Optical flow evaluation results: Average angle error

Optical flow evaluation results	Statistics: Average SD R2.5 R5.0 R10.0 A50 A75 A95 Error type: endpoint angle interpolation normalized interpolation
Show images: below table above table

Average angle error	avg.	Army (Hidden texture) GT im0 im1			Mequon (Hidden texture) GT im0 im1			Schefflera (Hidden texture) GT im0 im1			Wooden (Hidden texture) GT im0 im1			Grove (Synthetic) GT im0 im1			Urban (Synthetic) GT im0 im1			Yosemite (Synthetic) GT im0 im1			Teddy (Stereo) GT im0 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
NNF-Local [87]	4.5	2.69 2	7.56 3	1.98 3	1.97 4	7.01 4	1.59 4	2.18 2	5.36 3	1.53 4	1.87 2	9.14 5	1.06 4	2.28 2	2.94 1	1.57 2	2.39 5	6.78 2	2.15 8	2.00 15	3.36 15	1.62 13	0.99 1	2.16 2	0.57 2
NN-field [71]	9.0	2.89 7	8.13 15	2.11 5	2.10 6	7.15 9	1.77 14	2.27 4	5.59 5	1.61 8	1.58 1	8.52 4	0.79 1	2.35 4	3.05 5	1.60 3	1.89 1	5.20 1	1.37 1	2.43 41	3.70 44	1.95 31	1.01 2	2.25 3	0.53 1
OFLAF [77]	11.4	3.04 14	7.80 9	2.40 12	2.14 7	7.02 5	1.72 9	2.25 3	5.32 2	1.56 5	2.62 15	13.7 21	1.37 18	2.35 4	3.13 6	1.62 4	2.98 18	7.73 7	2.57 17	2.08 20	3.27 10	2.05 34	1.33 12	2.43 6	1.40 15
PMMST [115]	12.7	3.42 39	7.60 4	2.65 27	2.32 11	6.39 1	2.20 30	2.63 11	6.08 8	2.03 23	2.06 4	6.07 1	1.44 23	2.60 10	3.27 8	1.91 10	2.56 6	6.78 2	2.09 4	2.06 17	3.53 33	1.63 14	1.27 9	2.29 4	1.02 6
nLayers [57]	14.9	2.80 5	7.42 2	2.20 7	2.71 27	7.24 10	2.55 55	2.61 9	6.24 9	2.45 48	2.30 9	12.7 11	1.16 7	2.30 3	3.02 3	1.70 5	2.62 9	6.95 4	2.09 4	2.29 35	3.46 23	1.89 28	1.38 14	3.06 17	1.29 13
MDP-Flow2 [68]	16.8	3.23 29	7.93 12	2.60 19	1.92 2	6.64 2	1.52 1	2.46 7	5.91 7	1.56 5	3.05 38	15.8 47	1.51 33	2.77 22	3.50 17	2.16 23	2.86 14	8.58 18	2.70 27	2.00 15	3.50 30	1.59 11	1.28 10	2.67 11	0.89 4
ComponentFusion [96]	17.3	2.78 4	8.20 16	2.05 4	2.04 5	7.31 11	1.66 8	2.55 8	6.78 13	1.61 8	2.24 8	13.1 13	1.01 3	2.71 19	3.56 19	2.10 19	3.55 45	12.4 50	3.22 51	2.19 30	3.60 39	1.54 10	1.32 11	2.91 13	1.13 8
TC/T-Flow [76]	19.7	2.69 2	7.75 8	1.87 2	2.76 31	10.2 42	1.73 10	3.33 23	9.01 29	1.49 2	2.86 30	16.7 55	1.21 9	2.60 10	3.49 16	1.90 9	2.21 2	7.65 5	2.04 3	1.84 8	3.23 7	3.14 79	2.03 36	4.53 36	1.49 18
FC-2Layers-FF [74]	22.5	3.02 13	7.87 11	2.61 20	2.72 28	9.35 35	2.29 37	2.36 5	5.47 4	2.15 31	2.48 10	12.6 10	1.28 10	2.49 7	3.19 7	2.03 15	3.39 34	8.92 19	2.83 36	2.83 65	3.92 57	2.80 59	1.25 7	2.57 10	1.20 10
WLIF-Flow [93]	23.1	2.96 9	7.67 5	2.40 12	2.41 16	7.70 15	2.10 25	2.98 16	7.63 18	1.97 22	2.71 22	13.5 18	1.33 13	3.01 38	4.00 43	2.40 39	3.03 20	8.32 12	2.44 14	2.09 22	3.36 15	2.04 33	2.26 41	4.97 41	2.59 46
Layers++ [37]	23.8	3.11 16	8.22 19	2.79 37	2.43 19	7.02 5	2.24 32	2.43 6	5.77 6	2.18 34	2.13 6	9.71 6	1.15 6	2.35 4	3.02 3	1.96 11	3.81 51	11.4 38	3.22 51	2.74 60	4.01 62	2.35 44	1.45 15	3.05 16	1.79 25
HAST [110]	24.3	2.58 1	7.12 1	1.81 1	2.41 16	7.05 7	2.10 25	1.83 1	4.19 1	1.17 1	2.84 29	15.5 43	1.08 5	2.23 1	2.97 2	1.40 1	3.72 50	10.0 30	3.92 72	3.40 83	4.90 88	5.66 107	1.20 6	2.09 1	1.24 11
FESL [72]	25.3	2.96 9	7.70 6	2.54 15	3.26 62	10.4 43	2.56 56	3.25 21	8.39 21	2.17 32	2.56 11	13.2 14	1.31 12	2.57 9	3.40 11	2.12 22	2.60 8	7.65 5	2.30 9	2.64 55	4.22 70	2.47 46	1.75 25	3.49 24	1.71 22
Efficient-NL [60]	25.4	2.99 12	8.23 20	2.28 8	2.72 28	8.95 30	2.25 35	3.81 36	9.87 34	2.07 27	2.77 26	14.3 28	1.46 28	2.61 12	3.48 15	1.96 11	3.31 30	8.33 13	2.59 19	2.60 50	3.75 45	2.54 49	1.60 20	3.02 14	1.66 20
AGIF+OF [85]	25.5	3.06 15	8.20 16	2.55 17	3.17 52	10.6 45	2.46 50	3.46 28	8.97 28	2.24 37	2.61 13	13.7 21	1.33 13	2.63 14	3.46 14	2.11 20	2.88 16	8.34 14	2.35 11	2.10 24	3.56 35	2.09 36	1.80 28	3.68 28	2.24 36
LME [70]	25.5	3.15 21	8.04 14	2.31 10	1.95 3	6.65 3	1.59 4	4.03 41	9.31 30	4.57 84	2.69 20	13.6 19	1.42 21	2.85 28	3.61 22	2.42 40	3.47 41	12.8 55	3.17 47	2.12 26	3.53 33	1.73 16	1.34 13	2.75 12	1.18 9
ALD-Flow [66]	25.8	2.82 6	7.86 10	2.16 6	2.84 37	10.1 40	1.86 16	3.73 34	10.4 37	1.67 11	3.10 41	16.8 56	1.28 10	2.69 18	3.60 21	1.85 8	2.79 11	11.3 37	2.32 10	2.07 19	3.25 9	3.10 76	2.03 36	5.11 42	1.94 29
NNF-EAC [103]	27.0	3.31 32	8.21 18	2.68 28	2.19 9	7.49 13	1.76 12	2.73 13	6.62 12	1.70 12	3.18 46	15.8 47	1.64 44	2.87 30	3.66 26	2.24 26	3.02 19	8.07 10	2.59 19	2.19 30	3.48 26	1.74 17	2.85 52	6.52 52	3.12 56
IROF++ [58]	27.3	3.17 23	8.69 27	2.61 20	2.79 33	9.61 36	2.33 38	3.43 25	8.86 25	2.38 43	2.87 31	14.8 33	1.52 35	2.74 20	3.57 20	2.19 24	3.20 27	9.70 27	2.71 28	1.96 13	3.45 22	1.22 5	1.80 28	4.06 30	2.50 43
PH-Flow [101]	28.0	3.19 26	8.87 32	2.71 29	2.84 37	9.33 34	2.37 40	2.85 14	7.20 15	2.36 40	2.92 33	15.4 40	1.51 33	2.63 14	3.42 12	2.04 16	3.03 20	8.52 17	2.49 16	2.69 57	3.60 39	3.13 78	1.25 7	2.53 8	1.34 14
Classic+CPF [83]	28.1	3.14 19	8.60 25	2.63 24	3.03 50	10.6 45	2.33 38	3.66 31	9.58 31	2.20 35	2.61 13	14.1 26	1.34 16	2.68 17	3.53 18	2.21 25	2.85 13	7.95 9	2.38 12	2.44 43	3.49 28	2.90 69	1.67 23	3.40 22	2.43 42
Sparse-NonSparse [56]	29.4	3.14 19	8.75 29	2.76 36	3.02 48	10.6 45	2.43 45	3.45 27	8.96 26	2.36 40	2.66 17	13.7 21	1.42 21	2.85 28	3.75 32	2.33 31	3.28 29	9.40 24	2.73 29	2.42 40	3.31 12	2.69 53	1.47 16	3.07 18	1.66 20
TC-Flow [46]	30.1	2.91 8	8.00 13	2.34 11	2.18 8	8.77 25	1.52 1	3.84 38	10.7 42	1.49 2	3.13 42	16.6 54	1.46 28	2.78 23	3.73 31	1.96 11	3.08 23	11.4 38	2.66 22	1.94 11	3.43 19	3.20 83	3.06 56	7.04 54	4.08 79
LSM [39]	31.5	3.12 17	8.62 26	2.75 35	3.00 46	10.5 44	2.44 47	3.43 25	8.85 24	2.35 39	2.66 17	13.6 19	1.44 23	2.82 25	3.68 27	2.36 33	3.38 33	9.41 25	2.81 34	2.69 57	3.52 31	2.84 63	1.59 19	3.38 21	1.80 26
Correlation Flow [75]	32.2	3.38 37	8.40 22	2.64 25	2.23 10	7.54 14	1.56 3	5.14 59	13.1 59	1.60 7	2.09 5	8.15 3	1.35 17	3.12 44	4.09 49	2.34 32	4.01 61	11.5 41	4.00 74	2.59 49	3.61 41	3.00 74	1.49 17	3.04 15	1.42 16
SVFilterOh [112]	32.4	3.63 44	8.82 30	2.86 39	2.60 25	8.06 18	2.05 24	2.95 15	7.09 14	2.03 23	2.80 28	13.8 24	1.41 20	2.63 14	3.42 12	1.75 7	3.49 42	10.3 32	3.23 53	3.63 90	5.75 107	4.47 100	1.09 4	2.45 7	0.92 5
Ramp [62]	32.7	3.18 25	8.83 31	2.73 32	2.89 41	10.1 40	2.44 47	3.27 22	8.43 22	2.38 43	2.74 24	14.2 27	1.46 28	2.82 25	3.69 29	2.29 29	3.37 32	9.31 22	2.93 40	2.62 53	3.38 18	3.19 82	1.54 18	3.21 19	2.24 36
PMF [73]	33.2	3.61 42	9.07 34	2.62 22	2.40 14	8.05 17	1.83 15	2.61 9	6.27 10	1.65 10	3.35 53	15.4 40	1.58 39	2.54 8	3.27 8	1.71 6	3.59 46	11.1 36	3.46 59	4.07 99	6.18 112	4.02 98	1.06 3	2.38 5	1.25 12
COFM [59]	33.4	3.17 23	9.90 48	2.46 14	2.41 16	8.34 22	1.92 18	3.77 35	10.5 38	2.54 51	2.71 22	14.9 35	1.19 8	3.08 42	3.92 39	3.25 78	3.83 53	10.9 33	3.15 46	2.20 33	3.35 13	2.91 71	1.62 22	2.56 9	2.09 33
FMOF [94]	34.2	3.12 17	8.23 20	2.73 32	3.25 59	10.7 52	2.52 53	3.01 17	7.61 17	2.20 35	2.56 11	13.4 16	1.33 13	2.75 21	3.61 22	2.24 26	3.66 48	8.50 16	2.78 32	2.62 53	3.84 52	3.27 85	2.66 49	5.69 45	1.95 31
Classic+NL [31]	35.5	3.20 28	8.72 28	2.81 38	3.02 48	10.6 45	2.44 47	3.46 28	8.84 23	2.38 43	2.78 27	14.3 28	1.46 28	2.83 27	3.68 27	2.31 30	3.40 35	9.09 21	2.76 31	2.87 67	3.82 51	2.86 66	1.67 23	3.53 25	2.26 39
TV-L1-MCT [64]	36.3	3.16 22	8.48 24	2.71 29	3.28 63	10.8 56	2.60 62	3.95 40	10.5 38	2.38 43	2.69 20	13.9 25	1.45 27	2.94 34	3.79 33	2.63 56	3.50 43	9.75 28	3.06 44	2.08 20	3.35 13	2.29 42	1.95 33	3.89 29	2.71 48
SimpleFlow [49]	39.9	3.35 33	9.20 37	2.98 46	3.18 55	10.7 52	2.71 65	5.06 57	12.6 57	2.70 56	2.95 35	15.1 37	1.58 39	2.91 33	3.79 33	2.47 42	3.59 46	9.49 26	2.99 42	2.39 38	3.46 23	2.24 41	1.60 20	3.56 26	1.57 19
CostFilter [40]	40.2	3.84 47	9.64 44	3.06 48	2.55 24	8.09 19	2.03 22	2.69 12	6.47 11	1.88 18	3.66 62	16.8 56	1.88 54	2.62 13	3.34 10	1.99 14	4.05 62	11.0 35	3.65 66	4.16 101	7.18 118	4.66 102	1.16 5	3.36 20	0.87 3
2DHMM-SAS [92]	41.8	3.19 26	8.89 33	2.71 29	3.20 57	11.5 63	2.38 41	5.19 60	12.2 52	2.73 58	2.92 33	15.2 38	1.53 36	2.79 24	3.65 25	2.27 28	3.45 39	9.34 23	2.78 32	2.66 56	3.56 35	3.07 75	2.34 44	5.12 43	2.97 54
MDP-Flow [26]	43.0	3.48 40	9.46 41	3.10 50	2.45 20	7.36 12	2.41 42	3.21 20	8.31 20	2.78 60	3.18 46	17.8 61	1.70 49	3.03 39	3.87 35	2.60 52	3.43 37	12.6 53	2.81 34	2.19 30	3.88 55	1.60 12	4.13 72	9.96 77	3.86 75
MLDP_OF [89]	43.0	4.13 59	10.3 55	3.60 74	2.34 12	7.70 15	1.88 17	4.23 44	10.9 43	1.87 17	2.74 24	14.6 32	1.37 18	3.10 43	3.91 38	2.48 46	3.40 35	9.00 20	3.79 69	3.46 85	4.20 68	5.55 106	2.31 42	4.64 38	1.98 32
S2D-Matching [84]	43.1	3.36 34	9.66 45	2.86 39	3.19 56	11.1 59	2.46 50	4.86 55	12.9 58	2.47 49	2.67 19	13.2 14	1.44 23	2.87 30	3.72 30	2.38 35	3.45 39	9.76 29	2.95 41	3.05 73	3.79 49	3.30 87	1.95 33	4.16 33	3.00 55
AggregFlow [97]	44.0	4.25 68	11.9 74	3.26 53	4.46 84	13.7 83	3.43 81	4.76 52	12.4 54	3.93 81	3.28 50	15.6 44	1.68 46	2.89 32	3.89 36	2.08 17	2.32 3	7.75 8	2.14 6	2.06 17	3.77 47	1.48 9	2.07 39	4.11 31	2.36 40
IROF-TV [53]	44.2	3.40 38	9.29 39	2.95 45	2.99 45	11.1 59	2.53 54	3.81 36	9.81 33	2.44 47	3.25 49	16.9 58	1.78 52	3.27 60	4.10 50	2.93 70	4.47 68	16.0 81	3.53 61	1.70 3	3.21 5	1.12 3	1.91 32	4.75 39	2.19 35
CombBMOF [114]	44.5	3.94 51	10.6 59	2.74 34	2.80 34	8.55 24	2.16 28	3.10 19	7.99 19	1.76 13	2.99 36	13.4 16	1.95 58	3.04 40	3.89 36	2.49 47	5.64 88	12.3 48	6.74 102	3.54 86	5.16 95	2.81 60	1.85 31	4.60 37	1.10 7
Split-Bregman [107]	46.0	4.07 56	11.3 68	2.55 17	2.75 30	9.20 33	2.00 20	4.82 54	12.3 53	2.13 30	3.05 38	11.8 8	1.88 54	3.33 62	4.26 63	2.69 60	4.52 71	12.1 46	3.98 73	2.71 59	4.30 71	2.77 56	1.78 27	3.57 27	1.92 28
FlowFields [111]	47.1	4.57 78	13.7 87	3.38 62	3.01 47	10.6 45	2.59 60	4.19 43	11.1 44	3.30 68	3.17 45	15.0 36	1.96 59	3.21 53	4.24 61	2.61 55	2.91 17	12.4 50	2.66 22	1.84 8	3.46 23	1.84 23	2.50 45	6.15 50	2.79 49
NL-TV-NCC [25]	47.8	3.89 49	9.16 36	2.98 46	2.87 40	9.69 37	1.99 19	4.44 48	11.6 47	1.76 13	2.64 16	11.8 8	1.48 32	3.49 75	4.60 82	2.47 42	4.67 76	13.5 60	4.26 81	2.83 65	4.57 79	2.84 63	2.62 47	6.00 49	2.25 38
Sparse Occlusion [54]	48.2	3.62 43	9.12 35	2.90 41	2.92 43	9.08 31	2.56 56	4.49 49	11.8 50	2.11 29	3.14 43	15.8 47	1.57 38	3.26 58	4.22 57	2.36 33	3.52 44	10.9 33	2.66 22	5.10 114	6.32 113	3.15 80	2.02 35	4.92 40	1.71 22
OFH [38]	48.6	3.90 50	9.77 47	3.62 77	2.84 37	11.0 58	2.04 23	5.52 65	14.4 66	1.89 19	3.52 56	20.5 76	1.60 42	3.18 47	4.06 48	2.82 65	3.86 54	14.1 68	3.59 63	1.77 5	3.62 42	1.81 20	2.64 48	7.08 56	2.15 34
PGM-C [121]	48.9	4.62 82	14.0 91	3.39 64	3.29 65	12.3 68	2.70 64	4.39 47	11.7 48	3.43 71	4.00 70	19.8 68	2.15 64	3.19 48	4.23 58	2.54 48	2.79 11	11.9 44	2.45 15	1.83 6	3.21 5	1.83 21	2.31 42	5.87 46	1.82 27
EPPM w/o HM [88]	49.2	4.25 68	11.1 63	3.13 51	2.36 13	8.35 23	1.76 12	3.72 33	10.2 36	1.81 15	3.24 48	14.5 31	1.94 57	3.16 45	3.94 40	2.82 65	4.78 79	12.9 56	4.32 82	3.64 92	4.54 78	5.73 108	1.76 26	4.11 31	1.94 29
Occlusion-TV-L1 [63]	50.1	3.59 41	9.61 42	2.64 25	2.93 44	10.6 45	2.41 42	6.16 71	15.2 69	2.70 56	3.32 51	17.0 59	1.68 46	3.38 65	4.44 72	2.82 65	3.10 25	13.2 59	2.68 25	2.17 27	3.52 31	1.46 7	4.63 81	11.1 91	3.53 64
Complementary OF [21]	51.1	4.44 73	11.2 66	4.04 84	2.51 23	9.77 39	1.74 11	3.93 39	10.6 40	2.04 25	3.87 66	18.8 63	2.19 67	3.17 46	4.00 43	2.92 69	4.64 74	13.8 65	3.64 65	2.17 27	3.36 15	2.51 47	3.08 57	7.04 54	3.65 68
Aniso-Texture [82]	52.2	2.96 9	7.72 7	2.54 15	2.48 22	8.26 21	2.24 32	6.48 73	15.9 78	2.63 53	1.96 3	10.1 7	0.98 2	3.26 58	4.21 55	2.60 52	5.74 90	16.9 88	5.61 93	4.47 107	5.88 110	3.33 88	3.51 63	7.12 57	3.68 70
Adaptive [20]	52.4	3.29 30	9.43 40	2.28 8	3.10 51	11.4 62	2.46 50	6.58 75	15.7 75	2.52 50	3.14 43	15.6 44	1.56 37	3.67 84	4.46 74	3.48 87	3.32 31	13.0 58	2.38 12	2.76 63	4.39 75	1.93 30	3.58 64	8.18 63	2.88 51
ACK-Prior [27]	53.2	4.19 64	9.27 38	3.60 74	2.40 14	8.21 20	1.65 7	3.40 24	8.96 26	1.84 16	2.87 31	14.4 30	1.44 23	3.36 64	4.15 51	3.07 73	6.35 98	16.1 83	4.90 87	4.21 103	4.80 83	6.03 110	3.29 60	5.99 48	2.82 50
CPM-Flow [117]	53.9	4.63 83	14.1 94	3.39 64	3.33 66	12.5 72	2.73 66	4.37 45	11.7 48	3.43 71	4.00 70	19.9 70	2.14 63	3.19 48	4.23 58	2.54 48	3.08 23	12.0 45	2.88 38	1.87 10	3.44 20	1.84 23	2.91 54	7.48 61	2.91 53
EpicFlow [102]	54.2	4.61 81	14.0 91	3.39 64	3.33 66	12.5 72	2.74 67	5.37 62	14.8 68	3.46 74	3.94 69	19.2 65	2.13 62	3.20 50	4.23 58	2.58 51	2.87 15	12.2 47	2.64 21	1.83 6	3.28 11	1.83 21	3.21 59	7.12 57	3.61 65
DPOF [18]	55.2	4.67 86	12.6 80	3.30 55	3.57 71	10.6 45	3.12 78	3.09 18	7.50 16	2.32 38	3.06 40	14.8 33	1.82 53	3.21 53	4.18 53	2.79 64	4.47 68	12.5 52	3.33 54	4.09 100	3.92 57	6.96 112	2.09 40	4.39 35	1.74 24
DeepFlow2 [109]	55.3	4.04 55	11.2 66	3.38 62	3.80 73	12.4 71	2.86 71	5.12 58	13.4 62	3.00 62	4.17 75	20.1 72	2.18 66	2.96 35	3.97 41	2.08 17	3.06 22	12.6 53	2.69 26	2.17 27	3.24 8	2.71 54	4.74 83	10.4 84	4.38 84
ROF-ND [108]	56.4	4.12 57	10.0 49	3.37 61	2.78 32	8.82 27	2.12 27	4.61 51	11.9 51	2.09 28	2.23 7	6.56 2	1.69 48	3.60 81	4.75 92	2.85 68	4.92 82	13.6 63	3.75 67	4.59 109	5.18 96	4.10 99	2.67 50	5.19 44	3.46 63
TCOF [69]	56.6	4.17 62	10.4 57	3.71 79	3.17 52	10.7 52	2.59 60	6.58 75	15.7 75	3.82 79	3.69 64	16.1 52	2.37 74	3.78 87	4.95 99	2.47 42	2.59 7	8.47 15	2.58 18	3.66 93	4.83 84	2.67 52	1.83 30	4.20 34	1.46 17
RFlow [90]	57.0	3.82 46	10.0 49	3.44 69	2.61 26	9.73 38	2.02 21	5.66 67	14.5 67	2.05 26	3.93 68	23.1 90	1.90 56	3.24 55	4.19 54	2.66 59	4.12 65	15.2 77	3.34 56	2.61 51	3.56 35	2.65 51	4.48 77	10.5 87	3.93 78
HBM-GC [105]	58.1	5.25 91	10.5 58	4.34 91	3.17 52	8.78 26	2.94 74	4.38 46	10.6 40	2.68 55	3.59 59	12.8 12	2.47 77	2.96 35	3.64 24	2.64 57	3.96 60	8.26 11	3.56 62	4.40 105	5.92 111	3.62 92	2.55 46	6.34 51	3.29 59
Steered-L1 [119]	58.3	3.30 31	8.44 23	2.91 42	1.89 1	7.14 8	1.60 6	3.61 30	9.91 35	1.89 19	3.45 54	19.4 67	1.64 44	3.42 68	4.30 64	3.39 81	5.18 83	14.5 70	4.37 84	5.09 113	5.05 91	10.1 116	5.56 90	10.2 82	6.24 97
ComplOF-FED-GPU [35]	59.7	4.28 70	11.3 68	3.70 78	3.25 59	13.0 76	2.16 28	4.06 42	11.2 45	1.95 21	3.91 67	19.2 65	2.01 60	3.20 50	4.15 51	2.64 57	4.61 73	16.1 83	3.90 71	2.98 72	3.77 47	3.69 93	2.85 52	7.44 60	2.53 45
SRR-TVOF-NL [91]	60.0	4.47 75	10.9 61	3.32 58	4.04 77	13.2 79	2.90 72	4.81 53	12.5 55	3.15 66	3.33 52	15.3 39	1.61 43	3.24 55	4.03 47	2.70 62	3.94 58	11.8 43	3.33 54	4.16 101	5.21 98	3.44 91	2.06 38	3.48 23	2.42 41
TF+OM [100]	61.9	3.97 52	10.2 52	2.94 44	2.91 42	9.12 32	2.57 59	5.22 61	11.5 46	6.92 89	3.59 59	16.1 52	2.28 71	3.20 50	3.97 41	3.11 74	4.70 77	14.5 70	4.32 82	3.06 75	4.84 85	2.71 54	3.93 68	8.79 68	4.32 83
Aniso. Huber-L1 [22]	62.4	3.71 45	10.1 51	3.08 49	4.36 83	13.0 76	3.77 85	6.92 79	15.3 71	3.60 77	3.54 57	15.9 50	2.04 61	3.38 65	4.45 73	2.47 42	3.88 55	12.9 56	2.74 30	3.37 82	4.36 74	2.85 65	3.16 58	7.52 62	2.90 52
DeepFlow [86]	63.5	4.49 76	11.7 71	4.14 86	4.26 80	12.8 75	3.36 79	5.96 68	14.2 65	5.10 85	4.89 86	23.1 90	2.67 80	2.98 37	4.00 43	2.11 20	3.26 28	13.5 60	2.84 37	2.09 22	3.10 2	2.77 56	5.83 92	11.4 93	5.45 94
Classic++ [32]	64.2	3.37 36	9.67 46	2.91 42	3.28 63	12.1 66	2.61 63	5.46 64	14.1 64	3.00 62	3.63 61	20.2 74	1.70 49	3.24 55	4.34 66	2.60 52	4.65 75	16.0 81	3.60 64	3.09 76	3.94 60	3.28 86	4.64 82	10.4 84	3.71 71
TV-L1-improved [17]	64.5	3.36 34	9.63 43	2.62 22	2.82 35	10.7 52	2.23 31	6.50 74	15.8 77	2.73 58	3.80 65	21.3 81	1.76 51	3.34 63	4.38 70	2.39 36	5.97 92	18.1 93	5.67 94	3.57 88	4.92 89	3.43 90	4.01 71	9.84 76	3.44 62
LocallyOriented [52]	66.8	4.54 77	12.8 82	3.27 54	4.73 88	14.8 90	3.73 84	7.77 85	18.3 91	3.44 73	3.56 58	15.6 44	2.22 68	3.46 72	4.47 75	2.69 60	3.15 26	10.2 31	3.19 49	2.61 51	4.20 68	2.52 48	4.39 74	8.52 65	5.23 90
SIOF [67]	67.1	4.23 66	10.2 52	3.31 57	3.97 75	14.5 88	2.97 75	7.81 87	16.4 81	7.48 90	4.82 82	20.1 72	2.96 83	3.54 78	4.49 76	3.12 75	4.31 66	13.5 60	4.13 77	2.36 37	3.59 38	1.68 15	3.46 62	7.39 59	3.37 60
TriangleFlow [30]	68.9	4.12 57	10.6 59	3.47 70	3.47 70	13.1 78	2.41 42	6.00 69	15.2 69	2.17 32	2.99 36	16.0 51	1.58 39	4.46 106	5.79 111	4.15 99	5.42 86	13.9 67	5.24 88	3.10 78	5.47 103	2.90 69	3.02 55	6.82 53	3.64 67
CRTflow [80]	69.6	4.18 63	11.8 73	3.20 52	3.22 58	10.8 56	2.43 45	6.20 72	15.5 73	2.63 53	4.21 76	22.0 84	2.24 69	3.32 61	4.34 66	2.44 41	7.43 105	19.3 99	8.15 107	2.55 47	4.09 64	2.59 50	4.60 79	11.2 92	4.45 85
Brox et al. [5]	69.6	4.44 73	12.4 76	4.22 89	3.72 72	13.5 82	3.06 76	4.97 56	13.3 61	3.11 64	4.58 80	22.0 84	2.37 74	3.79 89	4.60 82	4.33 103	3.91 57	17.0 89	3.45 58	2.22 34	3.79 49	1.19 4	4.62 80	10.0 78	3.38 61
Rannacher [23]	71.5	4.13 59	11.0 62	3.61 76	3.39 68	12.3 68	2.80 70	7.26 81	17.4 87	3.59 76	4.40 78	23.1 90	2.24 69	3.43 70	4.54 79	2.56 50	5.41 85	18.5 94	4.23 79	2.92 69	3.91 56	2.82 61	3.45 61	9.14 69	3.27 58
F-TV-L1 [15]	72.5	5.44 94	12.5 79	5.69 98	5.46 92	15.0 93	4.03 87	7.48 82	16.3 80	3.42 70	5.08 88	23.3 93	2.81 82	3.42 68	4.34 66	3.03 71	4.05 62	15.1 76	3.18 48	2.43 41	3.92 57	1.87 25	3.90 67	9.35 73	2.61 47
Local-TV-L1 [65]	73.1	5.33 92	12.6 80	5.19 96	6.90 98	15.7 96	6.22 96	10.0 97	18.2 90	8.89 91	5.81 93	24.7 98	3.70 93	3.05 41	4.00 43	2.39 36	4.05 62	14.6 72	3.09 45	1.95 12	3.11 3	2.15 38	5.85 93	10.8 89	7.34 100
SuperFlow [81]	73.2	4.16 61	11.1 63	3.32 58	4.80 89	12.2 67	4.68 91	7.80 86	16.0 79	10.6 98	5.16 90	22.4 88	3.24 90	3.39 67	4.24 61	3.71 91	3.44 38	13.7 64	2.91 39	3.19 79	4.62 81	1.87 25	4.74 83	10.6 88	4.24 81
DF-Auto [116]	73.5	5.04 90	13.7 87	3.30 55	6.51 95	14.1 87	6.09 95	8.14 90	16.5 82	10.2 96	5.06 87	21.3 81	3.10 89	3.74 85	4.91 98	3.25 78	2.67 10	11.4 38	2.14 6	3.36 81	5.23 99	1.45 6	4.45 76	9.18 70	4.28 82
TriFlow [95]	73.9	4.73 88	12.4 76	3.49 72	4.03 76	12.5 72	3.70 83	8.18 92	17.2 85	10.4 97	3.50 55	15.4 40	2.32 73	3.43 70	4.21 55	3.42 82	3.90 56	12.3 48	3.76 68	7.86 119	5.72 106	16.2 119	2.80 51	5.89 47	2.50 43
CLG-TV [48]	74.2	4.00 53	10.3 55	3.40 67	4.33 82	12.3 68	4.08 88	6.78 77	15.5 73	3.64 78	4.07 72	17.7 60	2.39 76	3.79 89	4.86 95	3.23 77	4.48 70	16.5 86	3.80 70	3.55 87	4.65 82	2.89 68	4.00 70	10.1 80	3.18 57
CBF [12]	75.3	3.88 48	10.2 52	3.50 73	4.60 86	11.3 61	5.06 92	5.43 63	13.1 59	3.39 69	4.09 73	21.2 80	2.16 65	3.80 92	4.72 91	3.52 88	4.33 67	14.4 69	3.01 43	4.97 111	5.51 104	4.93 104	3.99 69	9.27 72	3.91 77
Bartels [41]	77.1	4.43 71	11.1 63	4.17 88	2.83 36	8.84 28	2.56 56	4.54 50	12.5 55	2.80 61	4.87 83	22.1 86	3.05 87	3.58 80	4.35 69	4.15 99	5.55 87	17.5 90	5.78 95	3.74 94	5.02 90	5.98 109	5.21 89	11.9 95	5.20 89
Fusion [6]	77.3	4.43 71	13.7 87	4.08 85	2.47 21	8.91 29	2.24 32	3.70 32	9.68 32	3.12 65	3.68 63	19.8 68	2.54 79	4.26 103	5.16 104	4.31 102	6.32 95	16.8 87	6.15 99	4.55 108	5.78 108	3.10 76	7.12 103	13.6 103	7.86 104
p-harmonic [29]	77.6	4.64 84	13.0 83	4.43 92	3.41 69	11.9 64	2.93 73	7.60 83	18.1 89	3.96 82	4.65 81	21.0 78	2.97 85	3.46 72	4.33 65	3.34 80	4.75 78	17.5 90	4.60 86	3.05 73	4.17 66	2.15 38	5.09 88	10.9 90	3.77 73
CNN-flow-warp+ref [118]	78.4	4.93 89	14.5 98	4.29 90	4.18 79	11.9 64	4.24 89	8.23 93	19.7 98	6.35 88	5.13 89	24.4 97	2.96 83	3.55 79	4.40 71	3.85 92	3.82 52	15.0 74	3.39 57	1.96 13	3.44 20	2.14 37	10.0 111	14.8 108	10.8 110
Dynamic MRF [7]	79.2	4.58 79	12.4 76	4.14 86	3.25 59	13.9 84	2.27 36	6.02 70	16.8 83	2.36 40	4.39 77	22.6 89	2.51 78	3.61 82	4.55 80	3.46 84	6.81 100	22.2 109	6.78 104	2.41 39	3.48 26	3.69 93	9.26 109	17.8 112	10.2 107
SegOF [10]	79.8	5.85 95	13.5 86	3.98 83	7.40 99	14.9 91	8.13 104	8.55 95	17.3 86	9.01 92	6.50 97	18.1 62	5.14 99	3.90 96	4.53 78	4.81 107	6.57 99	21.7 107	6.81 105	1.65 2	3.49 28	1.08 2	3.71 65	9.23 71	3.63 66
FlowNetS+ft+v [113]	80.3	4.22 65	12.1 75	3.48 71	4.50 85	13.4 80	3.85 86	8.29 94	18.4 92	6.20 87	4.87 83	21.6 83	3.01 86	3.93 97	5.04 101	3.47 86	3.71 49	15.3 78	3.21 50	3.32 80	5.12 93	3.87 95	3.76 66	9.44 74	3.74 72
LDOF [28]	81.0	4.60 80	13.0 83	3.77 80	4.67 87	15.5 95	3.67 82	5.63 66	14.0 63	4.21 83	5.80 92	27.1 106	3.43 91	3.52 77	4.50 77	3.46 84	4.84 81	17.8 92	4.04 75	2.46 45	4.14 65	3.25 84	4.85 86	12.0 96	3.78 74
Second-order prior [8]	81.2	4.03 54	11.6 70	3.35 60	3.88 74	14.0 86	3.08 77	7.21 80	17.6 88	3.57 75	4.14 74	19.9 70	2.31 72	3.66 83	4.86 95	2.73 63	7.32 103	21.2 105	6.76 103	4.02 97	4.58 80	4.01 97	4.27 73	10.4 84	5.12 86
StereoFlow [44]	84.5	17.1 121	28.1 121	17.9 120	18.7 118	29.7 119	16.5 113	20.1 118	30.9 118	17.5 115	21.2 119	38.3 120	17.9 117	4.60 107	5.05 102	5.52 109	2.38 4	11.5 41	1.77 2	1.25 1	2.92 1	0.71 1	4.49 78	10.3 83	4.23 80
Ad-TV-NDC [36]	85.6	8.36 108	14.0 91	11.1 114	12.9 111	19.9 105	12.8 111	14.4 107	23.1 101	12.1 101	7.40 100	20.6 77	6.33 100	3.47 74	4.66 87	2.39 36	3.95 59	13.8 65	3.51 60	2.48 46	3.75 45	2.05 34	9.75 110	12.1 97	16.7 116
Learning Flow [11]	88.9	4.23 66	11.7 71	3.41 68	4.16 78	15.3 94	3.42 80	6.78 77	16.9 84	3.83 80	6.41 96	25.3 101	4.25 95	4.66 109	6.01 115	4.00 95	6.33 97	20.7 104	5.30 89	3.09 76	4.84 85	2.91 71	7.08 102	15.0 109	5.27 91
Shiralkar [42]	89.2	4.64 84	14.1 94	3.94 81	4.29 81	16.9 98	2.77 68	7.75 84	18.8 94	3.19 67	5.54 91	25.0 100	3.56 92	3.51 76	4.55 80	3.04 72	7.41 104	20.1 103	6.41 100	3.76 95	4.35 73	5.28 105	6.56 99	14.4 107	5.30 92
StereoOF-V1MT [120]	89.7	4.71 87	14.1 94	3.95 82	5.10 91	20.3 107	2.78 69	7.98 89	20.7 99	2.57 52	4.48 79	21.1 79	2.79 81	4.20 102	5.29 106	4.10 97	6.85 102	22.3 110	6.42 101	2.45 44	4.17 66	3.15 80	10.5 112	18.4 115	10.5 108
IAOF2 [51]	90.3	5.38 93	13.7 87	4.50 93	5.95 94	14.6 89	5.61 94	8.80 96	18.8 94	9.40 93	12.2 109	23.8 96	13.1 112	3.86 93	4.89 97	3.12 75	5.21 84	14.9 73	4.54 85	4.33 104	5.15 94	3.93 96	4.39 74	8.57 66	3.87 76
Modified CLG [34]	91.6	7.17 103	17.1 106	6.47 101	6.85 97	14.9 91	7.48 100	14.0 104	24.8 105	15.7 111	8.35 103	27.3 107	6.36 101	3.96 98	4.99 100	4.08 96	4.54 72	19.3 99	4.15 78	2.33 36	3.86 54	2.40 45	6.00 94	13.8 105	5.40 93
2D-CLG [1]	92.0	10.1 110	22.6 114	7.59 106	9.84 106	16.9 98	11.1 110	16.9 113	28.2 113	18.8 117	14.1 112	31.1 111	13.1 112	3.86 93	4.62 85	4.53 104	5.98 93	21.2 105	5.97 97	1.76 4	3.14 4	1.46 7	6.29 96	12.9 102	5.81 95
Filter Flow [19]	92.7	6.48 97	14.6 99	4.96 94	5.73 93	15.7 96	5.07 93	10.1 98	18.6 93	14.3 107	9.04 105	23.3 93	7.80 105	3.98 99	4.71 89	4.21 101	5.86 91	15.0 74	5.41 92	4.98 112	6.87 115	2.78 58	4.82 85	8.66 67	3.65 68
GraphCuts [14]	92.7	6.25 96	14.3 97	5.53 97	8.60 102	20.1 106	6.61 98	7.91 88	15.4 72	10.9 99	4.88 85	19.0 64	3.05 87	3.78 87	4.71 89	3.94 93	8.74 110	16.4 85	5.39 91	4.04 98	4.87 87	4.85 103	6.35 97	12.2 98	6.05 96
SPSA-learn [13]	92.9	6.84 102	16.7 104	6.74 102	8.47 101	19.4 103	7.49 101	12.5 100	23.1 101	13.1 105	8.40 104	25.8 103	7.08 103	3.87 95	4.66 87	4.10 97	6.32 95	18.8 95	6.89 106	2.56 48	3.85 53	1.79 18	7.29 104	12.5 100	7.47 102
GroupFlow [9]	94.5	8.00 105	18.6 107	8.09 108	11.1 109	23.7 112	10.3 108	12.6 101	25.6 106	12.8 103	5.84 94	20.3 75	4.39 96	4.69 110	5.81 112	3.67 89	9.29 111	22.4 111	10.1 113	2.11 25	3.99 61	2.29 42	5.75 91	10.0 78	7.39 101
HBpMotionGpu [43]	94.5	6.57 99	15.0 101	5.17 95	8.29 100	18.0 100	8.29 105	14.1 105	26.5 108	13.2 106	6.12 95	25.3 101	3.94 94	3.79 89	4.62 85	3.97 94	4.80 80	15.7 79	4.11 76	4.40 105	5.20 97	2.87 67	6.28 95	11.7 94	7.31 99
IAOF [50]	95.1	6.49 98	14.6 99	6.42 100	9.22 105	18.5 101	7.94 103	16.4 112	27.4 111	13.0 104	8.22 101	22.2 87	7.73 104	3.77 86	4.76 93	3.42 82	6.84 101	18.8 95	4.23 79	3.59 89	4.46 76	2.83 62	7.51 106	10.1 80	10.6 109
Black & Anandan [4]	95.4	6.81 101	15.4 102	7.43 104	8.77 103	19.5 104	7.35 99	13.0 102	22.9 100	12.5 102	8.29 102	26.1 104	6.77 102	4.18 101	5.28 105	3.69 90	6.19 94	20.0 102	5.34 90	3.63 90	5.05 91	1.79 18	6.45 98	12.2 98	5.17 88
Nguyen [33]	98.3	7.88 104	16.8 105	7.02 103	13.4 112	19.0 102	15.3 112	17.6 114	28.9 114	17.2 114	12.0 108	26.9 105	11.6 110	4.38 104	5.07 103	5.58 112	5.69 89	19.7 101	5.93 96	2.75 61	4.02 63	1.91 29	6.59 100	12.5 100	6.52 98
BlockOverlap [61]	98.7	6.67 100	13.1 85	5.87 99	6.62 96	13.9 84	6.53 97	10.6 99	19.5 97	10.1 95	6.97 99	24.9 99	5.13 98	4.38 104	4.61 84	6.37 115	7.47 106	15.7 79	6.05 98	6.23 115	6.41 114	13.0 118	6.92 101	9.60 75	12.2 112
2bit-BM-tele [98]	100.8	8.00 105	15.8 103	8.40 110	4.91 90	13.4 80	4.67 90	8.14 90	19.0 96	5.12 86	6.62 98	23.5 95	5.04 97	4.08 100	4.78 94	4.61 106	8.68 109	18.8 95	8.31 108	6.46 117	7.08 117	9.47 115	7.36 105	14.1 106	9.62 106
Horn & Schunck [3]	103.8	8.01 107	19.9 108	8.38 109	9.13 104	23.2 111	7.71 102	14.2 106	25.9 107	14.6 109	12.4 110	30.6 109	11.3 109	4.64 108	5.64 108	4.60 105	8.21 108	24.4 113	8.45 109	4.01 96	5.41 100	1.95 31	9.16 108	17.5 110	8.86 105
SILK [79]	105.0	9.34 109	20.4 109	10.5 113	10.4 107	21.9 108	10.3 108	16.0 111	27.5 112	14.5 108	10.3 106	29.0 108	8.54 106	4.81 111	5.65 109	5.56 111	9.41 112	25.4 115	8.74 110	2.79 64	3.68 43	4.62 101	10.9 113	17.8 112	12.3 113
Heeger++ [104]	106.2	11.9 115	21.8 112	8.08 107	12.5 110	29.7 119	9.42 107	14.8 108	27.1 109	9.68 94	14.3 113	31.0 110	12.7 111	4.98 113	5.74 110	4.97 108	17.5 119	34.1 120	18.4 119	2.75 61	5.44 101	2.15 38	12.3 115	18.8 116	14.8 114
TI-DOFE [24]	107.2	13.4 118	23.2 115	16.5 119	16.5 115	24.1 113	18.2 117	20.2 119	31.1 119	20.6 118	19.9 118	32.9 114	20.8 119	4.89 112	5.90 113	5.54 110	8.04 107	23.9 112	8.81 111	2.97 71	4.34 72	1.88 27	10.9 113	17.7 111	11.9 111
HCIC-L [99]	110.5	15.7 120	22.0 113	10.1 112	31.5 121	26.6 117	41.0 121	14.8 108	23.1 101	16.8 113	18.4 117	34.4 116	18.2 118	5.94 116	6.35 116	6.35 114	10.6 115	19.2 98	11.4 115	18.7 121	17.8 121	19.2 120	4.93 87	8.34 64	5.16 87
SLK [47]	110.8	11.6 113	26.0 119	14.6 118	15.3 114	25.0 115	17.5 115	17.8 116	30.1 117	18.1 116	25.4 121	33.6 115	28.0 121	5.25 114	5.90 113	7.03 116	10.3 114	27.4 117	10.6 114	2.89 68	4.47 77	2.94 73	14.9 118	20.7 118	18.8 117
FFV1MT [106]	111.6	12.0 116	23.3 116	8.83 111	10.7 108	26.6 117	8.71 106	15.6 110	29.0 115	12.0 100	16.6 116	36.3 119	15.5 115	6.51 119	6.40 117	10.4 119	16.2 118	30.7 119	17.7 118	3.41 84	5.44 101	3.35 89	12.3 115	18.8 116	14.8 114
Adaptive flow [45]	113.1	13.2 117	20.8 110	14.0 117	17.1 117	22.0 109	17.9 116	18.1 117	27.1 109	22.8 120	11.8 107	31.1 111	10.5 107	6.35 118	7.13 119	6.25 113	9.87 113	21.8 108	9.44 112	12.6 120	11.4 120	20.0 121	7.75 107	13.6 103	7.73 103
PGAM+LK [55]	114.2	11.8 114	25.6 117	13.9 116	14.8 113	24.4 114	16.7 114	13.2 103	24.0 104	15.0 110	16.2 115	41.2 121	15.3 114	5.40 115	5.45 107	8.10 117	12.3 117	26.5 116	12.1 116	7.42 118	8.24 119	7.87 113	13.2 117	18.3 114	19.4 118
Periodicity [78]	115.1	11.2 112	27.0 120	7.46 105	16.6 116	29.8 121	18.2 117	25.3 121	31.2 121	24.9 121	12.7 111	35.7 118	11.1 108	31.7 121	41.4 121	25.1 121	23.8 121	41.5 121	23.8 121	2.92 69	5.62 105	6.90 111	18.6 120	33.1 121	22.3 119
FOLKI [16]	115.8	10.5 111	25.6 117	11.9 115	20.9 119	26.2 116	26.1 119	17.6 114	31.1 119	16.5 112	15.4 114	32.6 113	16.0 116	6.16 117	6.53 118	9.07 118	12.2 116	29.7 118	13.0 117	4.67 110	5.83 109	9.41 114	18.2 119	22.8 119	25.1 120
Pyramid LK [2]	118.2	13.9 119	20.9 111	21.4 121	24.1 120	23.1 110	30.2 120	20.9 120	29.5 116	21.9 119	22.2 120	34.6 117	25.0 120	18.7 120	23.1 120	20.2 120	21.2 120	24.5 114	21.0 120	6.41 116	7.02 116	10.8 117	25.6 121	31.5 120	34.5 121

Move the mouse over the numbers in the table to see the corresponding images. Click to compare with the ground truth.

References

Method	time*	frames	color	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] Split-Bregman	97	2	color	Anonymous. Optical flow estimation based on split Bregman method. ICCV 2015 submission 120.
[108] 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.
[109] DeepFlow2	16	2	color	J. Revaud, P. Weinzaepfel, Z. Harchaoui, and C. Schmid. Deep convolutional matching. Submitted to IJCV, 2015.
[110] HAST	2667	2	color	Anonymous. Highly accurate optical flow estimation on superpixel tree. ICCV 2015 submission 2221.
[111] FlowFields	15	2	color	Anonymous. Flow Fields: Dense unregularized correspondence fields for highly accurate large displacement optical flow estimation. ICCV 2015 submission 744.
[112] SVFilterOh	1.56	2	color	Anonymous. Fast estimation of large displacement optical flow using PatchMatch and dominant motion patterns. CVPR 2016 submission 1788.
[113] FlowNetS+ft+v	0.5	2	color	Anonymous. Learning optical flow with convolutional neural networks. ICCV 2015 submission 235.
[114] 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.)
[115] 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.
[116] DF-Auto	70	2	color	N. Monzon, A. Salgado, and J. Sanchez. Regularization strategies for discontinuity-preserving optical flow methods. Submitted to TIP 2015.
[117] CPM-Flow	3	2	color	Anonymous. Efficient coarse-to-fine PatchMatch for large displacement optical flow. CVPR 2016 submission 241.
[118] 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.
[119] Steered-L1	804	2	color	Anonymous. Optical flow estimation via steered-L1 norm. Submitted to WSCG 2016.
[120] 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.
[121] PGM-C	5	2	color	Y. Li. Pyramidal gradient matching for optical flow estimation. Submitted to PAMI 2016.

* 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.