7 63 52.4 79 0.50 72
SIOF [67]70.7 5.64 108 16.5 71 0.28 117 11.3 108 21.6 107 0.91 97 6.32 39 15.9 67 0.42 58 18.7 65 28.4 62 5.36 53 43.0 19 57.9 26 5.17 25 10.7 34 40.2 38 0.95 41 10.1 72 42.4 70 0.50 81 22.2 116 53.2 112 0.53 110
Efficient-NL [60]71.2 5.54 93 17.1 93 0.16 35 9.60 51 18.9 43 0.56 41 6.99 104 15.1 47 0.75 119 18.8 74 28.2 55 5.26 28 43.1 30 57.9 26 5.25 47 11.6 120 43.4 138 1.04 63 10.1 72 42.5 77 0.48 65 22.6 128 53.8 126 0.48 34
LDOF [28]71.6 5.53 91 15.6 29 0.32 129 11.1 104 20.3 77 1.45 142 6.89 98 17.3 108 0.59 100 19.0 89 28.9 87 5.63 95 43.4 78 58.2 68 5.40 100 10.4 11 39.0 10 0.83 13 9.92 46 42.4 70 0.46 40 21.6 50 52.3 69 0.46 14
ContinualFlow_ROB [153]71.9 5.85 124 19.2 134 0.16 35 10.4 84 21.5 102 0.82 84 7.31 121 18.8 134 0.51 83 18.7 65 29.7 110 5.52 84 43.1 30 58.1 56 5.33 67 10.5 16 40.3 43 0.86 17 9.97 59 41.6 41 0.43 14 21.5 37 52.1 56 0.55 130
Complementary OF [21]72.3 5.28 44 16.7 82 0.15 22 9.39 45 19.5 53 0.58 44 7.53 127 16.3 80 1.10 138 18.7 65 29.0 91 5.35 51 43.2 51 58.2 68 5.26 50 10.9 62 41.2 75 1.16 95 10.3 92 43.4 109 0.55 104 21.5 37 52.2 64 0.51 86
p-harmonic [29]72.3 5.17 28 15.5 25 0.16 35 11.2 106 21.4 100 0.94 102 6.55 67 17.4 113 0.55 93 19.2 98 28.6 73 5.45 76 43.3 61 58.2 68 5.27 55 10.7 34 40.2 38 1.04 63 10.4 100 43.4 109 0.50 81 21.8 76 52.6 83 0.49 52
CostFilter [40]72.5 5.44 73 17.7 115 0.13 6 9.64 54 20.1 73 0.45 16 6.96 102 19.1 135 0.47 76 18.5 49 28.9 87 5.13 8 43.6 121 58.5 116 5.32 66 11.1 94 40.5 54 1.48 135 9.94 54 42.1 63 0.45 32 21.8 76 52.6 83 0.49 52
F-TV-L1 [15]72.9 5.56 95 16.0 49 0.36 139 11.4 112 21.5 102 0.94 102 6.88 95 17.0 103 0.66 112 18.7 65 27.9 41 5.79 110 42.6 6 57.8 17 5.01 10 10.6 26 39.3 15 1.02 57 10.0 62 41.9 51 0.55 104 22.0 97 52.8 94 0.51 86
OFH [38]73.2 5.49 83 16.6 76 0.25 109 10.3 80 20.2 76 0.77 79 6.88 95 17.8 120 0.36 36 18.4 40 28.9 87 5.24 23 43.1 30 58.0 38 5.26 50 10.9 62 41.5 91 1.18 98 10.3 92 43.0 101 0.58 113 21.6 50 52.1 56 0.50 72
TC/T-Flow [76]74.2 5.73 117 17.3 100 0.22 91 9.66 56 19.7 63 0.63 57 6.24 32 14.9 37 0.32 14 18.6 55 28.7 76 5.38 56 43.5 105 58.4 107 5.50 129 11.0 76 41.4 85 0.89 24 10.2 84 43.0 101 0.58 113 21.9 88 53.0 106 0.45 8
CBF [12]74.2 4.98 9 14.8 15 0.18 59 10.2 74 19.9 68 0.71 70 6.63 84 15.2 51 0.42 58 19.0 89 28.5 67 6.39 130 43.4 78 58.3 92 5.49 127 10.7 34 40.4 49 0.95 41 10.1 72 42.6 81 0.50 81 22.3 121 53.5 122 0.53 110
HBM-GC [105]74.4 5.52 87 17.1 93 0.22 91 9.64 54 19.3 48 0.59 50 5.93 10 13.2 7 0.31 13 18.8 74 28.0 45 5.83 115 44.3 143 59.2 136 5.71 141 11.5 117 43.3 136 1.32 120 9.75 20 40.6 11 0.39 4 22.0 97 52.9 101 0.50 72
LFNet_ROB [150]74.4 5.45 75 17.6 113 0.13 6 10.4 84 21.2 95 0.73 73 6.75 89 18.1 124 0.47 76 18.4 40 28.7 76 5.27 29 43.1 30 58.0 38 5.20 33 11.1 94 41.8 105 1.10 82 10.4 100 42.7 89 0.50 81 21.7 63 52.0 49 0.60 142
Steered-L1 [118]74.8 5.12 21 16.0 49 0.17 46 9.62 53 19.5 53 0.88 92 7.15 110 15.6 61 1.00 130 19.4 105 28.5 67 6.39 130 43.5 105 58.5 116 5.19 30 10.8 50 40.8 64 1.20 102 9.95 56 42.6 81 0.52 95 21.7 63 52.6 83 0.48 34
GraphCuts [14]75.8 5.98 129 17.5 111 0.24 105 10.0 71 19.5 53 0.76 78 8.24 140 14.6 26 1.06 133 19.7 111 29.0 91 5.69 100 42.9 11 57.9 26 4.97 8 10.5 16 40.3 43 0.87 19 10.0 62 42.4 70 0.58 113 22.1 106 53.2 112 0.51 86
MLDP_OF [89]76.5 5.44 73 17.2 96 0.17 46 9.84 66 19.9 68 0.62 56 6.19 28 14.8 35 0.28 3 18.6 55 27.4 35 5.71 102 43.3 61 58.2 68 5.34 73 11.9 131 43.3 136 1.57 142 10.4 100 42.6 81 0.56 107 21.7 63 52.3 69 0.59 141
AdaConv-v1 [126]77.2 6.72 141 21.8 145 0.25 109 12.8 132 22.4 126 1.80 148 8.18 139 18.4 128 1.46 146 24.3 146 34.7 148 7.39 142 41.5 4 56.1 4 4.28 4 9.57 5 36.9 6 0.71 6 9.75 20 41.0 20 0.60 118 20.5 5 49.7 6 0.42 4
SRR-TVOF-NL [91]77.4 5.70 114 16.9 88 0.23 99 10.3 80 21.0 90 0.88 92 6.57 70 16.1 74 0.39 47 19.2 98 28.7 76 5.12 7 43.2 51 58.3 92 5.27 55 10.8 50 40.9 65 0.86 17 10.6 114 42.3 67 0.46 40 22.5 124 53.8 126 0.54 121
BlockOverlap [61]78.2 5.34 60 14.6 10 0.41 144 11.4 112 20.6 82 1.42 138 6.49 60 14.1 15 0.61 107 18.9 82 26.9 22 7.34 141 44.2 141 58.9 132 5.91 147 11.0 76 39.9 26 1.39 129 9.81 30 41.3 29 0.46 40 21.5 37 51.7 30 0.51 86
Sparse Occlusion [54]78.7 5.43 72 16.8 84 0.23 99 10.3 80 20.8 86 0.63 57 6.51 62 15.0 43 0.44 67 19.0 89 29.0 91 5.42 70 43.4 78 58.2 68 5.41 104 11.3 107 42.9 127 1.14 90 10.1 72 42.2 65 0.42 8 22.1 106 53.2 112 0.49 52
CRTflow [80]80.2 5.48 82 16.5 71 0.34 133 10.7 93 20.7 83 0.86 91 7.25 115 18.6 131 0.60 105 18.8 74 28.8 82 5.98 122 43.4 78 58.2 68 5.43 111 10.7 34 40.4 49 0.95 41 9.93 49 42.0 57 0.49 71 21.7 63 52.3 69 0.49 52
LiteFlowNet [143]81.1 5.61 102 18.9 133 0.15 22 9.94 68 20.9 88 0.65 62 6.33 43 17.5 114 0.39 47 19.2 98 30.9 129 5.94 121 43.1 30 57.9 26 5.36 81 11.3 107 42.2 112 1.06 71 10.7 116 43.5 111 0.62 122 21.2 10 51.2 13 0.54 121
SimpleFlow [49]81.9 5.52 87 17.5 111 0.18 59 10.2 74 19.7 63 0.73 73 7.32 122 15.8 65 1.05 132 18.0 19 26.8 18 5.44 74 43.3 61 58.1 56 5.33 67 11.3 107 42.9 127 1.22 108 10.3 92 44.6 124 1.04 147 21.8 76 52.6 83 0.47 20
AugFNG_ROB [144]82.2 5.68 113 18.7 130 0.15 22 10.9 98 21.8 111 0.93 100 7.28 117 20.6 144 0.48 78 19.3 103 30.7 122 5.40 62 43.6 121 58.6 123 5.47 122 10.6 26 40.1 32 0.82 10 10.5 110 43.0 101 0.50 81 20.9 6 50.7 7 0.48 34
FlowNet2 [122]83.0 6.90 143 21.5 144 0.25 109 10.6 91 20.7 83 0.82 84 7.10 107 17.3 108 0.54 89 19.4 105 31.8 136 5.57 89 43.4 78 58.3 92 5.39 93 10.7 34 40.3 43 0.90 25 10.0 62 42.0 57 0.46 40 21.6 50 51.9 45 0.51 86
IAOF [50]83.7 5.97 128 16.8 84 0.29 122 14.1 147 24.8 147 1.41 137 6.05 19 16.2 76 0.61 107 20.1 119 29.5 106 5.47 78 43.0 19 57.8 17 5.19 30 10.7 34 40.3 43 0.94 37 10.4 100 43.3 107 0.46 40 22.0 97 52.8 94 0.54 121
Modified CLG [34]84.0 5.05 15 15.1 17 0.19 67 12.3 128 22.2 120 1.30 129 6.81 92 18.3 127 0.66 112 19.3 103 29.7 110 5.34 47 43.4 78 58.2 68 5.29 62 10.8 50 40.6 58 1.15 92 10.2 84 43.6 112 0.47 58 21.9 88 52.7 90 0.53 110
Aniso-Texture [82]85.2 5.09 18 15.7 32 0.15 22 11.1 104 21.7 108 1.00 107 7.30 120 15.9 67 0.59 100 18.7 65 28.6 73 5.90 117 43.6 121 58.4 107 5.53 132 11.6 120 44.0 143 1.44 133 9.90 43 41.4 34 0.43 14 22.1 106 53.1 108 0.49 52
FlowNetS+ft+v [112]85.9 5.40 68 15.5 25 0.29 122 11.7 120 21.7 108 1.62 145 6.88 95 17.1 105 0.56 94 19.0 89 29.2 100 5.73 106 43.5 105 58.4 107 5.56 134 10.5 16 39.9 26 0.95 41 10.1 72 42.9 97 0.52 95 21.8 76 52.5 82 0.48 34
Occlusion-TV-L1 [63]86.5 5.32 51 16.2 64 0.28 117 11.3 108 21.9 114 0.96 106 6.60 75 16.9 100 0.58 98 19.1 95 28.9 87 5.72 104 43.4 78 58.2 68 5.24 44 10.9 62 40.3 43 1.26 114 10.9 123 42.6 81 0.81 138 21.8 76 52.4 79 0.49 52
EPMNet [133]87.5 6.85 142 22.5 146 0.21 79 10.5 87 20.3 77 0.84 89 7.10 107 17.3 108 0.54 89 19.9 112 33.4 144 5.56 88 43.4 78 58.3 92 5.39 93 11.0 76 41.6 97 0.92 28 10.0 62 42.0 57 0.46 40 21.6 50 51.8 38 0.54 121
Shiralkar [42]88.5 5.73 117 18.1 122 0.21 79 11.6 116 22.0 116 0.88 92 6.74 87 19.9 139 0.73 117 20.3 122 30.1 117 5.46 77 42.6 6 57.5 7 4.99 9 11.3 107 41.5 91 1.35 123 11.0 126 44.9 127 0.67 125 21.5 37 51.7 30 0.48 34
TCOF [69]88.6 5.56 95 16.8 84 0.17 46 11.8 121 22.1 118 1.02 111 6.09 21 15.0 43 0.30 8 19.0 89 29.4 104 5.67 99 43.4 78 58.3 92 5.17 25 11.4 114 43.1 132 1.02 57 11.0 126 43.9 114 0.48 65 23.1 140 55.1 145 0.52 100
HBpMotionGpu [43]88.9 5.80 121 16.3 66 0.42 145 13.1 135 23.8 138 1.34 132 6.32 39 14.9 37 0.38 43 19.9 112 30.4 120 5.80 111 43.1 30 58.3 92 5.39 93 11.3 107 41.0 71 1.21 105 9.94 54 41.9 51 0.43 14 22.1 106 52.9 101 0.53 110
3DFlow [135]89.2 5.58 99 17.4 106 0.16 35 9.35 43 19.1 45 0.61 55 6.93 101 15.0 43 0.44 67 18.6 55 28.4 62 5.54 87 43.4 78 58.2 68 5.40 100 12.1 138 44.7 151 1.35 123 11.3 133 44.6 124 0.57 110 22.4 122 53.7 125 0.50 72
Adaptive [20]89.5 5.50 85 16.7 82 0.30 124 11.8 121 22.2 120 1.02 111 6.58 73 16.5 90 0.53 88 18.6 55 28.0 45 5.60 93 43.5 105 58.3 92 5.21 36 11.0 76 41.3 81 1.09 79 10.4 100 42.8 94 0.46 40 22.2 116 53.5 122 0.54 121
Fusion [6]89.9 5.37 63 16.9 88 0.21 79 9.33 42 18.3 36 0.54 36 6.39 50 15.1 47 0.54 89 20.0 117 29.8 113 5.41 65 43.5 105 59.2 136 5.14 21 11.5 117 43.7 141 1.21 105 10.5 110 44.1 117 0.52 95 23.1 140 55.4 146 0.52 100
CNN-flow-warp+ref [117]90.0 4.95 8 14.4 8 0.22 91 10.9 98 21.2 95 1.23 125 7.43 125 18.0 122 0.79 122 20.9 133 29.8 113 6.84 137 43.5 105 58.3 92 5.57 135 10.7 34 40.3 43 1.22 108 10.3 92 44.4 122 0.67 125 21.6 50 52.1 56 0.47 20
BriefMatch [124]92.1 5.45 75 16.5 71 0.31 128 9.84 66 19.6 59 1.43 139 7.55 129 15.6 61 1.08 135 20.3 122 29.2 100 7.97 149 43.3 61 58.3 92 5.43 111 12.0 135 41.5 91 2.37 150 9.84 34 41.5 36 0.56 107 21.4 29 51.7 30 0.52 100
ResPWCR_ROB [145]92.3 5.54 93 17.8 117 0.20 73 10.6 91 21.5 102 0.80 83 7.77 133 17.7 119 0.44 67 19.4 105 30.7 122 5.93 118 42.7 8 57.6 9 5.10 17 12.4 147 41.7 104 2.49 151 10.9 123 42.7 89 0.58 113 21.7 63 52.3 69 0.52 100
Nguyen [33]93.3 5.63 107 15.9 41 0.23 99 13.8 141 23.8 138 1.37 134 6.89 98 18.7 132 0.59 100 20.8 132 30.8 126 5.44 74 43.1 30 58.1 56 5.14 21 10.6 26 40.4 49 0.93 32 11.9 144 45.9 134 0.73 134 22.0 97 52.8 94 0.52 100
2D-CLG [1]95.5 5.27 43 15.7 32 0.21 79 13.1 135 22.8 128 1.37 134 7.29 118 17.3 108 0.94 128 20.3 122 30.2 118 5.34 47 43.5 105 58.4 107 5.37 84 10.8 50 40.7 61 1.22 108 10.5 110 44.3 120 0.59 117 22.0 97 52.3 69 0.50 72
TV-L1-improved [17]95.6 5.26 41 16.0 49 0.28 117 11.6 116 22.0 116 1.06 116 7.21 113 16.3 80 0.79 122 18.8 74 28.5 67 5.70 101 43.5 105 58.5 116 5.22 38 11.0 76 41.5 91 1.05 69 10.4 100 44.6 124 0.74 136 22.1 106 53.2 112 0.53 110
SPSA-learn [13]96.1 5.45 75 15.4 21 0.25 109 11.6 116 21.4 100 1.15 122 7.65 131 16.6 91 1.26 141 20.1 119 28.2 55 5.30 34 43.3 61 58.2 68 5.42 108 10.9 62 41.0 71 1.14 90 11.6 139 50.4 152 1.71 153 22.2 116 53.3 120 0.49 52
SegOF [10]96.5 5.25 38 15.9 41 0.20 73 10.9 98 20.8 86 0.82 84 8.07 138 18.4 128 1.18 140 20.0 117 32.3 138 5.52 84 43.3 61 58.2 68 5.35 77 11.4 114 43.1 132 1.38 128 10.7 116 46.3 135 0.96 144 21.5 37 51.7 30 0.53 110
IIOF-NLDP [131]96.9 5.65 111 17.8 117 0.15 22 10.5 87 21.5 102 0.72 72 6.98 103 15.2 51 0.42 58 19.5 108 29.3 103 6.15 126 43.1 30 58.0 38 5.20 33 12.2 142 44.1 145 1.54 139 11.9 144 49.2 150 1.34 151 22.2 116 53.0 106 0.50 72
TriangleFlow [30]98.7 5.85 124 18.2 124 0.26 113 11.0 102 21.8 111 0.79 81 7.17 111 16.3 80 0.58 98 19.6 110 30.7 122 5.74 107 42.8 9 57.8 17 4.95 7 11.6 120 42.8 121 1.05 69 10.8 119 45.8 132 0.73 134 22.8 134 54.3 138 0.51 86
Rannacher [23]99.6 5.39 66 16.6 76 0.30 124 11.6 116 22.2 120 1.01 109 7.17 111 16.9 100 0.92 127 18.6 55 28.4 62 5.74 107 43.6 121 58.5 116 5.33 67 11.0 76 41.6 97 1.11 84 10.4 100 44.3 120 0.72 133 21.9 88 52.8 94 0.54 121
Black & Anandan [4]100.0 5.71 115 15.5 25 0.35 135 12.7 131 22.3 124 1.12 119 7.89 134 18.1 124 1.06 133 20.5 128 30.3 119 5.42 70 43.6 121 58.6 123 5.35 77 10.6 26 39.7 21 0.91 27 10.9 123 44.1 117 0.50 81 22.2 116 52.9 101 0.53 110
ROF-ND [107]100.1 6.15 131 16.4 68 0.14 14 10.4 84 21.1 91 0.70 69 7.09 106 15.9 67 0.40 51 20.7 131 32.9 141 5.82 113 43.4 78 58.2 68 5.37 84 11.6 120 43.4 138 1.16 95 11.6 139 46.4 136 0.55 104 22.6 128 53.8 126 0.54 121
TVL1_ROB [139]101.2 5.72 116 15.6 29 0.35 135 13.7 140 23.8 138 1.40 136 6.64 85 17.6 116 0.60 105 20.5 128 29.7 110 5.58 90 43.6 121 58.5 116 5.42 108 10.8 50 40.4 49 1.09 79 10.5 110 44.5 123 0.65 123 21.9 88 52.6 83 0.49 52
OFRF [134]102.5 6.29 134 18.2 124 0.38 141 11.8 121 22.3 124 1.17 124 6.61 76 17.3 108 0.43 63 19.2 98 29.4 104 5.31 37 43.4 78 58.4 107 5.35 77 11.7 127 42.8 121 1.28 116 10.4 100 43.2 105 0.49 71 22.0 97 53.3 120 0.51 86
Ad-TV-NDC [36]103.0 6.08 130 15.9 41 0.60 147 13.0 133 22.8 128 1.36 133 6.55 67 16.4 86 0.56 94 20.9 133 30.6 121 6.29 128 44.1 137 59.0 134 5.43 111 10.7 34 39.4 18 1.11 84 10.4 100 43.3 107 0.51 90 22.1 106 52.9 101 0.53 110
IAOF2 [51]104.2 6.17 132 18.3 127 0.30 124 12.0 124 23.3 134 0.93 100 5.90 9 16.1 74 0.42 58 20.4 127 31.2 134 5.75 109 43.7 130 58.9 132 5.39 93 11.2 99 42.0 108 1.08 75 10.3 92 42.7 89 0.48 65 22.7 131 54.2 135 0.52 100
Correlation Flow [75]104.7 5.61 102 17.8 117 0.15 22 10.8 95 21.7 108 0.82 84 6.40 53 14.8 35 0.42 58 19.1 95 29.0 91 6.04 124 43.9 133 58.6 123 6.05 149 12.0 135 43.9 142 1.29 118 11.0 126 45.3 129 0.70 130 22.5 124 54.1 133 0.51 86
Filter Flow [19]105.9 5.64 108 16.4 68 0.32 129 12.2 127 22.2 120 1.08 117 6.61 76 16.2 76 0.57 97 20.3 122 29.0 91 6.32 129 44.1 137 59.1 135 5.74 143 10.9 62 40.7 61 1.04 63 10.2 84 43.2 105 0.54 101 22.7 131 54.3 138 0.54 121
Bartels [41]107.6 5.52 87 17.2 96 0.40 143 10.0 71 20.7 83 0.94 102 6.50 61 15.8 65 0.54 89 19.9 112 30.0 116 7.79 146 44.8 147 59.2 136 6.72 152 12.8 151 42.4 114 3.06 153 10.0 62 42.0 57 0.54 101 22.1 106 53.2 112 0.54 121
LocallyOriented [52]108.9 5.79 120 17.9 120 0.26 113 12.1 126 23.2 132 1.01 109 7.05 105 17.6 116 0.51 83 19.9 112 30.9 129 5.72 104 43.3 61 58.2 68 5.23 41 11.9 131 42.6 117 1.52 138 10.8 119 44.0 115 0.53 99 22.5 124 54.0 131 0.52 100
Dynamic MRF [7]109.0 5.39 66 17.4 106 0.20 73 10.5 87 21.8 111 0.74 76 7.60 130 20.3 143 0.99 129 21.3 136 31.1 132 7.06 139 43.0 19 58.1 56 5.34 73 11.6 120 43.0 130 1.49 136 10.7 116 45.8 132 0.85 139 22.5 124 53.2 112 0.55 130
ACK-Prior [27]112.5 5.46 78 17.7 115 0.15 22 9.70 58 20.3 77 0.67 65 7.76 132 16.4 86 1.08 135 19.9 112 31.0 131 6.01 123 44.7 146 59.6 143 5.78 144 12.1 138 44.2 147 1.33 121 10.6 114 44.2 119 0.53 99 23.4 145 56.1 150 0.52 100
StereoOF-V1MT [119]112.9 5.94 127 18.8 132 0.20 73 11.3 108 22.6 127 0.94 102 7.95 135 19.6 137 1.00 130 21.6 137 30.7 122 6.76 135 43.3 61 58.3 92 5.37 84 12.1 138 42.6 117 1.82 147 11.6 139 46.7 139 0.90 141 21.8 76 51.8 38 0.50 72
StereoFlow [44]115.5 10.4 153 27.1 153 0.35 135 16.3 152 28.4 153 1.03 113 6.55 67 16.8 97 0.50 81 18.8 74 28.2 55 5.38 56 45.7 152 62.1 152 5.58 136 13.6 152 50.3 153 1.28 116 10.0 62 42.4 70 0.49 71 23.0 136 55.5 147 0.56 136
TI-DOFE [24]115.5 6.39 135 18.7 130 0.36 139 14.8 148 25.5 150 1.66 146 7.45 126 20.2 141 0.78 121 22.8 143 32.5 139 6.04 124 43.2 51 58.4 107 5.17 25 10.9 62 40.4 49 0.92 28 11.2 131 45.6 131 0.65 123 23.2 142 54.2 135 0.65 147
UnFlow [129]115.7 6.39 135 20.9 138 0.21 79 13.0 133 24.4 146 1.15 122 8.06 137 21.1 145 0.82 125 19.2 98 29.6 108 5.64 96 43.1 30 58.0 38 5.40 100 11.8 130 42.8 121 1.36 126 11.0 126 42.4 70 0.70 130 24.3 152 54.8 143 0.70 149
Horn & Schunck [3]116.2 5.81 122 17.3 100 0.21 79 13.1 135 23.5 136 1.26 127 8.03 136 19.7 138 1.08 135 22.6 141 32.7 140 5.59 91 43.6 121 58.7 127 5.39 93 10.9 62 40.6 58 1.02 57 11.7 142 46.5 137 0.60 118 22.8 134 53.9 129 0.55 130
2bit-BM-tele [98]116.2 5.61 102 15.9 41 0.50 146 11.5 114 21.9 114 1.04 114 6.57 70 15.1 47 0.79 122 20.1 119 29.8 113 7.50 143 44.8 147 59.6 143 6.26 150 12.2 142 42.8 121 2.11 149 11.2 131 49.2 150 1.26 149 21.8 76 52.1 56 0.55 130
WRT [151]119.5 5.83 123 18.2 124 0.17 46 11.3 108 21.5 102 0.92 99 8.29 141 15.2 51 1.12 139 19.5 108 29.6 108 5.93 118 43.6 121 58.7 127 5.31 63 12.4 147 45.8 152 1.43 132 12.1 147 51.8 153 1.66 152 22.6 128 54.6 141 0.58 139
WOLF_ROB [149]127.5 6.71 140 21.0 139 0.32 129 12.5 129 23.2 132 1.12 119 7.54 128 17.6 116 0.65 111 20.3 122 33.1 143 6.43 132 43.6 121 58.8 129 5.47 122 11.9 131 42.8 121 1.56 140 12.1 147 46.5 137 0.71 132 22.1 106 52.8 94 0.58 139
NL-TV-NCC [25]129.8 6.44 137 20.3 137 0.24 105 10.7 93 22.1 118 0.68 67 7.38 124 17.2 106 0.59 100 22.2 140 34.7 148 6.82 136 45.5 151 60.2 150 6.68 151 12.3 146 44.6 149 1.19 100 14.4 153 48.1 146 0.67 125 24.0 151 56.4 151 0.55 130
Adaptive flow [45]130.8 7.18 147 19.2 134 0.69 148 15.0 149 25.0 148 2.11 150 7.29 118 16.7 95 0.87 126 22.6 141 31.3 135 7.85 148 44.8 147 60.2 150 5.63 138 11.7 127 43.4 138 1.36 126 10.4 100 43.7 113 0.57 110 23.0 136 54.7 142 0.50 72
SILK [79]130.9 6.21 133 19.3 136 0.39 142 13.8 141 24.0 141 1.73 147 8.85 144 20.2 141 1.41 143 21.8 138 31.1 132 7.10 140 43.5 105 58.5 116 5.45 116 11.9 131 41.4 85 2.03 148 10.8 119 45.5 130 0.77 137 22.4 122 53.2 112 0.60 142
HCIC-L [99]130.9 8.84 152 25.2 151 1.06 152 14.0 145 24.1 143 1.43 139 9.42 147 19.3 136 0.69 116 24.3 146 34.1 146 6.48 133 45.1 150 60.1 148 5.86 146 12.1 138 44.1 145 1.06 71 10.2 84 42.6 81 0.51 90 23.6 148 56.0 149 0.51 86
H+S_ROB [138]131.8 6.50 138 21.0 139 0.14 14 13.6 139 23.3 134 1.26 127 9.85 149 24.2 150 1.51 150 26.5 152 32.2 137 6.19 127 43.9 133 59.3 140 5.46 119 11.7 127 42.9 127 1.23 113 12.1 147 48.5 148 1.02 146 23.5 147 53.9 129 0.55 130
Learning Flow [11]134.2 5.91 126 18.6 129 0.30 124 12.0 124 22.9 130 1.00 107 8.30 142 20.0 140 1.33 142 21.9 139 32.9 141 6.94 138 44.5 145 59.7 146 5.97 148 11.5 117 42.6 117 1.35 123 11.3 133 46.8 140 0.69 129 23.7 149 55.9 148 0.62 145
GroupFlow [9]135.0 7.04 145 22.5 146 0.28 117 12.5 129 24.0 141 1.13 121 9.10 145 22.0 147 1.45 144 21.0 135 33.6 145 5.93 118 44.1 137 59.3 140 5.50 129 12.2 142 44.4 148 1.42 131 11.1 130 45.2 128 0.61 121 22.7 131 54.1 133 0.56 136
SLK [47]136.1 6.55 139 21.1 142 0.32 129 13.5 138 23.1 131 1.44 141 9.16 146 21.2 146 1.49 148 24.9 148 34.2 147 7.81 147 43.5 105 58.8 129 5.34 73 12.2 142 43.1 132 1.45 134 11.9 144 48.9 149 0.96 144 23.0 136 54.0 131 0.64 146
Heeger++ [104]138.5 7.79 150 25.2 151 0.17 46 13.9 144 24.2 144 1.33 131 11.8 151 28.7 152 1.49 148 23.4 144 30.8 126 7.63 144 44.4 144 59.9 147 5.62 137 12.6 150 43.1 132 1.77 146 12.6 151 46.9 141 0.87 140 23.2 142 53.5 122 0.60 142
FFV1MT [106]139.0 6.93 144 22.8 148 0.24 105 14.0 145 23.5 136 1.48 143 11.2 150 27.7 151 1.52 151 23.4 144 30.8 126 7.63 144 44.0 135 59.2 136 5.69 140 12.0 135 41.6 97 1.56 140 12.1 147 47.3 144 0.95 143 23.4 145 54.2 135 0.79 151
FOLKI [16]142.1 7.10 146 21.1 142 0.94 151 15.3 150 25.5 150 2.28 151 8.49 143 22.2 148 1.47 147 26.3 150 35.2 150 10.6 152 44.0 135 59.6 143 5.54 133 11.6 120 41.8 105 1.49 136 11.4 136 47.7 145 0.90 141 23.3 144 54.9 144 0.67 148
Pyramid LK [2]143.5 7.19 148 21.0 139 0.93 150 16.2 151 25.1 149 2.91 152 14.0 152 18.5 130 2.57 152 32.5 153 46.2 153 13.7 153 44.2 141 60.1 148 5.48 125 11.6 120 42.5 116 1.40 130 11.4 136 47.2 143 1.28 150 23.7 149 56.7 152 1.08 152
PGAM+LK [55]144.2 7.51 149 23.5 150 0.73 149 13.8 141 24.2 144 1.92 149 9.44 148 22.7 149 1.45 144 26.4 151 36.9 151 10.5 151 44.1 137 59.5 142 5.72 142 12.4 147 44.0 143 1.75 145 11.3 133 47.0 142 0.68 128 23.0 136 54.5 140 0.76 150
Periodicity [78]151.0 8.05 151 23.2 149 1.34 153 20.5 153 27.4 152 3.39 153 15.2 153 30.5 153 4.22 153 26.2 149 43.5 152 9.47 150 46.4 153 62.7 153 6.92 153 13.7 153 44.6 149 2.88 152 11.4 136 48.3 147 1.18 148 25.7 153 59.2 153 1.29 153
AVG_FLOW_ROB [142]154.0 30.2 154 60.4 154 6.56 154 42.6 154 49.8 154 9.03 154 34.7 154 42.2 154 9.09 154 57.3 154 72.3 154 20.9 154 51.6 154 69.0 154 7.96 154 25.2 154 71.8 154 4.67 154 39.2 154 64.3 154 3.36 154 43.7 154 66.4 154 8.60 154
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.