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        
A95
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]6.6 5.00 3 9.68 12 2.00 2 6.88 21 11.0 10 2.08 1 5.69 4 9.00 2 1.73 1 8.21 4 12.0 4 5.07 2 17.4 3 23.4 3 5.07 10 9.29 6 22.6 3 3.74 19 8.66 9 37.1 11 2.45 1 13.9 13 21.3 13 2.16 1
MDP-Flow2 [68]8.9 4.97 2 9.42 7 2.00 2 6.68 9 11.0 10 2.08 1 5.69 4 9.04 7 1.73 1 8.19 1 12.0 4 5.10 22 17.5 5 23.5 6 5.07 10 9.95 32 24.7 32 3.74 19 8.60 6 36.4 3 2.45 1 13.9 13 21.5 15 2.16 1
NNF-Local [87]17.5 5.07 11 10.1 25 2.00 2 6.40 1 10.0 3 2.08 1 5.69 4 9.00 2 1.73 1 8.66 35 14.5 77 5.10 22 17.6 8 23.8 15 5.07 10 10.4 65 25.8 61 3.74 19 8.66 9 37.5 17 2.45 1 13.9 13 21.6 18 2.16 1
PH-Flow [101]18.2 5.20 37 10.7 55 2.00 2 6.45 3 10.3 5 2.08 1 5.69 4 9.38 13 1.73 1 8.19 1 11.9 1 5.07 2 17.7 21 24.0 26 5.03 6 10.6 78 26.5 79 3.70 1 8.68 12 38.8 46 2.45 1 14.0 20 21.7 22 2.16 1
NN-field [71]18.5 5.07 11 10.4 41 2.00 2 6.45 3 10.0 3 2.08 1 5.97 48 9.00 2 1.73 1 8.76 47 15.0 86 5.10 22 17.6 8 23.7 11 5.07 10 10.1 44 25.0 40 3.74 19 8.54 4 36.9 8 2.45 1 13.9 13 21.6 18 2.16 1
NNF-EAC [103]20.5 5.35 71 10.0 21 2.08 57 7.05 31 11.6 22 2.08 1 6.00 49 9.35 10 1.73 1 8.35 7 12.4 9 5.23 75 17.7 21 23.9 23 5.07 10 9.47 7 22.9 4 3.70 1 8.83 22 37.0 10 2.45 1 14.0 20 21.6 18 2.16 1
IROF++ [58]22.5 5.23 55 10.8 64 2.00 2 6.88 21 11.5 20 2.08 1 6.00 49 10.0 28 1.73 1 8.19 1 11.9 1 5.07 2 17.9 41 24.4 49 5.10 29 9.49 9 24.2 24 3.74 19 9.09 52 37.2 14 2.45 1 14.0 20 22.1 36 2.16 1
SepConv-v1 [127]22.6 3.87 1 8.50 1 1.73 1 7.05 31 11.4 16 2.16 59 3.46 1 6.56 1 2.00 76 8.58 30 12.6 17 5.26 90 17.5 5 23.6 7 4.97 2 8.35 1 22.4 2 3.70 1 8.08 2 33.3 1 2.52 94 12.8 1 19.1 1 2.38 102
DF-Auto [115]25.5 5.03 8 8.87 2 2.16 83 7.72 61 13.1 57 2.38 91 5.69 4 9.20 9 1.73 1 8.68 37 12.5 13 5.10 22 17.4 3 23.4 3 5.16 54 9.47 7 24.0 17 3.74 19 8.98 36 38.4 37 2.45 1 14.0 20 21.8 26 2.16 1
DeepFlow2 [108]26.0 5.07 11 9.85 17 2.08 57 7.53 56 13.1 57 2.16 59 5.69 4 10.0 28 1.73 1 8.83 64 13.4 45 5.10 22 17.6 8 23.7 11 5.20 61 9.24 4 23.0 5 3.74 19 9.00 39 37.9 27 2.45 1 13.9 13 21.5 15 2.16 1
COFM [59]27.7 5.07 11 10.7 55 2.00 2 6.86 20 11.4 16 2.08 1 5.69 4 9.75 20 1.73 1 8.35 7 12.5 13 5.07 2 18.1 63 24.7 66 5.03 6 11.0 98 27.5 102 3.70 1 8.06 1 39.1 50 2.45 1 14.4 62 22.7 62 2.16 1
WLIF-Flow [93]28.3 5.10 29 10.2 31 2.00 2 7.00 30 11.9 33 2.08 1 5.69 4 9.68 16 1.73 1 8.29 5 12.2 6 5.23 75 17.8 31 24.0 26 5.10 29 10.6 78 26.6 83 3.83 93 8.83 22 37.5 17 2.45 1 14.1 32 21.9 33 2.16 1
DeepFlow [86]29.0 5.07 11 9.63 11 2.08 57 7.44 52 13.0 51 2.16 59 5.74 41 10.0 28 1.73 1 8.96 77 13.0 29 5.20 50 17.6 8 23.8 15 5.20 61 9.15 3 23.2 7 3.87 101 8.81 19 35.6 2 2.45 1 13.7 5 21.1 7 2.16 1
LME [70]29.1 5.07 11 10.1 25 2.00 2 7.05 31 12.0 36 2.16 59 5.69 4 10.7 76 1.73 1 8.35 7 12.8 26 5.10 22 18.0 50 24.4 49 5.29 121 10.2 51 25.3 49 3.74 19 8.70 13 36.4 3 2.45 1 14.0 20 21.7 22 2.16 1
Layers++ [37]29.7 5.10 29 10.1 25 2.08 57 6.45 3 9.88 1 2.08 1 5.69 4 10.0 28 1.73 1 8.37 14 12.7 22 5.10 22 18.1 63 24.9 83 5.10 29 10.7 86 28.3 113 3.74 19 8.76 14 38.0 31 2.45 1 14.1 32 21.9 33 2.16 1
SuperFlow [81]29.8 5.00 3 9.35 4 2.16 83 7.85 66 13.1 57 2.38 91 6.00 49 9.47 14 2.00 76 8.70 41 12.7 22 5.20 50 17.6 8 23.7 11 5.20 61 9.27 5 23.9 15 3.70 1 8.81 19 37.6 20 2.45 1 13.8 8 21.2 10 2.16 1
ProbFlowFields [128]29.9 5.03 8 10.7 55 2.00 2 6.68 9 11.3 14 2.08 1 5.69 4 9.47 14 1.73 1 8.52 27 13.3 44 5.20 50 18.2 82 24.9 83 5.23 104 10.5 69 26.2 75 3.74 19 8.60 6 37.7 22 2.45 1 13.8 8 21.6 18 2.16 1
nLayers [57]30.3 5.16 35 10.5 47 2.00 2 6.66 8 10.9 9 2.08 1 5.69 4 9.00 2 1.73 1 8.49 26 13.0 29 5.10 22 18.3 91 25.2 101 5.20 61 10.4 65 25.6 55 3.74 19 8.66 9 38.5 41 2.45 1 14.2 46 22.4 52 2.16 1
CombBMOF [113]30.3 5.35 71 10.5 47 2.00 2 6.83 18 11.4 16 2.08 1 5.80 44 10.0 28 1.73 1 8.83 64 14.4 72 5.10 22 17.9 41 24.3 43 5.07 10 9.88 26 24.1 20 3.70 1 10.7 112 38.3 34 2.45 1 14.0 20 21.9 33 2.16 1
Aniso. Huber-L1 [22]31.2 5.26 57 10.0 21 2.08 57 8.81 89 14.5 92 2.16 59 6.00 49 9.75 20 1.73 1 8.72 43 13.0 29 5.16 43 17.6 8 23.8 15 5.10 29 9.87 25 23.2 7 3.70 1 9.26 64 37.8 24 2.45 1 13.8 8 21.0 5 2.16 1
IROF-TV [53]31.4 5.20 37 10.7 55 2.08 57 7.05 31 11.9 33 2.08 1 6.00 49 10.3 54 1.73 1 8.37 14 12.6 17 5.16 43 17.8 31 24.1 34 5.23 104 10.1 44 25.0 40 3.70 1 9.04 46 39.1 50 2.45 1 13.7 5 21.0 5 2.16 1
Brox et al. [5]31.5 5.20 37 9.83 14 2.00 2 7.62 60 12.6 42 2.16 59 6.00 49 10.2 50 2.00 76 8.76 47 12.6 17 5.07 2 17.5 5 23.6 7 5.16 54 10.1 44 25.3 49 3.74 19 9.00 39 40.1 62 2.45 1 13.8 8 21.3 13 2.16 1
FMOF [94]31.9 5.42 88 11.0 71 2.00 2 6.76 14 11.0 10 2.08 1 6.00 49 10.3 54 1.73 1 8.83 64 14.1 65 5.10 22 17.8 31 24.1 34 5.07 10 10.0 42 25.6 55 3.74 19 8.58 5 37.7 22 2.45 1 14.3 53 22.4 52 2.16 1
Sparse-NonSparse [56]32.0 5.20 37 10.7 55 2.00 2 6.78 15 11.6 22 2.08 1 5.69 4 10.0 28 1.73 1 8.43 22 12.5 13 5.07 2 18.1 63 24.7 66 5.10 29 10.5 69 26.7 86 3.74 19 8.76 14 42.1 89 2.45 1 14.3 53 23.0 76 2.16 1
TV-L1-MCT [64]32.2 5.48 101 11.4 92 2.00 2 7.35 41 13.1 57 2.08 1 5.48 2 10.3 54 1.73 1 8.35 7 12.4 9 5.07 2 18.3 91 25.3 105 5.10 29 9.49 9 23.5 10 3.79 74 8.81 19 39.2 53 2.45 1 13.7 5 21.1 7 2.16 1
FlowFields [110]32.7 5.10 29 11.1 81 2.00 2 6.88 21 11.5 20 2.08 1 5.69 4 10.0 28 1.73 1 8.76 47 14.9 83 5.20 50 18.0 50 24.4 49 5.16 54 10.3 58 25.8 61 3.74 19 8.76 14 37.8 24 2.45 1 14.1 32 22.5 55 2.16 1
ComponentFusion [96]33.6 5.07 11 11.2 86 2.00 2 6.81 17 11.6 22 2.08 1 5.72 40 9.81 24 1.73 1 8.37 14 13.2 41 5.07 2 18.1 63 24.7 66 5.10 29 9.90 27 24.9 38 3.74 19 9.20 61 44.1 107 2.45 1 14.2 46 23.3 88 2.16 1
MDP-Flow [26]33.9 5.03 8 9.95 19 2.00 2 6.68 9 11.3 14 2.08 1 5.69 4 9.04 7 1.73 1 8.89 72 13.7 51 5.20 50 17.8 31 24.2 41 5.20 61 11.3 110 27.9 108 3.74 19 9.27 68 39.3 55 2.45 1 14.1 32 22.3 48 2.16 1
CLG-TV [48]35.8 5.20 37 9.49 8 2.08 57 8.43 80 14.3 86 2.16 59 6.00 49 10.1 46 2.00 76 8.76 47 13.1 37 5.20 50 17.6 8 23.8 15 5.10 29 9.59 16 23.1 6 3.74 19 9.20 61 38.4 37 2.45 1 14.0 20 21.5 15 2.16 1
2DHMM-SAS [92]35.8 5.42 88 11.2 86 2.00 2 7.90 68 13.7 70 2.08 1 5.60 3 9.85 25 1.73 1 8.35 7 12.2 6 5.10 22 18.0 50 24.6 64 5.10 29 9.93 30 25.7 58 3.74 19 8.96 31 39.8 60 2.45 1 14.4 62 23.0 76 2.16 1
PGM-C [120]36.5 5.07 11 10.9 69 2.00 2 6.93 24 11.6 22 2.08 1 6.00 49 10.3 54 1.73 1 8.76 47 15.2 88 5.16 43 18.0 50 24.7 66 5.20 61 9.97 37 24.8 36 3.74 19 9.00 39 40.1 62 2.45 1 14.1 32 22.7 62 2.16 1
ALD-Flow [66]37.5 5.20 37 10.7 55 2.08 57 7.35 41 12.9 47 2.16 59 6.00 49 10.1 46 1.73 1 8.39 19 13.0 29 5.16 43 17.9 41 24.3 43 5.20 61 9.56 13 23.5 10 3.79 74 8.79 18 36.8 7 2.45 1 14.5 73 23.0 76 2.16 1
CPM-Flow [116]37.8 5.07 11 10.9 69 2.00 2 6.95 26 11.6 22 2.08 1 5.80 44 10.0 28 1.73 1 9.00 80 15.9 102 5.20 50 18.1 63 24.7 66 5.20 61 9.81 20 24.3 25 3.79 74 9.26 64 38.3 34 2.45 1 14.0 20 22.2 42 2.16 1
HAST [109]38.4 5.07 11 10.5 47 2.00 2 6.68 9 10.7 8 2.08 1 6.00 49 10.3 54 1.73 1 8.29 5 12.4 9 5.00 1 18.4 101 25.3 105 5.03 6 11.0 98 30.7 122 3.70 1 8.60 6 41.8 84 2.45 1 14.9 99 23.9 101 2.16 1
Second-order prior [8]38.5 5.20 37 9.83 14 2.08 57 8.43 80 14.5 92 2.08 1 6.35 91 11.0 92 2.00 76 8.83 64 13.8 58 5.07 2 17.7 21 23.8 15 5.07 10 9.70 18 24.1 20 3.74 19 9.33 71 38.4 37 2.45 1 14.0 20 21.8 26 2.16 1
CBF [12]38.9 5.00 3 9.40 6 2.08 57 7.77 64 13.0 51 2.16 59 6.00 49 9.68 16 1.73 1 8.68 37 12.5 13 5.35 105 17.6 8 23.4 3 5.20 61 9.85 24 24.3 25 3.74 19 9.11 55 39.3 55 2.52 94 14.0 20 21.1 7 2.38 102
S2F-IF [123]39.1 5.10 29 11.6 101 2.00 2 6.78 15 11.4 16 2.08 1 5.69 4 10.3 54 1.73 1 8.74 44 15.2 88 5.07 2 18.3 91 25.1 97 5.20 61 10.5 69 26.1 70 3.74 19 9.02 44 38.5 41 2.45 1 14.1 32 22.6 56 2.16 1
Local-TV-L1 [65]39.5 5.20 37 9.38 5 2.16 83 8.96 95 14.5 92 2.38 91 5.69 4 9.35 10 1.73 1 8.70 41 13.0 29 5.45 110 17.6 8 23.8 15 5.16 54 9.54 12 24.0 17 4.08 122 8.76 14 37.2 14 2.45 1 13.6 4 20.9 4 2.31 84
Ramp [62]39.6 5.29 64 10.8 64 2.00 2 6.83 18 11.6 22 2.08 1 5.69 4 10.1 46 1.73 1 8.35 7 12.2 6 5.07 2 18.1 63 24.7 66 5.10 29 10.9 93 27.8 107 3.79 74 8.83 22 43.0 100 2.45 1 14.5 73 23.2 84 2.16 1
SIOF [67]40.0 5.42 88 10.4 41 2.08 57 8.83 90 15.0 105 2.38 91 5.69 4 10.4 73 1.73 1 8.68 37 13.1 37 5.20 50 17.3 2 23.2 2 5.07 10 9.83 21 23.6 12 3.74 19 9.00 39 36.9 8 2.45 1 14.3 53 22.1 36 2.31 84
p-harmonic [29]40.5 5.07 11 9.98 20 2.00 2 8.68 86 14.4 88 2.16 59 6.00 49 10.7 76 1.91 72 9.20 88 13.7 51 5.20 50 17.8 31 24.0 26 5.10 29 9.90 27 23.7 13 3.74 19 9.61 89 38.5 41 2.45 1 14.0 20 21.7 22 2.16 1
LDOF [28]41.6 5.35 71 9.83 14 2.16 83 7.94 69 12.1 37 2.52 107 6.00 49 10.3 54 2.00 76 8.91 75 13.6 50 5.23 75 17.6 8 23.6 7 5.20 61 9.49 9 24.5 29 3.74 19 8.96 31 37.9 27 2.45 1 14.0 20 21.8 26 2.16 1
DPOF [18]42.3 5.35 71 11.7 103 2.08 57 6.56 7 10.4 6 2.08 1 6.00 49 9.71 19 1.91 72 8.76 47 14.4 72 5.20 50 17.7 21 24.1 34 5.07 10 10.3 58 26.7 86 3.70 1 9.33 71 39.1 50 2.45 1 14.4 62 22.8 67 2.16 1
ComplOF-FED-GPU [35]42.7 5.20 37 11.1 81 2.00 2 7.19 38 12.6 42 2.08 1 6.35 91 10.0 28 2.00 76 8.68 37 14.0 64 5.10 22 17.9 41 24.5 56 5.10 29 9.97 37 25.1 43 3.74 19 9.40 76 38.8 46 2.45 1 14.5 73 23.2 84 2.16 1
OFLAF [77]43.2 5.07 11 10.6 52 2.00 2 6.48 6 10.5 7 2.08 1 5.69 4 10.0 28 1.73 1 8.37 14 12.6 17 5.07 2 18.4 101 25.4 111 5.20 61 10.9 93 27.4 100 3.74 19 9.59 88 44.9 111 2.45 1 15.1 104 24.1 103 2.16 1
AGIF+OF [85]43.4 5.42 88 11.1 81 2.00 2 6.98 27 11.8 30 2.08 1 5.69 4 10.0 28 1.73 1 8.43 22 12.8 26 5.07 2 18.5 108 25.2 101 5.20 61 10.8 91 27.6 103 3.74 19 8.98 36 37.9 27 2.45 1 14.7 89 23.4 92 2.16 1
LSM [39]43.7 5.35 71 11.5 97 2.00 2 6.98 27 11.9 33 2.08 1 5.80 44 10.7 76 1.73 1 8.58 30 13.4 45 5.07 2 18.1 63 24.9 83 5.10 29 10.6 78 27.1 94 3.74 19 8.83 22 42.2 91 2.45 1 14.4 62 23.0 76 2.16 1
FC-2Layers-FF [74]43.8 5.26 57 11.0 71 2.00 2 6.40 1 9.88 1 2.08 1 5.69 4 10.3 54 1.73 1 8.39 19 12.8 26 5.10 22 18.2 82 25.0 91 5.20 61 11.0 98 28.1 110 3.79 74 8.91 29 42.8 95 2.45 1 14.5 73 23.0 76 2.16 1
Classic+NL [31]44.0 5.35 71 11.0 71 2.08 57 6.98 27 11.7 28 2.08 1 5.69 4 10.2 50 1.73 1 8.43 22 12.4 9 5.20 50 18.1 63 24.8 76 5.10 29 10.6 78 26.8 89 3.79 74 8.83 22 42.9 96 2.45 1 14.4 62 22.9 74 2.16 1
RFlow [90]44.2 5.07 11 10.2 31 2.08 57 8.58 84 14.7 98 2.08 1 6.00 49 10.3 54 1.73 1 8.91 75 14.4 72 5.20 50 17.7 21 23.9 23 5.10 29 9.95 32 25.4 52 3.70 1 9.13 57 40.4 68 2.45 1 14.3 53 22.6 56 2.31 84
OAR-Flow [125]44.2 5.20 37 10.7 55 2.08 57 7.44 52 13.0 51 2.16 59 5.74 41 10.0 28 1.73 1 8.35 7 13.0 29 5.10 22 18.1 63 24.9 83 5.23 104 10.2 51 24.7 32 3.74 19 9.54 86 39.4 58 2.45 1 14.4 62 22.7 62 2.16 1
RNLOD-Flow [121]45.1 5.20 37 11.0 71 2.00 2 7.53 56 13.4 63 2.08 1 6.00 49 11.0 92 1.73 1 8.52 27 13.0 29 5.07 2 18.2 82 25.0 91 5.10 29 10.6 78 26.9 91 3.74 19 8.96 31 38.4 37 2.45 1 14.9 99 23.5 94 2.16 1
TF+OM [100]45.5 5.00 3 10.2 31 2.08 57 6.93 24 11.7 28 2.16 59 5.69 4 10.5 74 1.73 1 8.81 59 14.6 78 5.20 50 18.0 50 24.4 49 5.20 61 9.95 32 26.1 70 3.79 74 9.09 52 41.0 74 2.45 1 14.1 32 21.8 26 2.38 102
TC/T-Flow [76]45.9 5.45 96 11.5 97 2.00 2 7.42 49 13.0 51 2.08 1 5.69 4 9.76 22 1.73 1 8.60 33 13.7 51 5.16 43 18.3 91 24.9 83 5.20 61 10.1 44 24.9 38 3.74 19 9.75 92 42.6 92 2.45 1 14.5 73 22.6 56 2.16 1
F-TV-L1 [15]46.0 5.35 71 10.3 38 2.16 83 8.83 90 14.6 97 2.16 59 6.00 49 10.3 54 2.00 76 8.76 47 13.2 41 5.26 90 17.6 8 23.8 15 5.03 6 9.57 15 23.2 7 3.79 74 9.18 60 37.6 20 2.45 1 13.8 8 21.2 10 2.31 84
EpicFlow [102]46.8 5.07 11 11.0 71 2.00 2 7.39 45 12.9 47 2.08 1 5.80 44 10.3 54 1.73 1 8.85 70 15.5 96 5.20 50 18.1 63 24.8 76 5.20 61 10.2 51 25.1 43 3.74 19 9.33 71 40.4 68 2.45 1 14.5 73 24.1 103 2.16 1
S2D-Matching [84]47.5 5.35 71 11.2 86 2.00 2 7.75 63 13.5 66 2.08 1 5.69 4 10.0 28 1.73 1 8.37 14 12.6 17 5.20 50 18.3 91 25.2 101 5.07 10 11.0 98 27.7 106 3.79 74 9.09 52 40.3 66 2.45 1 14.4 62 23.0 76 2.16 1
TC-Flow [46]47.6 5.07 11 10.8 64 2.00 2 7.39 45 13.2 62 2.16 59 6.00 49 10.3 54 1.73 1 8.66 35 13.7 51 5.23 75 18.2 82 25.0 91 5.20 61 10.2 51 24.5 29 3.79 74 9.04 46 38.1 32 2.45 1 14.5 73 23.5 94 2.16 1
Fusion [6]49.9 5.20 37 10.4 41 2.00 2 7.14 37 11.8 30 2.08 1 5.74 41 9.68 16 1.73 1 9.33 91 14.2 66 5.20 50 18.3 91 24.7 66 5.07 10 11.6 114 28.1 110 3.70 1 9.63 90 41.4 79 2.45 1 15.3 115 24.2 106 2.16 1
Modified CLG [34]50.2 5.07 11 9.49 8 2.16 83 9.42 108 14.2 84 2.65 111 6.00 49 11.5 101 2.00 76 9.15 85 14.3 68 5.10 22 17.7 21 23.9 23 5.10 29 10.1 44 24.7 32 3.74 19 9.31 70 37.5 17 2.45 1 14.1 32 21.8 26 2.31 84
Classic++ [32]50.6 5.20 37 10.3 38 2.08 57 7.94 69 13.8 72 2.08 1 6.00 49 10.1 46 1.73 1 8.89 72 13.7 51 5.23 75 18.0 50 24.5 56 5.10 29 10.3 58 25.8 61 3.87 101 9.13 57 40.1 62 2.45 1 14.2 46 22.2 42 2.31 84
Sparse Occlusion [54]51.9 5.26 57 10.5 47 2.08 57 8.04 71 14.4 88 2.08 1 6.00 49 10.0 28 1.73 1 8.83 64 13.7 51 5.20 50 18.1 63 24.7 66 5.20 61 11.0 98 26.5 79 3.74 19 9.42 77 42.0 88 2.45 1 14.4 62 22.8 67 2.16 1
AggregFlow [97]52.0 5.45 96 13.8 119 2.08 57 7.44 52 13.1 57 2.16 59 5.69 4 9.95 27 1.73 1 9.15 85 16.1 103 5.10 22 18.0 50 24.5 56 5.20 61 9.90 27 24.6 31 3.83 93 8.98 36 40.7 71 2.45 1 14.4 62 23.0 76 2.16 1
FESL [72]52.9 5.42 88 11.0 71 2.00 2 7.05 31 11.8 30 2.08 1 5.69 4 10.7 76 1.73 1 8.81 59 13.5 48 5.20 50 18.4 101 25.1 97 5.20 61 11.0 98 27.0 93 3.74 19 9.06 50 42.9 96 2.45 1 14.8 94 23.7 98 2.16 1
BlockOverlap [61]53.1 5.20 37 9.29 3 2.16 83 8.74 88 14.1 79 2.65 111 6.00 49 9.35 10 2.00 76 8.52 27 11.9 1 5.60 117 17.8 31 24.0 26 5.32 123 9.83 21 25.0 40 4.04 118 8.83 22 37.1 11 2.52 94 13.5 3 20.6 3 2.38 102
PMF [73]53.1 5.20 37 11.4 92 2.00 2 7.35 41 12.4 40 2.08 1 6.00 49 12.0 108 1.73 1 8.76 47 14.4 72 5.07 2 18.4 101 25.0 91 5.10 29 10.2 51 25.8 61 3.87 101 9.04 46 41.3 78 2.45 1 15.2 113 24.5 110 2.16 1
Classic+CPF [83]53.6 5.35 71 11.3 90 2.00 2 7.07 36 12.1 37 2.08 1 5.69 4 10.5 74 1.73 1 8.43 22 12.7 22 5.07 2 18.7 117 25.7 120 5.20 61 11.2 107 28.7 116 3.74 19 9.42 77 42.9 96 2.45 1 15.1 104 24.2 106 2.16 1
TCOF [69]54.1 5.35 71 10.7 55 2.00 2 9.27 102 15.4 112 2.16 59 5.69 4 10.2 50 1.73 1 8.74 44 13.1 37 5.23 75 17.7 21 23.8 15 5.07 10 10.7 86 26.6 83 3.70 1 10.0 100 44.7 110 2.45 1 14.6 84 22.9 74 2.38 102
FlowNetS+ft+v [112]54.5 5.26 57 10.1 25 2.16 83 9.11 99 14.5 92 2.45 102 6.00 49 10.3 54 2.00 76 8.96 77 13.5 48 5.26 90 17.8 31 24.1 34 5.23 104 9.76 19 23.9 15 3.74 19 9.38 75 41.6 82 2.45 1 14.1 32 22.2 42 2.16 1
SVFilterOh [111]55.2 5.20 37 10.6 52 2.00 2 6.73 13 11.0 10 2.08 1 6.00 49 10.0 28 1.73 1 8.76 47 13.8 58 5.26 90 18.4 101 25.3 105 5.26 114 10.6 78 28.0 109 3.74 19 8.45 3 39.2 53 2.52 94 14.7 89 23.3 88 2.31 84
OFH [38]55.2 5.35 71 11.0 71 2.08 57 8.06 74 13.7 70 2.08 1 6.00 49 11.6 102 1.73 1 8.58 30 13.9 62 5.07 2 18.2 82 24.9 83 5.16 54 10.3 58 25.1 43 3.74 19 9.88 96 42.7 94 2.45 1 14.8 94 24.7 111 2.16 1
CRTflow [80]55.8 5.29 64 10.5 47 2.16 83 8.43 80 14.5 92 2.16 59 6.35 91 11.1 98 2.00 76 8.64 34 13.0 29 5.29 99 18.0 50 24.5 56 5.20 61 9.68 17 23.8 14 3.74 19 9.00 39 40.9 73 2.45 1 14.1 32 22.2 42 2.31 84
EPPM w/o HM [88]56.4 5.23 55 12.6 111 2.00 2 7.39 45 13.0 51 2.08 1 6.35 91 14.0 122 1.91 72 8.83 64 15.3 92 5.10 22 18.0 50 24.5 56 5.10 29 10.5 69 27.6 103 3.74 19 9.11 55 41.9 85 2.45 1 14.5 73 23.2 84 2.16 1
SRR-TVOF-NL [91]56.5 5.45 96 12.1 109 2.08 57 7.77 64 13.5 66 2.16 59 6.00 49 10.3 54 1.73 1 9.26 89 14.7 80 5.07 2 18.1 63 24.6 64 5.10 29 10.4 65 26.6 83 3.70 1 9.42 77 38.5 41 2.45 1 15.1 104 23.9 101 2.16 1
Efficient-NL [60]56.6 5.35 71 10.7 55 2.00 2 7.42 49 13.0 51 2.08 1 6.35 91 10.7 76 2.00 76 8.81 59 13.4 45 5.10 22 18.1 63 24.7 66 5.10 29 11.2 107 27.6 103 3.70 1 9.47 81 43.6 105 2.45 1 15.1 104 23.8 100 2.16 1
2D-CLG [1]57.0 5.16 35 10.0 21 2.16 83 9.90 113 14.2 84 2.83 118 6.35 91 10.7 76 2.00 76 10.0 105 15.2 88 5.10 22 17.7 21 24.1 34 5.20 61 10.1 44 24.1 20 3.74 19 9.81 93 43.6 105 2.45 1 14.1 32 21.8 26 2.16 1
Steered-L1 [118]57.2 5.07 11 9.81 13 2.00 2 7.35 41 12.8 46 2.16 59 6.35 91 10.3 54 2.00 76 9.31 90 14.3 68 5.35 105 18.2 82 24.7 66 5.07 10 10.2 51 25.7 58 3.79 74 9.33 71 40.4 68 2.45 1 14.6 84 22.8 67 2.31 84
MLDP_OF [89]57.3 5.32 68 11.1 81 2.00 2 7.55 59 13.6 68 2.08 1 5.69 4 10.0 28 1.73 1 8.76 47 13.1 37 5.26 90 18.0 50 24.5 56 5.20 61 11.0 98 26.9 91 4.08 122 9.26 64 38.2 33 2.52 94 14.4 62 22.6 56 2.38 102
Occlusion-TV-L1 [63]58.8 5.20 37 10.2 31 2.08 57 8.89 92 15.3 110 2.16 59 6.00 49 10.3 54 2.00 76 9.15 85 15.4 93 5.26 90 17.6 8 23.7 11 5.10 29 9.98 39 25.5 54 3.87 101 10.3 105 39.3 55 2.52 94 14.1 32 22.3 48 2.16 1
IAOF [50]59.3 5.60 106 11.0 71 2.16 83 12.0 127 16.9 128 2.52 107 5.69 4 11.0 92 2.00 76 9.76 102 14.3 68 5.20 50 17.7 21 24.0 26 5.07 10 10.0 42 25.2 47 3.74 19 9.47 81 41.4 79 2.45 1 14.2 46 22.1 36 2.16 1
Complementary OF [21]60.8 5.20 37 12.0 107 2.00 2 7.19 38 12.9 47 2.08 1 6.68 108 10.8 88 2.00 76 8.76 47 14.6 78 5.16 43 18.2 82 25.2 101 5.10 29 10.3 58 25.9 65 3.74 19 9.97 99 42.6 92 2.45 1 15.6 118 28.0 123 2.16 1
Ad-TV-NDC [36]61.5 5.66 108 9.88 18 2.52 122 10.1 116 15.1 106 2.71 114 6.00 49 10.7 76 1.73 1 9.49 98 14.2 66 5.35 105 17.7 21 24.0 26 5.20 61 9.56 13 24.0 17 3.87 101 9.56 87 38.6 45 2.45 1 13.9 13 21.2 10 2.38 102
Adaptive [20]61.8 5.32 68 10.3 38 2.16 83 9.29 105 15.4 112 2.16 59 6.00 49 10.7 76 1.73 1 8.81 59 13.8 58 5.20 50 17.9 41 24.3 43 5.07 10 10.4 65 26.0 66 3.79 74 9.83 94 44.6 108 2.45 1 14.5 73 22.8 67 2.31 84
Black & Anandan [4]63.2 5.45 96 10.1 25 2.16 83 10.2 119 15.3 110 2.45 102 6.68 108 11.3 99 2.00 76 10.2 106 15.6 97 5.20 50 17.8 31 24.0 26 5.16 54 9.83 21 24.7 32 3.74 19 10.2 104 41.9 85 2.45 1 14.2 46 21.8 26 2.16 1
HBM-GC [105]63.2 5.35 71 10.6 52 2.16 83 7.42 49 13.4 63 2.16 59 5.69 4 9.00 2 1.73 1 8.74 44 13.2 41 5.26 90 18.6 113 25.5 116 5.26 114 11.8 120 31.5 125 3.83 93 8.83 22 41.1 76 2.45 1 14.3 53 22.2 42 2.31 84
CNN-flow-warp+ref [117]63.3 5.00 3 9.59 10 2.16 83 8.35 78 13.6 68 2.16 59 6.35 91 11.8 107 2.00 76 10.6 111 15.4 93 5.48 114 17.8 31 24.3 43 5.23 104 9.95 32 24.3 25 3.83 93 9.83 94 44.6 108 2.45 1 14.2 46 22.3 48 2.16 1
CostFilter [40]63.3 5.32 68 13.2 116 2.00 2 7.33 40 12.3 39 2.08 1 6.06 87 13.5 121 1.73 1 8.96 77 16.1 103 5.07 2 18.6 113 25.6 119 5.16 54 9.98 39 24.8 36 4.04 118 9.20 61 43.5 104 2.45 1 15.1 104 24.9 113 2.16 1
BriefMatch [124]64.0 5.29 64 11.4 92 2.08 57 7.44 52 12.7 44 2.16 59 6.38 106 9.93 26 2.00 76 9.83 104 14.9 83 5.83 124 18.0 50 24.4 49 5.20 61 10.5 69 27.3 98 4.32 127 9.04 46 37.9 27 2.45 1 14.3 53 22.8 67 2.16 1
AdaConv-v1 [126]65.7 6.24 119 14.4 120 2.38 116 9.02 96 12.7 44 3.11 123 7.00 116 11.0 92 2.38 122 13.1 124 18.8 117 5.83 124 16.8 1 22.5 1 4.83 1 8.79 2 22.0 1 3.70 1 8.91 29 36.6 6 2.58 116 13.3 2 20.2 2 2.38 102
HBpMotionGpu [43]66.0 5.48 101 10.8 64 2.38 116 10.1 116 15.4 112 2.71 114 5.69 4 10.0 28 1.73 1 9.40 93 16.2 107 5.23 75 17.9 41 24.3 43 5.20 61 10.5 69 26.4 78 3.83 93 8.96 31 37.8 24 2.45 1 14.3 53 22.6 56 2.38 102
TriFlow [95]67.1 5.26 57 12.0 107 2.16 83 8.39 79 14.4 88 2.38 91 6.00 49 11.0 92 1.73 1 9.02 81 15.4 93 5.10 22 18.5 108 25.4 111 5.20 61 10.6 78 27.3 98 3.74 19 9.26 64 39.7 59 2.45 1 14.6 84 23.1 83 2.16 1
Nguyen [33]67.5 5.42 88 10.0 21 2.38 116 10.9 122 15.1 106 2.65 111 6.00 49 12.0 108 2.00 76 10.4 110 16.1 103 5.20 50 17.8 31 24.1 34 5.07 10 9.98 39 25.3 49 3.70 1 10.9 116 46.9 117 2.52 94 14.1 32 22.1 36 2.16 1
Aniso-Texture [82]68.5 5.07 11 10.2 31 2.00 2 8.89 92 15.2 109 2.16 59 6.35 91 10.3 54 1.73 1 9.04 82 16.1 103 5.29 99 18.3 91 24.9 83 5.23 104 11.8 120 30.0 120 3.83 93 9.06 50 40.2 65 2.45 1 14.7 89 23.5 94 2.16 1
TV-L1-improved [17]70.8 5.10 29 10.2 31 2.08 57 9.20 101 15.4 112 2.16 59 6.35 91 10.3 54 2.00 76 8.85 70 13.8 58 5.23 75 18.0 50 24.4 49 5.10 29 10.6 78 26.5 79 3.79 74 9.93 98 46.9 117 2.52 94 14.3 53 22.7 62 2.38 102
FlowNet2 [122]70.8 6.45 123 19.1 127 2.16 83 7.85 66 13.4 63 2.38 91 6.06 87 11.7 103 1.73 1 9.40 93 18.2 113 5.23 75 18.5 108 25.3 105 5.20 61 10.3 58 25.2 47 3.74 19 9.27 68 41.9 85 2.45 1 14.3 53 22.8 67 2.16 1
Bartels [41]71.9 5.35 71 11.4 92 2.16 83 7.72 61 14.0 78 2.38 91 6.00 49 10.3 54 2.00 76 9.11 84 15.0 86 5.69 119 17.6 8 23.6 7 5.45 127 10.7 86 27.2 95 4.55 129 8.96 31 36.4 3 2.65 125 14.1 32 22.1 36 2.38 102
GraphCuts [14]72.1 5.66 108 11.9 106 2.16 83 7.53 56 12.5 41 2.38 91 7.68 122 10.2 50 2.00 76 9.47 97 14.9 83 5.23 75 18.1 63 24.5 56 5.00 3 10.1 44 25.7 58 3.70 1 9.02 44 42.1 89 2.52 94 15.1 104 24.1 103 2.31 84
SimpleFlow [49]72.7 5.35 71 11.0 71 2.00 2 8.04 71 13.9 75 2.08 1 6.56 107 11.3 99 2.00 76 8.41 21 12.7 22 5.20 50 18.4 101 25.4 111 5.20 61 11.4 112 28.9 117 3.74 19 10.1 102 53.7 126 2.52 94 15.3 115 26.5 119 2.16 1
Filter Flow [19]75.2 5.42 88 10.2 31 2.16 83 9.40 107 14.7 98 2.71 114 6.00 49 10.7 76 2.00 76 9.49 98 13.9 62 5.35 105 18.1 63 24.3 43 5.26 114 10.2 51 25.6 55 3.83 93 9.52 85 41.4 79 2.45 1 14.6 84 22.3 48 2.38 102
ROF-ND [107]75.7 5.74 111 10.4 41 2.00 2 8.04 71 14.1 79 2.16 59 6.06 87 10.7 76 1.73 1 10.6 111 19.9 122 5.26 90 18.1 63 24.8 76 5.20 61 11.7 117 28.6 114 3.74 19 11.1 118 41.0 74 2.52 94 15.3 115 25.3 115 2.16 1
Shiralkar [42]77.1 5.48 101 12.7 112 2.08 57 9.06 98 14.7 98 2.08 1 6.00 49 12.8 116 2.00 76 10.7 113 19.7 121 5.20 50 18.1 63 24.8 76 5.00 3 10.8 91 26.1 70 3.87 101 10.8 115 47.5 121 2.45 1 14.9 99 25.8 117 2.16 1
TriangleFlow [30]77.6 5.60 106 11.6 101 2.16 83 8.50 83 14.4 88 2.08 1 6.35 91 10.7 76 2.00 76 9.42 95 15.8 99 5.23 75 18.0 50 24.5 56 5.00 3 11.1 106 27.2 95 3.74 19 10.4 106 47.2 120 2.52 94 15.6 118 26.7 120 2.16 1
Rannacher [23]78.2 5.26 57 10.8 64 2.16 83 9.27 102 15.5 119 2.16 59 6.35 91 10.9 90 2.00 76 8.76 47 14.4 72 5.23 75 17.9 41 24.4 49 5.20 61 10.5 69 26.7 86 3.79 74 9.90 97 45.9 114 2.52 94 14.4 62 23.5 94 2.38 102
Correlation Flow [75]78.2 5.42 88 11.7 103 2.00 2 8.58 84 15.4 112 2.08 1 5.69 4 9.80 23 1.73 1 8.89 72 14.7 80 5.32 103 18.1 63 24.8 76 5.32 123 12.3 127 30.3 121 3.83 93 10.5 109 48.8 123 2.52 94 14.8 94 23.7 98 2.31 84
Horn & Schunck [3]80.8 5.48 101 10.4 41 2.16 83 10.5 121 15.4 112 2.52 107 6.68 108 12.0 108 2.00 76 11.5 119 17.6 112 5.23 75 17.9 41 24.0 26 5.20 61 9.93 30 24.1 20 3.79 74 11.1 118 42.9 96 2.52 94 14.5 73 22.2 42 2.38 102
IAOF2 [51]81.5 5.74 111 11.5 97 2.16 83 9.49 109 15.9 126 2.38 91 5.69 4 11.0 92 2.00 76 9.61 101 15.8 99 5.26 90 18.7 117 25.3 105 5.20 61 10.9 93 27.4 100 3.74 19 9.47 81 41.1 76 2.45 1 14.5 73 22.8 67 2.31 84
TI-DOFE [24]83.6 5.80 113 11.0 71 2.52 122 11.5 125 15.8 124 3.11 123 6.35 91 12.3 112 2.00 76 11.4 118 17.4 110 5.29 99 17.9 41 24.2 41 5.07 10 9.95 32 24.4 28 3.79 74 10.5 109 39.9 61 2.52 94 14.8 94 22.1 36 2.38 102
LocallyOriented [52]84.8 5.45 96 11.2 86 2.16 83 9.49 109 15.7 122 2.16 59 6.06 87 11.7 103 1.91 72 9.42 95 17.0 108 5.23 75 18.2 82 24.8 76 5.07 10 11.0 98 26.5 79 4.04 118 10.4 106 43.0 100 2.45 1 14.8 94 23.4 92 2.31 84
SegOF [10]85.4 5.10 29 11.4 92 2.16 83 8.29 76 13.9 75 2.38 91 7.00 116 12.1 111 2.00 76 9.81 103 21.0 123 5.20 50 18.2 82 25.1 97 5.20 61 10.9 93 26.1 70 3.79 74 10.4 106 48.4 122 2.58 116 14.7 89 25.1 114 2.16 1
SPSA-learn [13]86.0 5.29 64 10.4 41 2.16 83 9.04 97 14.1 79 2.45 102 6.68 108 11.7 103 2.00 76 10.3 109 15.8 99 5.10 22 18.4 101 25.3 105 5.20 61 10.5 69 26.8 89 3.74 19 12.3 126 58.4 128 2.71 128 17.6 126 35.0 128 2.16 1
2bit-BM-tele [98]88.3 5.35 71 10.1 25 2.16 83 8.91 94 15.4 112 2.45 102 6.00 49 10.0 28 2.00 76 9.04 82 14.3 68 5.60 117 18.3 91 24.9 83 5.35 126 11.7 117 31.3 124 4.24 126 12.0 125 58.7 129 2.83 129 13.9 13 21.7 22 2.45 127
StereoOF-V1MT [119]88.6 5.69 110 13.0 115 2.08 57 8.68 86 14.1 79 2.08 1 6.73 115 12.4 114 2.00 76 11.6 120 19.1 120 5.45 110 18.5 108 25.4 111 5.20 61 11.3 110 26.1 70 3.92 114 11.2 120 44.9 111 2.58 116 14.2 46 22.6 56 2.16 1
StereoFlow [44]89.1 8.68 129 20.4 129 2.45 120 10.3 120 16.1 127 2.71 114 6.00 49 10.7 76 1.73 1 8.81 59 13.7 51 5.16 43 22.6 127 31.6 127 5.26 114 14.3 129 35.7 129 3.79 74 9.13 57 38.8 46 2.45 1 15.6 118 25.3 115 2.31 84
ACK-Prior [27]89.7 5.35 71 11.7 103 2.00 2 7.39 45 12.9 47 2.08 1 6.68 108 10.8 88 2.00 76 9.54 100 15.7 98 5.32 103 18.7 117 25.5 116 5.29 121 11.9 123 29.5 118 3.87 101 10.1 102 41.7 83 2.52 94 16.1 121 24.8 112 2.38 102
UnFlow [129]90.5 5.97 115 15.5 121 2.16 83 9.13 100 14.1 79 2.38 91 6.68 108 13.0 118 2.00 76 9.35 92 17.1 109 5.23 75 18.6 113 25.8 122 5.20 61 11.5 113 29.5 118 3.74 19 9.66 91 37.4 16 2.45 1 16.9 125 28.1 124 2.38 102
Dynamic MRF [7]91.0 5.26 57 11.5 97 2.00 2 8.12 75 14.3 86 2.16 59 6.68 108 12.8 116 2.00 76 10.9 116 18.3 115 5.51 116 18.3 91 25.0 91 5.20 61 11.6 114 28.6 114 3.87 101 10.7 112 45.7 113 2.52 94 14.9 99 23.3 88 2.31 84
NL-TV-NCC [25]93.5 6.03 116 12.8 114 2.00 2 8.29 76 14.7 98 2.16 59 6.35 91 11.7 103 2.00 76 10.7 113 18.6 116 5.45 110 18.1 63 24.1 34 5.45 127 12.0 124 28.2 112 3.79 74 13.0 128 43.4 103 2.58 116 15.1 104 23.2 84 2.38 102
SILK [79]95.0 5.80 113 12.7 112 2.38 116 11.1 123 15.6 121 2.83 118 7.35 121 13.0 118 2.00 76 10.8 115 17.5 111 5.48 114 18.3 91 24.8 76 5.20 61 10.5 69 26.0 66 4.20 125 10.0 100 37.1 11 2.52 94 14.6 84 22.7 62 2.31 84
Learning Flow [11]96.4 5.57 105 11.1 81 2.16 83 9.27 102 15.1 106 2.16 59 7.00 116 13.3 120 2.00 76 10.2 106 15.2 88 5.45 110 18.5 108 25.1 97 5.32 123 10.7 86 26.2 75 3.87 101 10.6 111 40.8 72 2.52 94 15.1 104 23.3 88 2.38 102
FOLKI [16]103.7 6.14 118 12.4 110 3.11 126 11.5 125 15.5 119 3.32 126 7.00 116 14.7 124 2.38 122 13.5 125 18.2 113 6.27 127 18.6 113 25.0 91 5.23 104 10.3 58 25.1 43 4.04 118 11.0 117 38.3 34 2.58 116 14.7 89 22.4 52 2.38 102
Adaptive flow [45]103.8 6.24 119 11.3 90 2.71 124 11.2 124 15.7 122 3.42 127 6.35 91 10.9 90 2.00 76 10.2 106 14.7 80 5.72 120 18.7 117 25.4 111 5.23 104 11.7 117 30.8 123 3.87 101 9.42 77 38.8 46 2.58 116 14.9 99 24.3 108 2.38 102
GroupFlow [9]105.2 6.56 124 19.6 128 2.16 83 9.38 106 14.7 98 2.52 107 7.68 122 16.8 127 2.00 76 11.1 117 23.5 127 5.29 99 20.7 126 29.3 126 5.23 104 12.4 128 32.8 127 3.87 101 11.3 122 49.6 125 2.45 1 16.8 124 30.4 127 2.16 1
Heeger++ [104]108.2 7.16 127 18.5 126 2.16 83 9.75 111 13.8 72 2.45 102 9.35 126 16.1 126 2.38 122 13.0 122 18.9 118 5.74 121 19.8 125 27.4 125 5.23 104 12.2 125 26.0 66 3.92 114 13.5 129 46.5 115 2.52 94 16.1 121 27.4 121 2.16 1
SLK [47]111.4 6.03 116 13.6 117 2.45 120 10.1 116 13.8 72 2.89 120 7.68 122 12.4 114 2.38 122 13.8 127 21.0 123 5.77 123 19.1 124 26.4 124 5.20 61 11.2 107 26.2 75 3.87 101 11.8 123 46.9 117 2.58 116 15.2 113 26.1 118 2.38 102
FFV1MT [106]113.0 6.40 122 16.8 123 2.16 83 9.87 112 13.9 75 2.89 120 9.35 126 18.7 128 2.52 127 13.0 122 18.9 118 5.74 121 18.8 121 25.7 120 5.26 114 10.9 93 26.0 66 3.92 114 12.8 127 46.5 115 2.52 94 16.2 123 27.4 121 2.45 127
HCIC-L [99]114.5 7.62 128 17.7 125 3.16 127 9.98 114 14.8 104 3.16 125 7.14 120 14.0 122 2.00 76 12.4 121 21.5 126 5.35 105 18.9 123 25.5 116 5.26 114 11.8 120 31.9 126 3.87 101 9.47 81 43.2 102 2.58 116 18.7 128 30.1 126 2.38 102
PGAM+LK [55]114.7 6.56 124 16.0 122 2.71 124 10.0 115 14.7 98 3.00 122 7.75 125 15.7 125 2.38 122 13.7 126 21.1 125 6.27 127 18.8 121 25.8 122 5.26 114 11.6 114 27.2 95 4.08 122 10.7 112 40.3 66 2.58 116 15.1 104 24.4 109 2.38 102
Pyramid LK [2]116.5 6.24 119 13.7 118 3.16 127 12.7 128 15.8 124 3.79 128 11.8 128 12.3 112 3.00 128 25.5 129 41.4 128 7.14 129 22.9 128 33.6 128 5.20 61 10.7 86 25.4 52 3.92 114 11.2 120 49.2 124 2.65 125 19.6 129 37.8 129 2.38 102
Periodicity [78]127.4 6.81 126 17.5 124 3.27 129 15.3 129 16.9 128 4.24 129 13.7 129 22.7 129 4.36 129 18.0 128 41.4 128 6.16 126 23.9 129 34.4 129 5.60 129 12.2 125 34.5 128 4.51 128 11.8 123 55.6 127 2.65 125 17.9 127 29.7 125 2.71 129
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 Anonymous. Flow Fields: Dense unregularized correspondence fields for highly accurate large displacement optical flow estimation. ICCV 2015 submission 744.
[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. Submitted to TIP 2016.
[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.
* 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.