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]	5.8	2.69 3	7.56 4	1.98 3	1.97 4	7.01 4	1.59 4	2.18 2	5.36 3	1.53 4	1.87 4	9.14 8	1.06 4	2.28 2	2.94 1	1.57 2	2.39 5	6.78 2	2.15 9	2.00 26	3.36 18	1.62 21	0.99 1	2.16 3	0.57 2
NN-field [71]	11.0	2.89 8	8.13 16	2.11 5	2.10 6	7.15 9	1.77 15	2.27 4	5.59 5	1.61 8	1.58 1	8.52 7	0.79 1	2.35 4	3.05 5	1.60 3	1.89 1	5.20 1	1.37 1	2.43 57	3.70 55	1.95 44	1.01 2	2.25 4	0.53 1
OFLAF [77]	14.1	3.04 15	7.80 10	2.40 14	2.14 7	7.02 5	1.72 9	2.25 3	5.32 2	1.56 5	2.62 21	13.7 28	1.37 23	2.35 4	3.13 6	1.62 4	2.98 23	7.73 7	2.57 22	2.08 32	3.27 12	2.05 49	1.33 13	2.43 7	1.40 17
PMMST [114]	15.2	3.42 43	7.60 5	2.65 32	2.32 11	6.39 1	2.20 34	2.63 11	6.08 8	2.03 28	2.06 7	6.07 3	1.44 31	2.60 10	3.27 8	1.91 10	2.56 7	6.78 2	2.09 5	2.06 28	3.53 38	1.63 22	1.27 10	2.29 5	1.02 7
nLayers [57]	17.6	2.80 6	7.42 3	2.20 8	2.71 31	7.24 10	2.55 61	2.61 9	6.24 9	2.45 54	2.30 13	12.7 15	1.16 8	2.30 3	3.02 3	1.70 5	2.62 11	6.95 4	2.09 5	2.29 49	3.46 27	1.89 41	1.38 15	3.06 18	1.29 15
MDP-Flow2 [68]	20.3	3.23 31	7.93 13	2.60 24	1.92 2	6.64 2	1.52 1	2.46 7	5.91 7	1.56 5	3.05 47	15.8 60	1.51 41	2.77 24	3.50 18	2.16 28	2.86 19	8.58 18	2.70 33	2.00 26	3.50 35	1.59 19	1.28 11	2.67 12	0.89 4
ComponentFusion [96]	20.5	2.78 5	8.20 17	2.05 4	2.04 5	7.31 11	1.66 8	2.55 8	6.78 13	1.61 8	2.24 12	13.1 18	1.01 3	2.71 21	3.56 20	2.10 23	3.55 53	12.4 61	3.22 60	2.19 43	3.60 47	1.54 18	1.32 12	2.91 14	1.13 9
TC/T-Flow [76]	23.0	2.69 3	7.75 9	1.87 2	2.76 34	10.2 48	1.73 10	3.33 25	9.01 31	1.49 2	2.86 37	16.7 70	1.21 10	2.60 10	3.49 17	1.90 9	2.21 2	7.65 5	2.04 4	1.84 14	3.23 9	3.14 101	2.03 39	4.53 40	1.49 21
FC-2Layers-FF [74]	27.0	3.02 14	7.87 12	2.61 25	2.72 32	9.35 36	2.29 42	2.36 5	5.47 4	2.15 35	2.48 14	12.6 14	1.28 14	2.49 7	3.19 7	2.03 17	3.39 42	8.92 20	2.83 44	2.83 81	3.92 71	2.80 80	1.25 8	2.57 11	1.20 12
WLIF-Flow [93]	28.0	2.96 10	7.67 6	2.40 14	2.41 16	7.70 15	2.10 28	2.98 17	7.63 19	1.97 27	2.71 29	13.5 24	1.33 17	3.01 45	4.00 51	2.40 47	3.03 26	8.32 12	2.44 17	2.09 34	3.36 18	2.04 48	2.26 47	4.97 49	2.59 56
Layers++ [37]	28.5	3.11 18	8.22 20	2.79 44	2.43 19	7.02 5	2.24 37	2.43 6	5.77 6	2.18 38	2.13 9	9.71 10	1.15 7	2.35 4	3.02 3	1.96 11	3.81 61	11.4 46	3.22 60	2.74 76	4.01 76	2.35 63	1.45 16	3.05 17	1.79 31
HAST [109]	29.2	2.58 1	7.12 1	1.81 1	2.41 16	7.05 7	2.10 28	1.83 1	4.19 1	1.17 1	2.84 36	15.5 54	1.08 5	2.23 1	2.97 2	1.40 1	3.72 58	10.0 33	3.92 85	3.40 104	4.90 110	5.66 133	1.20 7	2.09 1	1.24 13
AGIF+OF [85]	30.1	3.06 16	8.20 17	2.55 22	3.17 60	10.6 55	2.46 55	3.46 30	8.97 30	2.24 41	2.61 19	13.7 28	1.33 17	2.63 15	3.46 15	2.11 24	2.88 21	8.34 14	2.35 13	2.10 36	3.56 42	2.09 51	1.80 30	3.68 30	2.24 42
Efficient-NL [60]	30.4	2.99 13	8.23 21	2.28 9	2.72 32	8.95 32	2.25 40	3.81 40	9.87 37	2.07 32	2.77 33	14.3 37	1.46 36	2.61 12	3.48 16	1.96 11	3.31 38	8.33 13	2.59 24	2.60 67	3.75 56	2.54 70	1.60 22	3.02 15	1.66 23
LME [70]	30.4	3.15 23	8.04 15	2.31 11	1.95 3	6.65 3	1.59 4	4.03 46	9.31 32	4.57 98	2.69 27	13.6 26	1.42 28	2.85 31	3.61 23	2.42 49	3.47 49	12.8 68	3.17 56	2.12 38	3.53 38	1.73 24	1.34 14	2.75 13	1.18 11
FESL [72]	30.5	2.96 10	7.70 7	2.54 20	3.26 71	10.4 52	2.56 62	3.25 23	8.39 23	2.17 36	2.56 16	13.2 19	1.31 16	2.57 9	3.40 11	2.12 26	2.60 9	7.65 5	2.30 11	2.64 72	4.22 85	2.47 67	1.75 28	3.49 27	1.71 26
ALD-Flow [66]	30.7	2.82 7	7.86 11	2.16 6	2.84 41	10.1 45	1.86 18	3.73 38	10.4 41	1.67 12	3.10 49	16.8 71	1.28 14	2.69 20	3.60 22	1.85 8	2.79 15	11.3 45	2.32 12	2.07 30	3.25 11	3.10 98	2.03 39	5.11 50	1.94 34
RNLOD-Flow [121]	31.8	2.66 2	7.33 2	2.17 7	2.53 27	9.46 38	1.86 18	3.94 44	10.7 47	1.95 24	2.50 15	13.5 24	1.21 10	2.68 18	3.62 25	2.05 19	2.99 24	8.59 19	2.75 37	3.00 91	4.54 97	3.25 106	1.48 18	3.24 21	1.76 30
IROF++ [58]	32.0	3.17 25	8.69 30	2.61 25	2.79 36	9.61 39	2.33 43	3.43 27	8.86 27	2.38 48	2.87 39	14.8 42	1.52 43	2.74 22	3.57 21	2.19 30	3.20 34	9.70 30	2.71 34	1.96 24	3.45 26	1.22 9	1.80 30	4.06 32	2.50 52
NNF-EAC [103]	32.5	3.31 35	8.21 19	2.68 34	2.19 9	7.49 13	1.76 13	2.73 13	6.62 12	1.70 14	3.18 54	15.8 60	1.64 53	2.87 34	3.66 28	2.24 32	3.02 25	8.07 10	2.59 24	2.19 43	3.48 31	1.74 25	2.85 64	6.52 66	3.12 68
ProFlow_ROB [146]	32.5	3.29 32	9.91 56	2.35 13	2.50 25	10.0 44	1.83 16	4.04 47	11.6 55	1.96 26	2.86 37	15.0 45	1.22 12	2.87 34	3.89 42	1.97 14	2.60 9	10.5 36	2.20 10	1.53 4	3.54 40	1.53 17	2.50 53	6.37 65	2.33 47
PH-Flow [101]	32.8	3.19 28	8.87 35	2.71 35	2.84 41	9.33 35	2.37 45	2.85 14	7.20 15	2.36 45	2.92 42	15.4 51	1.51 41	2.63 15	3.42 12	2.04 18	3.03 26	8.52 17	2.49 19	2.69 74	3.60 47	3.13 100	1.25 8	2.53 9	1.34 16
Classic+CPF [83]	33.4	3.14 21	8.60 27	2.63 29	3.03 57	10.6 55	2.33 43	3.66 34	9.58 33	2.20 39	2.61 19	14.1 33	1.34 20	2.68 18	3.53 19	2.21 31	2.85 18	7.95 9	2.38 14	2.44 59	3.49 33	2.90 91	1.67 26	3.40 24	2.43 50
Sparse-NonSparse [56]	35.1	3.14 21	8.75 32	2.76 42	3.02 55	10.6 55	2.43 50	3.45 29	8.96 28	2.36 45	2.66 24	13.7 28	1.42 28	2.85 31	3.75 36	2.33 37	3.28 37	9.40 26	2.73 35	2.42 56	3.31 14	2.69 75	1.47 17	3.07 19	1.66 23
TC-Flow [46]	36.4	2.91 9	8.00 14	2.34 12	2.18 8	8.77 27	1.52 1	3.84 42	10.7 47	1.49 2	3.13 50	16.6 69	1.46 36	2.78 25	3.73 35	1.96 11	3.08 29	11.4 46	2.66 28	1.94 22	3.43 23	3.20 105	3.06 69	7.04 71	4.08 93
3DFlow [135]	36.9	3.44 44	8.63 29	2.46 16	2.43 19	8.59 26	1.75 12	3.71 36	9.93 39	1.64 10	1.61 3	4.58 1	1.23 13	2.86 33	3.72 33	2.16 28	4.52 86	11.6 52	4.20 93	3.16 99	4.02 77	4.44 125	1.13 5	2.14 2	0.89 4
LSM [39]	37.5	3.12 19	8.62 28	2.75 41	3.00 53	10.5 54	2.44 52	3.43 27	8.85 26	2.35 44	2.66 24	13.6 26	1.44 31	2.82 27	3.68 29	2.36 40	3.38 41	9.41 27	2.81 42	2.69 74	3.52 36	2.84 84	1.59 21	3.38 23	1.80 32
Ramp [62]	38.4	3.18 27	8.83 34	2.73 38	2.89 46	10.1 45	2.44 52	3.27 24	8.43 24	2.38 48	2.74 31	14.2 34	1.46 36	2.82 27	3.69 32	2.29 35	3.37 40	9.31 24	2.93 48	2.62 70	3.38 21	3.19 104	1.54 20	3.21 20	2.24 42
Correlation Flow [75]	38.5	3.38 41	8.40 23	2.64 30	2.23 10	7.54 14	1.56 3	5.14 69	13.1 68	1.60 7	2.09 8	8.15 6	1.35 22	3.12 53	4.09 58	2.34 38	4.01 72	11.5 50	4.00 87	2.59 66	3.61 49	3.00 96	1.49 19	3.04 16	1.42 19
SVFilterOh [111]	38.5	3.63 50	8.82 33	2.86 46	2.60 29	8.06 18	2.05 27	2.95 15	7.09 14	2.03 28	2.80 35	13.8 31	1.41 27	2.63 15	3.42 12	1.75 7	3.49 50	10.3 35	3.23 63	3.63 113	5.75 132	4.47 126	1.09 4	2.45 8	0.92 6
ProbFlowFields [128]	39.2	4.18 70	12.4 85	3.40 76	2.43 19	8.16 20	2.19 33	3.65 33	9.72 35	2.86 69	2.22 10	9.42 9	1.42 28	3.01 45	3.96 48	2.36 40	2.73 14	10.9 38	2.51 20	1.89 20	3.39 22	1.82 30	2.59 56	6.21 63	2.75 59
PMF [73]	39.4	3.61 48	9.07 38	2.62 27	2.40 14	8.05 17	1.83 16	2.61 9	6.27 10	1.65 11	3.35 64	15.4 51	1.58 48	2.54 8	3.27 8	1.71 6	3.59 54	11.1 43	3.46 69	4.07 123	6.18 138	4.02 121	1.06 3	2.38 6	1.25 14
COFM [59]	39.5	3.17 25	9.90 55	2.46 16	2.41 16	8.34 23	1.92 21	3.77 39	10.5 42	2.54 57	2.71 29	14.9 44	1.19 9	3.08 50	3.92 46	3.25 92	3.83 64	10.9 38	3.15 55	2.20 46	3.35 16	2.91 93	1.62 25	2.56 10	2.09 38
FMOF [94]	40.8	3.12 19	8.23 21	2.73 38	3.25 68	10.7 62	2.52 59	3.01 18	7.61 18	2.20 39	2.56 16	13.4 22	1.33 17	2.75 23	3.61 23	2.24 32	3.66 56	8.50 16	2.78 40	2.62 70	3.84 64	3.27 108	2.66 61	5.69 54	1.95 36
JOF [140]	41.0	3.08 17	8.56 26	2.51 19	3.27 72	10.2 48	2.81 82	3.02 19	7.55 17	2.42 52	2.64 22	14.2 34	1.34 20	2.62 13	3.42 12	2.08 20	3.26 35	8.96 21	2.56 21	3.12 98	4.26 86	4.09 123	2.11 45	4.58 42	2.18 40
OAR-Flow [125]	41.7	3.37 39	9.87 54	2.67 33	4.22 93	12.8 90	2.87 84	4.95 65	13.4 71	2.66 61	3.23 56	16.4 68	1.37 23	2.83 29	3.82 39	1.97 14	2.49 6	10.9 38	1.87 3	1.52 3	2.82 1	1.86 35	1.85 33	4.35 37	1.68 25
Classic+NL [31]	42.1	3.20 30	8.72 31	2.81 45	3.02 55	10.6 55	2.44 52	3.46 30	8.84 25	2.38 48	2.78 34	14.3 37	1.46 36	2.83 29	3.68 29	2.31 36	3.40 43	9.09 23	2.76 39	2.87 83	3.82 63	2.86 88	1.67 26	3.53 28	2.26 46
TV-L1-MCT [64]	43.2	3.16 24	8.48 25	2.71 35	3.28 73	10.8 66	2.60 70	3.95 45	10.5 42	2.38 48	2.69 27	13.9 32	1.45 35	2.94 41	3.79 37	2.63 69	3.50 51	9.75 31	3.06 52	2.08 32	3.35 16	2.29 60	1.95 36	3.89 31	2.71 58
PWC-Net_ROB [147]	45.8	4.86 102	12.4 85	3.56 85	3.14 59	10.3 51	2.60 70	4.38 55	11.6 55	3.18 76	2.56 16	10.6 12	1.52 43	3.25 69	4.18 62	2.46 51	3.10 31	10.6 37	2.75 37	1.44 2	3.56 42	1.01 3	1.60 22	3.41 25	1.14 10
IIOF-NLDP [131]	46.6	3.65 51	9.81 53	2.56 23	2.79 36	9.36 37	2.00 23	4.28 53	11.3 53	1.69 13	2.02 6	7.52 5	1.38 26	3.36 76	4.52 91	2.40 47	3.82 62	11.2 44	3.67 78	2.07 30	3.79 60	1.88 39	2.91 66	5.30 53	4.17 94
SimpleFlow [49]	47.5	3.35 36	9.20 41	2.98 53	3.18 63	10.7 62	2.71 76	5.06 67	12.6 66	2.70 63	2.95 44	15.1 47	1.58 48	2.91 39	3.79 37	2.47 52	3.59 54	9.49 28	2.99 50	2.39 54	3.46 27	2.24 59	1.60 22	3.56 29	1.57 22
CostFilter [40]	47.7	3.84 55	9.64 49	3.06 55	2.55 28	8.09 19	2.03 25	2.69 12	6.47 11	1.88 20	3.66 75	16.8 71	1.88 65	2.62 13	3.34 10	1.99 16	4.05 73	11.0 42	3.65 77	4.16 125	7.18 145	4.66 128	1.16 6	3.36 22	0.87 3
2DHMM-SAS [92]	49.9	3.19 28	8.89 36	2.71 35	3.20 66	11.5 76	2.38 46	5.19 70	12.2 62	2.73 65	2.92 42	15.2 48	1.53 45	2.79 26	3.65 27	2.27 34	3.45 47	9.34 25	2.78 40	2.66 73	3.56 42	3.07 97	2.34 50	5.12 51	2.97 66
S2D-Matching [84]	51.0	3.36 37	9.66 50	2.86 46	3.19 65	11.1 70	2.46 55	4.86 64	12.9 67	2.47 55	2.67 26	13.2 19	1.44 31	2.87 34	3.72 33	2.38 43	3.45 47	9.76 32	2.95 49	3.05 92	3.79 60	3.30 110	1.95 36	4.16 35	3.00 67
FlowFields+ [130]	51.0	4.57 88	13.7 99	3.35 68	2.94 51	10.1 45	2.58 66	4.05 48	10.6 44	3.26 79	2.90 41	13.2 19	1.81 63	3.18 57	4.20 66	2.54 58	2.68 13	11.4 46	2.40 16	1.84 14	3.62 50	1.77 26	2.48 52	5.86 56	2.77 60
MLDP_OF [89]	51.1	4.13 66	10.3 63	3.60 86	2.34 12	7.70 15	1.88 20	4.23 52	10.9 50	1.87 19	2.74 31	14.6 41	1.37 23	3.10 51	3.91 45	2.48 56	3.40 43	9.00 22	3.79 82	3.46 106	4.20 83	5.55 132	2.31 48	4.64 44	1.98 37
MDP-Flow [26]	51.8	3.48 46	9.46 46	3.10 57	2.45 22	7.36 12	2.41 47	3.21 22	8.31 22	2.78 67	3.18 54	17.8 79	1.70 58	3.03 47	3.87 40	2.60 65	3.43 45	12.6 65	2.81 42	2.19 43	3.88 68	1.60 20	4.13 90	9.96 98	3.86 88
AggregFlow [97]	51.9	4.25 76	11.9 82	3.26 60	4.46 99	13.7 101	3.43 95	4.76 62	12.4 63	3.93 95	3.28 59	15.6 56	1.68 55	2.89 37	3.89 42	2.08 20	2.32 3	7.75 8	2.14 7	2.06 28	3.77 58	1.48 15	2.07 43	4.11 33	2.36 48
IROF-TV [53]	52.2	3.40 42	9.29 43	2.95 52	2.99 52	11.1 70	2.53 60	3.81 40	9.81 36	2.44 53	3.25 58	16.9 73	1.78 62	3.27 73	4.10 59	2.93 83	4.47 83	16.0 101	3.53 71	1.70 6	3.21 7	1.12 6	1.91 35	4.75 46	2.19 41
CombBMOF [113]	53.7	3.94 59	10.6 67	2.74 40	2.80 38	8.55 25	2.16 31	3.10 21	7.99 21	1.76 15	2.99 45	13.4 22	1.95 69	3.04 48	3.89 42	2.49 57	5.64 111	12.3 59	6.74 126	3.54 109	5.16 118	2.81 81	1.85 33	4.60 43	1.10 8
S2F-IF [123]	53.7	4.51 86	13.6 98	3.31 64	2.90 47	10.4 52	2.48 58	4.07 50	10.8 49	3.15 74	3.31 60	15.7 59	1.90 66	3.17 55	4.19 64	2.55 61	2.81 17	11.6 52	2.60 26	1.86 17	3.67 53	1.87 36	2.11 45	4.64 44	2.54 55
WRT [150]	53.9	3.74 53	9.34 44	2.48 18	3.37 78	10.2 48	2.58 66	6.80 94	15.3 86	2.24 41	1.58 1	5.01 2	1.09 6	2.89 37	3.68 29	2.35 39	5.52 109	12.0 56	4.21 95	2.30 50	3.85 65	2.34 62	3.20 72	4.91 47	4.21 95
FlowFields [110]	56.2	4.57 88	13.7 99	3.38 71	3.01 54	10.6 55	2.59 68	4.19 51	11.1 51	3.30 80	3.17 53	15.0 45	1.96 70	3.21 64	4.24 73	2.61 68	2.91 22	12.4 61	2.66 28	1.84 14	3.46 27	1.84 33	2.50 53	6.15 61	2.79 61
Sparse Occlusion [54]	57.6	3.62 49	9.12 39	2.90 48	2.92 49	9.08 33	2.56 62	4.49 59	11.8 60	2.11 34	3.14 51	15.8 60	1.57 47	3.26 70	4.22 69	2.36 40	3.52 52	10.9 38	2.66 28	5.10 141	6.32 139	3.15 102	2.02 38	4.92 48	1.71 26
NL-TV-NCC [25]	57.9	3.89 57	9.16 40	2.98 53	2.87 45	9.69 40	1.99 22	4.44 58	11.6 55	1.76 15	2.64 22	11.8 13	1.48 40	3.49 88	4.60 98	2.47 52	4.67 93	13.5 73	4.26 99	2.83 81	4.57 99	2.84 84	2.62 59	6.00 60	2.25 44
EPPM w/o HM [88]	58.4	4.25 76	11.1 71	3.13 58	2.36 13	8.35 24	1.76 13	3.72 37	10.2 40	1.81 17	3.24 57	14.5 40	1.94 68	3.16 54	3.94 47	2.82 78	4.78 97	12.9 69	4.32 100	3.64 115	4.54 97	5.73 134	1.76 29	4.11 33	1.94 34
OFH [38]	58.5	3.90 58	9.77 52	3.62 89	2.84 41	11.0 69	2.04 26	5.52 77	14.4 79	1.89 21	3.52 67	20.5 100	1.60 51	3.18 57	4.06 56	2.82 78	3.86 65	14.1 82	3.59 73	1.77 10	3.62 50	1.81 29	2.64 60	7.08 74	2.15 39
PGM-C [120]	58.8	4.62 93	14.0 104	3.39 73	3.29 75	12.3 83	2.70 75	4.39 57	11.7 58	3.43 83	4.00 83	19.8 89	2.15 75	3.19 59	4.23 70	2.54 58	2.79 15	11.9 55	2.45 18	1.83 12	3.21 7	1.83 31	2.31 48	5.87 57	1.82 33
Occlusion-TV-L1 [63]	60.6	3.59 47	9.61 47	2.64 30	2.93 50	10.6 55	2.41 47	6.16 85	15.2 84	2.70 63	3.32 62	17.0 74	1.68 55	3.38 78	4.44 84	2.82 78	3.10 31	13.2 72	2.68 31	2.17 39	3.52 36	1.46 13	4.63 104	11.1 113	3.53 77
Complementary OF [21]	61.2	4.44 82	11.2 74	4.04 98	2.51 26	9.77 42	1.74 11	3.93 43	10.6 44	2.04 30	3.87 79	18.8 81	2.19 79	3.17 55	4.00 51	2.92 82	4.64 91	13.8 79	3.64 76	2.17 39	3.36 18	2.51 68	3.08 70	7.04 71	3.65 81
Adaptive [20]	63.0	3.29 32	9.43 45	2.28 9	3.10 58	11.4 73	2.46 55	6.58 90	15.7 91	2.52 56	3.14 51	15.6 56	1.56 46	3.67 99	4.46 86	3.48 102	3.32 39	13.0 71	2.38 14	2.76 79	4.39 91	1.93 43	3.58 78	8.18 84	2.88 63
Aniso-Texture [82]	64.1	2.96 10	7.72 8	2.54 20	2.48 24	8.26 22	2.24 37	6.48 88	15.9 96	2.63 59	1.96 5	10.1 11	0.98 2	3.26 70	4.21 67	2.60 65	5.74 114	16.9 110	5.61 116	4.47 133	5.88 135	3.33 111	3.51 77	7.12 75	3.68 83
ACK-Prior [27]	64.3	4.19 72	9.27 42	3.60 86	2.40 14	8.21 21	1.65 7	3.40 26	8.96 28	1.84 18	2.87 39	14.4 39	1.44 31	3.36 76	4.15 60	3.07 87	6.35 122	16.1 103	4.90 110	4.21 128	4.80 104	6.03 136	3.29 74	5.99 59	2.82 62
CPM-Flow [116]	65.1	4.63 94	14.1 107	3.39 73	3.33 76	12.5 87	2.73 77	4.37 54	11.7 58	3.43 83	4.00 83	19.9 92	2.14 74	3.19 59	4.23 70	2.54 58	3.08 29	12.0 56	2.88 46	1.87 18	3.44 24	1.84 33	2.91 66	7.48 81	2.91 65
DPOF [18]	65.3	4.67 97	12.6 91	3.30 62	3.57 83	10.6 55	3.12 91	3.09 20	7.50 16	2.32 43	3.06 48	14.8 42	1.82 64	3.21 64	4.18 62	2.79 77	4.47 83	12.5 63	3.33 64	4.09 124	3.92 71	6.96 138	2.09 44	4.39 38	1.74 28
EpicFlow [102]	65.5	4.61 92	14.0 104	3.39 73	3.33 76	12.5 87	2.74 78	5.37 73	14.8 82	3.46 86	3.94 82	19.2 85	2.13 73	3.20 61	4.23 70	2.58 64	2.87 20	12.2 58	2.64 27	1.83 12	3.28 13	1.83 31	3.21 73	7.12 75	3.61 78
DeepFlow2 [108]	67.0	4.04 63	11.2 74	3.38 71	3.80 86	12.4 86	2.86 83	5.12 68	13.4 71	3.00 70	4.17 89	20.1 94	2.18 78	2.96 42	3.97 49	2.08 20	3.06 28	12.6 65	2.69 32	2.17 39	3.24 10	2.71 76	4.74 106	10.4 106	4.38 102
ROF-ND [107]	67.5	4.12 64	10.0 57	3.37 70	2.78 35	8.82 29	2.12 30	4.61 61	11.9 61	2.09 33	2.23 11	6.56 4	1.69 57	3.60 95	4.75 108	2.85 81	4.92 100	13.6 76	3.75 80	4.59 135	5.18 119	4.10 124	2.67 62	5.19 52	3.46 76
TCOF [69]	67.7	4.17 69	10.4 65	3.71 92	3.17 60	10.7 62	2.59 68	6.58 90	15.7 91	3.82 93	3.69 77	16.1 65	2.37 88	3.78 103	4.95 120	2.47 52	2.59 8	8.47 15	2.58 23	3.66 117	4.83 105	2.67 74	1.83 32	4.20 36	1.46 20
HBM-GC [105]	68.9	5.25 105	10.5 66	4.34 105	3.17 60	8.78 28	2.94 87	4.38 55	10.6 44	2.68 62	3.59 71	12.8 16	2.47 91	2.96 42	3.64 26	2.64 70	3.96 71	8.26 11	3.56 72	4.40 131	5.92 136	3.62 115	2.55 55	6.34 64	3.29 71
RFlow [90]	69.2	3.82 54	10.0 57	3.44 79	2.61 30	9.73 41	2.02 24	5.66 79	14.5 80	2.05 31	3.93 81	23.1 114	1.90 66	3.24 66	4.19 64	2.66 72	4.12 76	15.2 97	3.34 66	2.61 68	3.56 42	2.65 73	4.48 99	10.5 109	3.93 92
Steered-L1 [118]	70.7	3.30 34	8.44 24	2.91 49	1.89 1	7.14 8	1.60 6	3.61 32	9.91 38	1.89 21	3.45 65	19.4 88	1.64 53	3.42 81	4.30 76	3.39 95	5.18 104	14.5 85	4.37 103	5.09 140	5.05 114	10.1 142	5.56 114	10.2 104	6.24 120
SRR-TVOF-NL [91]	71.5	4.47 84	10.9 69	3.32 66	4.04 90	13.2 96	2.90 85	4.81 63	12.5 64	3.15 74	3.33 63	15.3 49	1.61 52	3.24 66	4.03 55	2.70 74	3.94 69	11.8 54	3.33 64	4.16 125	5.21 122	3.44 114	2.06 42	3.48 26	2.42 49
DMF_ROB [139]	71.5	4.37 79	12.3 84	3.62 89	3.46 81	12.9 92	2.60 70	5.98 82	15.8 93	3.23 78	4.05 85	19.8 89	2.15 75	3.10 51	4.06 56	2.57 63	3.79 60	14.3 83	3.13 54	1.88 19	3.12 5	1.99 47	4.34 93	10.0 99	3.87 89
ComplOF-FED-GPU [35]	71.9	4.28 78	11.3 76	3.70 91	3.25 68	13.0 93	2.16 31	4.06 49	11.2 52	1.95 24	3.91 80	19.2 85	2.01 71	3.20 61	4.15 60	2.64 70	4.61 89	16.1 103	3.90 84	2.98 89	3.77 58	3.69 116	2.85 64	7.44 80	2.53 54
FF++_ROB [145]	73.0	4.84 101	14.8 114	3.46 80	3.18 63	11.4 73	2.69 74	5.30 72	14.1 76	3.73 92	3.31 60	14.2 34	2.20 80	3.26 70	4.29 75	2.72 75	4.58 88	12.7 67	3.70 79	1.91 21	3.46 27	2.19 58	3.65 80	7.31 77	5.97 117
TF+OM [100]	75.0	3.97 60	10.2 60	2.94 51	2.91 48	9.12 34	2.57 65	5.22 71	11.5 54	6.92 107	3.59 71	16.1 65	2.28 85	3.20 61	3.97 49	3.11 88	4.70 95	14.5 85	4.32 100	3.06 94	4.84 107	2.71 76	3.93 85	8.79 89	4.32 100
Aniso. Huber-L1 [22]	75.5	3.71 52	10.1 59	3.08 56	4.36 98	13.0 93	3.77 99	6.92 95	15.3 86	3.60 89	3.54 68	15.9 63	2.04 72	3.38 78	4.45 85	2.47 52	3.88 66	12.9 69	2.74 36	3.37 103	4.36 89	2.85 87	3.16 71	7.52 82	2.90 64
DeepFlow [86]	76.8	4.49 85	11.7 79	4.14 100	4.26 94	12.8 90	3.36 93	5.96 81	14.2 78	5.10 99	4.89 103	23.1 114	2.67 94	2.98 44	4.00 51	2.11 24	3.26 35	13.5 73	2.84 45	2.09 34	3.10 3	2.77 78	5.83 116	11.4 115	5.45 114
Classic++ [32]	77.7	3.37 39	9.67 51	2.91 49	3.28 73	12.1 80	2.61 73	5.46 76	14.1 76	3.00 70	3.63 73	20.2 97	1.70 58	3.24 66	4.34 78	2.60 65	4.65 92	16.0 101	3.60 74	3.09 95	3.94 74	3.28 109	4.64 105	10.4 106	3.71 84
TV-L1-improved [17]	78.6	3.36 37	9.63 48	2.62 27	2.82 39	10.7 62	2.23 35	6.50 89	15.8 93	2.73 65	3.80 78	21.3 105	1.76 61	3.34 75	4.38 82	2.39 44	5.97 116	18.1 118	5.67 117	3.57 111	4.92 112	3.43 113	4.01 88	9.84 97	3.44 75
LocallyOriented [52]	80.7	4.54 87	12.8 93	3.27 61	4.73 104	14.8 108	3.73 98	7.77 102	18.3 111	3.44 85	3.56 69	15.6 56	2.22 81	3.46 85	4.47 87	2.69 73	3.15 33	10.2 34	3.19 58	2.61 68	4.20 83	2.52 69	4.39 96	8.52 86	5.23 110
SIOF [67]	80.9	4.23 74	10.2 60	3.31 64	3.97 88	14.5 106	2.97 88	7.81 104	16.4 99	7.48 109	4.82 99	20.1 94	2.96 97	3.54 91	4.49 88	3.12 89	4.31 78	13.5 73	4.13 91	2.36 53	3.59 46	1.68 23	3.46 76	7.39 78	3.37 73
LiteFlowNet [142]	82.2	6.29 112	16.5 120	4.45 107	3.68 84	10.8 66	3.13 92	5.43 74	13.7 74	3.60 89	3.57 70	12.8 16	2.25 84	3.85 110	4.78 110	3.61 105	4.37 80	12.5 63	3.63 75	2.55 63	4.51 96	1.52 16	4.05 89	7.05 73	5.16 106
Brox et al. [5]	83.8	4.44 82	12.4 85	4.22 103	3.72 85	13.5 100	3.06 89	4.97 66	13.3 70	3.11 72	4.58 95	22.0 108	2.37 88	3.79 105	4.60 98	4.33 127	3.91 68	17.0 111	3.45 68	2.22 47	3.79 60	1.19 7	4.62 103	10.0 99	3.38 74
TriangleFlow [30]	84.1	4.12 64	10.6 67	3.47 81	3.47 82	13.1 95	2.41 47	6.00 83	15.2 84	2.17 36	2.99 45	16.0 64	1.58 48	4.46 131	5.79 136	4.15 123	5.42 108	13.9 81	5.24 111	3.10 97	5.47 128	2.90 91	3.02 68	6.82 68	3.64 80
CRTflow [80]	84.5	4.18 70	11.8 81	3.20 59	3.22 67	10.8 66	2.43 50	6.20 86	15.5 89	2.63 59	4.21 90	22.0 108	2.24 82	3.32 74	4.34 78	2.44 50	7.43 129	19.3 124	8.15 132	2.55 63	4.09 79	2.59 72	4.60 102	11.2 114	4.45 103
OFRF [134]	85.6	4.77 100	11.6 77	4.03 97	8.72 124	15.3 113	8.51 127	8.49 115	16.7 101	7.32 108	4.55 94	15.3 49	3.16 104	2.92 40	3.87 40	2.13 27	3.76 59	9.69 29	3.22 60	2.98 89	4.50 95	4.04 122	4.59 101	5.76 55	8.61 128
BriefMatch [124]	86.5	3.44 44	9.01 37	2.77 43	2.85 44	9.93 43	2.23 35	2.97 16	7.65 20	1.94 23	3.64 74	20.1 94	1.75 60	4.10 121	4.90 118	5.82 137	7.95 131	17.8 114	8.08 131	4.73 137	5.20 120	12.2 144	7.88 133	12.0 119	13.7 139
Rannacher [23]	86.7	4.13 66	11.0 70	3.61 88	3.39 79	12.3 83	2.80 81	7.26 97	17.4 107	3.59 88	4.40 92	23.1 114	2.24 82	3.43 83	4.54 94	2.56 62	5.41 107	18.5 119	4.23 96	2.92 86	3.91 70	2.82 82	3.45 75	9.14 90	3.27 70
F-TV-L1 [15]	87.6	5.44 108	12.5 90	5.69 117	5.46 109	15.0 111	4.03 102	7.48 99	16.3 98	3.42 82	5.08 106	23.3 117	2.81 96	3.42 81	4.34 78	3.03 85	4.05 73	15.1 94	3.18 57	2.43 57	3.92 71	1.87 36	3.90 84	9.35 94	2.61 57
TriFlow [95]	88.7	4.73 99	12.4 85	3.49 83	4.03 89	12.5 87	3.70 97	8.18 112	17.2 105	10.4 119	3.50 66	15.4 51	2.32 87	3.43 83	4.21 67	3.42 96	3.90 67	12.3 59	3.76 81	7.86 146	5.72 131	16.2 146	2.80 63	5.89 58	2.50 52
SuperFlow [81]	88.8	4.16 68	11.1 71	3.32 66	4.80 105	12.2 81	4.68 108	7.80 103	16.0 97	10.6 120	5.16 110	22.4 112	3.24 106	3.39 80	4.24 73	3.71 109	3.44 46	13.7 78	2.91 47	3.19 100	4.62 101	1.87 36	4.74 106	10.6 110	4.24 97
Local-TV-L1 [65]	89.2	5.33 106	12.6 91	5.19 112	6.90 117	15.7 116	6.22 116	10.0 120	18.2 110	8.89 112	5.81 116	24.7 122	3.70 114	3.05 49	4.00 51	2.39 44	4.05 73	14.6 87	3.09 53	1.95 23	3.11 4	2.15 54	5.85 117	10.8 111	7.34 123
DF-Auto [115]	89.2	5.04 104	13.7 99	3.30 62	6.51 114	14.1 105	6.09 115	8.14 108	16.5 100	10.2 118	5.06 105	21.3 105	3.10 103	3.74 101	4.91 119	3.25 92	2.67 12	11.4 46	2.14 7	3.36 102	5.23 124	1.45 12	4.45 98	9.18 91	4.28 99
ContinualFlow_ROB [152]	89.2	7.36 121	17.7 125	5.46 114	5.94 112	12.2 81	5.98 114	8.16 111	18.3 111	7.89 110	5.11 107	19.3 87	3.18 105	4.15 124	5.04 124	3.68 107	5.65 112	15.1 94	6.17 123	1.72 7	3.34 15	1.11 5	2.34 50	4.48 39	2.25 44
CLG-TV [48]	89.8	4.00 61	10.3 63	3.40 76	4.33 97	12.3 83	4.08 103	6.78 92	15.5 89	3.64 91	4.07 86	17.7 78	2.39 90	3.79 105	4.86 113	3.23 91	4.48 85	16.5 108	3.80 83	3.55 110	4.65 102	2.89 90	4.00 87	10.1 102	3.18 69
CBF [12]	90.5	3.88 56	10.2 60	3.50 84	4.60 101	11.3 72	5.06 109	5.43 74	13.1 68	3.39 81	4.09 87	21.2 104	2.16 77	3.80 108	4.72 107	3.52 103	4.33 79	14.4 84	3.01 51	4.97 138	5.51 129	4.93 130	3.99 86	9.27 93	3.91 91
Bartels [41]	93.0	4.43 80	11.1 71	4.17 102	2.83 40	8.84 30	2.56 62	4.54 60	12.5 64	2.80 68	4.87 100	22.1 110	3.05 101	3.58 94	4.35 81	4.15 123	5.55 110	17.5 112	5.78 118	3.74 118	5.02 113	5.98 135	5.21 113	11.9 118	5.20 109
p-harmonic [29]	93.7	4.64 95	13.0 94	4.43 106	3.41 80	11.9 77	2.93 86	7.60 100	18.1 109	3.96 96	4.65 96	21.0 102	2.97 99	3.46 85	4.33 77	3.34 94	4.75 96	17.5 112	4.60 107	3.05 92	4.17 81	2.15 54	5.09 112	10.9 112	3.77 86
Fusion [6]	93.7	4.43 80	13.7 99	4.08 99	2.47 23	8.91 31	2.24 37	3.70 35	9.68 34	3.12 73	3.68 76	19.8 89	2.54 93	4.26 128	5.16 129	4.31 126	6.32 119	16.8 109	6.15 122	4.55 134	5.78 133	3.10 98	7.12 127	13.6 128	7.86 127
CNN-flow-warp+ref [117]	94.6	4.93 103	14.5 111	4.29 104	4.18 92	11.9 77	4.24 105	8.23 113	19.7 119	6.35 106	5.13 108	24.4 121	2.96 97	3.55 92	4.40 83	3.85 113	3.82 62	15.0 91	3.39 67	1.96 24	3.44 24	2.14 53	10.0 137	14.8 134	10.8 135
CompactFlow_ROB [162]	94.8	8.85 132	18.7 128	5.45 113	5.55 110	12.0 79	5.64 113	8.73 117	17.0 104	11.7 124	5.19 111	17.5 76	3.62 112	4.11 122	4.99 122	3.72 110	4.37 80	14.6 87	4.01 88	1.75 8	3.64 52	0.96 2	4.14 91	7.40 79	5.55 115
EAI-Flow [151]	95.2	7.40 122	16.3 118	6.04 120	5.29 108	15.0 111	4.27 106	6.28 87	15.0 83	5.22 102	4.99 104	19.1 84	3.49 109	3.55 92	4.55 95	3.01 84	4.69 94	14.8 89	4.25 98	4.16 125	4.83 105	2.55 71	2.61 58	6.99 70	2.48 51
Dynamic MRF [7]	95.6	4.58 90	12.4 85	4.14 100	3.25 68	13.9 102	2.27 41	6.02 84	16.8 102	2.36 45	4.39 91	22.6 113	2.51 92	3.61 96	4.55 95	3.46 98	6.81 124	22.2 134	6.78 128	2.41 55	3.48 31	3.69 116	9.26 135	17.8 138	10.2 132
FlowNetS+ft+v [112]	96.9	4.22 73	12.1 83	3.48 82	4.50 100	13.4 98	3.85 100	8.29 114	18.4 113	6.20 105	4.87 100	21.6 107	3.01 100	3.93 115	5.04 124	3.47 101	3.71 57	15.3 98	3.21 59	3.32 101	5.12 116	3.87 118	3.76 82	9.44 95	3.74 85
SegOF [10]	97.0	5.85 110	13.5 97	3.98 96	7.40 118	14.9 109	8.13 125	8.55 116	17.3 106	9.01 113	6.50 121	18.1 80	5.14 123	3.90 114	4.53 92	4.81 131	6.57 123	21.7 132	6.81 129	1.65 5	3.49 33	1.08 4	3.71 81	9.23 92	3.63 79
ResPWCR_ROB [144]	97.1	7.29 120	16.3 118	6.15 121	4.28 95	11.4 73	3.95 101	5.85 80	13.6 73	5.20 101	4.75 98	17.5 76	3.50 110	3.80 108	4.53 92	4.12 122	4.96 103	15.0 91	4.81 109	3.52 108	5.22 123	2.40 64	3.61 79	6.77 67	4.27 98
LDOF [28]	97.4	4.60 91	13.0 94	3.77 93	4.67 102	15.5 115	3.67 96	5.63 78	14.0 75	4.21 97	5.80 115	27.1 131	3.43 108	3.52 90	4.50 90	3.46 98	4.84 99	17.8 114	4.04 89	2.46 61	4.14 80	3.25 106	4.85 109	12.0 119	3.78 87
Second-order prior [8]	98.3	4.03 62	11.6 77	3.35 68	3.88 87	14.0 104	3.08 90	7.21 96	17.6 108	3.57 87	4.14 88	19.9 92	2.31 86	3.66 98	4.86 113	2.73 76	7.32 127	21.2 130	6.76 127	4.02 121	4.58 100	4.01 120	4.27 92	10.4 106	5.12 105
WOLF_ROB [148]	99.2	5.79 109	16.6 121	4.49 108	7.62 119	21.2 133	5.10 111	9.70 119	21.0 123	5.66 104	5.32 112	19.0 82	3.78 115	3.61 96	4.49 88	3.54 104	4.63 90	13.6 76	4.34 102	2.30 50	3.89 69	2.16 57	4.37 94	7.52 82	6.03 118
AugFNG_ROB [143]	100.1	8.29 128	19.2 130	5.66 116	7.67 120	16.0 118	8.01 124	10.1 121	20.5 121	11.0 122	5.13 108	15.5 54	3.64 113	4.11 122	4.97 121	3.93 114	4.45 82	15.1 94	4.20 93	2.27 48	4.37 90	1.23 10	3.80 83	6.87 69	4.34 101
FlowNet2 [122]	103.4	8.58 131	18.6 126	6.31 122	9.39 129	17.6 122	9.09 130	8.06 107	15.8 93	9.81 116	5.61 114	16.2 67	4.12 117	4.04 118	4.88 115	3.79 111	4.92 100	16.2 105	4.50 104	4.28 129	6.73 141	2.84 84	2.05 41	4.54 41	1.41 18
StereoFlow [44]	103.8	17.1 148	28.1 148	17.9 147	18.7 145	29.7 146	16.5 140	20.1 145	30.9 144	17.5 140	21.2 145	38.3 147	17.9 143	4.60 132	5.05 126	5.52 133	2.38 4	11.5 50	1.77 2	1.25 1	2.92 2	0.71 1	4.49 100	10.3 105	4.23 96
EPMNet [133]	104.5	8.37 130	18.8 129	6.44 124	9.35 128	18.4 124	8.78 129	7.42 98	14.7 81	8.61 111	5.98 118	20.4 99	4.27 119	4.04 118	4.88 115	3.79 111	4.92 100	16.2 105	4.50 104	3.65 116	6.14 137	2.42 66	2.60 57	6.15 61	1.74 28
Ad-TV-NDC [36]	105.3	8.36 129	14.0 104	11.1 140	12.9 136	19.9 130	12.8 136	14.4 132	23.1 125	12.1 126	7.40 124	20.6 101	6.33 124	3.47 87	4.66 103	2.39 44	3.95 70	13.8 79	3.51 70	2.48 62	3.75 56	2.05 49	9.75 136	12.1 121	16.7 143
LFNet_ROB [149]	106.2	7.69 123	19.8 131	5.72 118	4.70 103	13.3 97	4.13 104	8.15 110	20.0 120	5.42 103	4.73 97	17.1 75	3.42 107	4.15 124	5.10 128	4.05 118	5.28 106	18.0 116	4.64 108	2.87 83	4.74 103	1.98 46	4.92 110	11.4 115	5.01 104
Shiralkar [42]	107.8	4.64 95	14.1 107	3.94 94	4.29 96	16.9 120	2.77 79	7.75 101	18.8 115	3.19 77	5.54 113	25.0 124	3.56 111	3.51 89	4.55 95	3.04 86	7.41 128	20.1 128	6.41 124	3.76 119	4.35 88	5.28 131	6.56 123	14.4 133	5.30 112
Learning Flow [11]	108.4	4.23 74	11.7 79	3.41 78	4.16 91	15.3 113	3.42 94	6.78 92	16.9 103	3.83 94	6.41 120	25.3 125	4.25 118	4.66 134	6.01 141	4.00 117	6.33 121	20.7 129	5.30 112	3.09 95	4.84 107	2.91 93	7.08 126	15.0 135	5.27 111
StereoOF-V1MT [119]	109.0	4.71 98	14.1 107	3.95 95	5.10 107	20.3 132	2.78 80	7.98 106	20.7 122	2.57 58	4.48 93	21.1 103	2.79 95	4.20 127	5.29 131	4.10 120	6.85 126	22.3 135	6.42 125	2.45 60	4.17 81	3.15 102	10.5 138	18.4 141	10.5 133
IAOF2 [51]	110.1	5.38 107	13.7 99	4.50 109	5.95 113	14.6 107	5.61 112	8.80 118	18.8 115	9.40 114	12.2 134	23.8 120	13.1 138	3.86 111	4.89 117	3.12 89	5.21 105	14.9 90	4.54 106	4.33 130	5.15 117	3.93 119	4.39 96	8.57 87	3.87 89
TVL1_ROB [138]	111.4	11.3 137	19.8 131	13.0 142	12.9 136	19.6 129	13.7 138	17.4 139	27.8 138	18.0 141	12.6 136	28.9 133	11.8 136	3.71 100	4.78 110	3.46 98	4.21 77	18.0 116	3.99 86	1.79 11	3.54 40	1.21 8	7.58 131	13.9 131	8.92 130
Modified CLG [34]	112.2	7.17 119	17.1 124	6.47 125	6.85 116	14.9 109	7.48 120	14.0 128	24.8 129	15.7 136	8.35 127	27.3 132	6.36 125	3.96 116	4.99 122	4.08 119	4.54 87	19.3 124	4.15 92	2.33 52	3.86 67	2.40 64	6.00 118	13.8 130	5.40 113
GraphCuts [14]	112.9	6.25 111	14.3 110	5.53 115	8.60 123	20.1 131	6.61 118	7.91 105	15.4 88	10.9 121	4.88 102	19.0 82	3.05 101	3.78 103	4.71 105	3.94 115	8.74 136	16.4 107	5.39 114	4.04 122	4.87 109	4.85 129	6.35 121	12.2 122	6.05 119
Filter Flow [19]	113.3	6.48 113	14.6 112	4.96 110	5.73 111	15.7 116	5.07 110	10.1 121	18.6 114	14.3 132	9.04 129	23.3 117	7.80 129	3.98 117	4.71 105	4.21 125	5.86 115	15.0 91	5.41 115	4.98 139	6.87 142	2.78 79	4.82 108	8.66 88	3.65 81
2D-CLG [1]	113.4	10.1 134	22.6 139	7.59 130	9.84 131	16.9 120	11.1 135	16.9 138	28.2 139	18.8 144	14.1 138	31.1 137	13.1 138	3.86 111	4.62 101	4.53 128	5.98 117	21.2 130	5.97 120	1.76 9	3.14 6	1.46 13	6.29 120	12.9 127	5.81 116
SPSA-learn [13]	114.0	6.84 118	16.7 122	6.74 126	8.47 122	19.4 127	7.49 121	12.5 124	23.1 125	13.1 130	8.40 128	25.8 128	7.08 127	3.87 113	4.66 103	4.10 120	6.32 119	18.8 120	6.89 130	2.56 65	3.85 65	1.79 27	7.29 128	12.5 124	7.47 125
HBpMotionGpu [43]	115.4	6.57 115	15.0 115	5.17 111	8.29 121	18.0 123	8.29 126	14.1 129	26.5 132	13.2 131	6.12 119	25.3 125	3.94 116	3.79 105	4.62 101	3.97 116	4.80 98	15.7 99	4.11 90	4.40 131	5.20 120	2.87 89	6.28 119	11.7 117	7.31 122
IAOF [50]	116.3	6.49 114	14.6 112	6.42 123	9.22 127	18.5 125	7.94 123	16.4 137	27.4 136	13.0 129	8.22 125	22.2 111	7.73 128	3.77 102	4.76 109	3.42 96	6.84 125	18.8 120	4.23 96	3.59 112	4.46 93	2.83 83	7.51 130	10.1 102	10.6 134
GroupFlow [9]	116.9	8.00 125	18.6 126	8.09 132	11.1 134	23.7 138	10.3 133	12.6 125	25.6 130	12.8 128	5.84 117	20.3 98	4.39 120	4.69 135	5.81 137	3.67 106	9.29 137	22.4 136	10.1 139	2.11 37	3.99 75	2.29 60	5.75 115	10.0 99	7.39 124
Black & Anandan [4]	117.5	6.81 117	15.4 116	7.43 128	8.77 125	19.5 128	7.35 119	13.0 126	22.9 124	12.5 127	8.29 126	26.1 129	6.77 126	4.18 126	5.28 130	3.69 108	6.19 118	20.0 127	5.34 113	3.63 113	5.05 114	1.79 27	6.45 122	12.2 122	5.17 108
BlockOverlap [61]	120.6	6.67 116	13.1 96	5.87 119	6.62 115	13.9 102	6.53 117	10.6 123	19.5 118	10.1 117	6.97 123	24.9 123	5.13 122	4.38 129	4.61 100	6.37 140	7.47 130	15.7 99	6.05 121	6.23 142	6.41 140	13.0 145	6.92 125	9.60 96	12.2 137
Nguyen [33]	121.2	7.88 124	16.8 123	7.02 127	13.4 138	19.0 126	15.3 139	17.6 140	28.9 140	17.2 139	12.0 133	26.9 130	11.6 135	4.38 129	5.07 127	5.58 136	5.69 113	19.7 126	5.93 119	2.75 77	4.02 77	1.91 42	6.59 124	12.5 124	6.52 121
UnFlow [129]	122.1	14.6 146	25.8 144	9.09 136	9.40 130	16.8 119	9.89 132	14.2 130	26.9 133	11.2 123	10.0 130	25.4 127	8.67 131	5.43 141	5.90 138	6.72 141	8.64 134	24.0 138	9.41 137	3.51 107	4.90 110	1.37 11	4.37 94	12.6 126	3.33 72
2bit-BM-tele [98]	122.6	8.00 125	15.8 117	8.40 134	4.91 106	13.4 98	4.67 107	8.14 108	19.0 117	5.12 100	6.62 122	23.5 119	5.04 121	4.08 120	4.78 110	4.61 130	8.68 135	18.8 120	8.31 133	6.46 144	7.08 144	9.47 141	7.36 129	14.1 132	9.62 131
Horn & Schunck [3]	127.8	8.01 127	19.9 133	8.38 133	9.13 126	23.2 137	7.71 122	14.2 130	25.9 131	14.6 134	12.4 135	30.6 135	11.3 134	4.64 133	5.64 133	4.60 129	8.21 133	24.4 139	8.45 134	4.01 120	5.41 125	1.95 44	9.16 134	17.5 136	8.86 129
SILK [79]	129.2	9.34 133	20.4 134	10.5 139	10.4 132	21.9 134	10.3 133	16.0 136	27.5 137	14.5 133	10.3 131	29.0 134	8.54 130	4.81 136	5.65 134	5.56 135	9.41 138	25.4 141	8.74 135	2.79 80	3.68 54	4.62 127	10.9 139	17.8 138	12.3 138
Heeger++ [104]	130.8	11.9 140	21.8 137	8.08 131	12.5 135	29.7 146	9.42 131	14.8 133	27.1 134	9.68 115	14.3 139	31.0 136	12.7 137	4.98 138	5.74 135	4.97 132	17.5 146	34.1 147	18.4 146	2.75 77	5.44 126	2.15 54	12.3 141	18.8 142	14.8 141
TI-DOFE [24]	131.7	13.4 144	23.2 140	16.5 146	16.5 142	24.1 139	18.2 144	20.2 146	31.1 146	20.6 145	19.9 144	32.9 140	20.8 145	4.89 137	5.90 138	5.54 134	8.04 132	23.9 137	8.81 136	2.97 88	4.34 87	1.88 39	10.9 139	17.7 137	11.9 136
H+S_ROB [137]	133.1	13.0 142	27.1 147	9.66 137	13.4 138	24.8 141	13.4 137	18.7 144	30.9 144	18.3 143	25.8 148	35.7 144	26.4 147	7.08 146	8.13 146	9.10 145	14.6 144	31.3 146	16.3 144	2.17 39	4.44 92	2.11 52	15.1 145	19.9 144	14.2 140
HCIC-L [99]	135.9	15.7 147	22.0 138	10.1 138	31.5 148	26.6 144	41.0 148	14.8 133	23.1 125	16.8 138	18.4 143	34.4 142	18.2 144	5.94 142	6.35 142	6.35 139	10.6 141	19.2 123	11.4 141	18.7 148	17.8 148	19.2 147	4.93 111	8.34 85	5.16 106
SLK [47]	136.0	11.6 138	26.0 145	14.6 145	15.3 141	25.0 142	17.5 142	17.8 142	30.1 143	18.1 142	25.4 147	33.6 141	28.0 148	5.25 139	5.90 138	7.03 142	10.3 140	27.4 143	10.6 140	2.89 85	4.47 94	2.94 95	14.9 144	20.7 145	18.8 144
FFV1MT [106]	137.0	12.0 141	23.3 141	8.83 135	10.7 133	26.6 144	8.71 128	15.6 135	29.0 141	12.0 125	16.6 142	36.3 146	15.5 141	6.51 145	6.40 143	10.4 146	16.2 145	30.7 145	17.7 145	3.41 105	5.44 126	3.35 112	12.3 141	18.8 142	14.8 141
Adaptive flow [45]	138.9	13.2 143	20.8 135	14.0 144	17.1 144	22.0 135	17.9 143	18.1 143	27.1 134	22.8 147	11.8 132	31.1 137	10.5 132	6.35 144	7.13 145	6.25 138	9.87 139	21.8 133	9.44 138	12.6 147	11.4 147	20.0 148	7.75 132	13.6 128	7.73 126
PGAM+LK [55]	140.2	11.8 139	25.6 142	13.9 143	14.8 140	24.4 140	16.7 141	13.2 127	24.0 128	15.0 135	16.2 141	41.2 148	15.3 140	5.40 140	5.45 132	8.10 143	12.3 143	26.5 142	12.1 142	7.42 145	8.24 146	7.87 139	13.2 143	18.3 140	19.4 145
Periodicity [78]	141.1	11.2 136	27.0 146	7.46 129	16.6 143	29.8 148	18.2 144	25.3 148	31.2 148	24.9 148	12.7 137	35.7 144	11.1 133	31.7 148	41.4 148	25.1 148	23.8 148	41.5 148	23.8 148	2.92 86	5.62 130	6.90 137	18.6 147	33.1 148	22.3 146
FOLKI [16]	141.9	10.5 135	25.6 142	11.9 141	20.9 146	26.2 143	26.1 146	17.6 140	31.1 146	16.5 137	15.4 140	32.6 139	16.0 142	6.16 143	6.53 144	9.07 144	12.2 142	29.7 144	13.0 143	4.67 136	5.83 134	9.41 140	18.2 146	22.8 146	25.1 147
Pyramid LK [2]	144.8	13.9 145	20.9 136	21.4 148	24.1 147	23.1 136	30.2 147	20.9 147	29.5 142	21.9 146	22.2 146	34.6 143	25.0 146	18.7 147	23.1 147	20.2 147	21.2 147	24.5 140	21.0 147	6.41 143	7.02 143	10.8 143	25.6 148	31.5 147	34.5 148
AdaConv-v1 [126]	149.4	39.2 149	39.9 149	41.8 149	73.0 149	74.5 149	71.1 149	70.1 149	67.3 149	71.8 149	64.4 149	66.2 149	65.9 149	76.5 150	78.1 150	72.0 150	68.2 150	64.9 150	66.5 150	52.3 150	45.1 150	70.9 150	81.8 149	81.6 149	82.3 149
SepConv-v1 [127]	149.4	39.2 149	39.9 149	41.8 149	73.0 149	74.5 149	71.1 149	70.1 149	67.3 149	71.8 149	64.4 149	66.2 149	65.9 149	76.5 150	78.1 150	72.0 150	68.2 150	64.9 150	66.5 150	52.3 150	45.1 150	70.9 150	81.8 149	81.6 149	82.3 149
SuperSlomo [132]	149.4	39.2 149	39.9 149	41.8 149	73.0 149	74.5 149	71.1 149	70.1 149	67.3 149	71.8 149	64.4 149	66.2 149	65.9 149	76.5 150	78.1 150	72.0 150	68.2 150	64.9 150	66.5 150	52.3 150	45.1 150	70.9 150	81.8 149	81.6 149	82.3 149
CtxSyn [136]	149.4	39.2 149	39.9 149	41.8 149	73.0 149	74.5 149	71.1 149	70.1 149	67.3 149	71.8 149	64.4 149	66.2 149	65.9 149	76.5 150	78.1 150	72.0 150	68.2 150	64.9 150	66.5 150	52.3 150	45.1 150	70.9 150	81.8 149	81.6 149	82.3 149
CyclicGen [153]	149.4	39.2 149	39.9 149	41.8 149	73.0 149	74.5 149	71.1 149	70.1 149	67.3 149	71.8 149	64.4 149	66.2 149	65.9 149	76.5 150	78.1 150	72.0 150	68.2 150	64.9 150	66.5 150	52.3 150	45.1 150	70.9 150	81.8 149	81.6 149	82.3 149
TOF-M [154]	149.4	39.2 149	39.9 149	41.8 149	73.0 149	74.5 149	71.1 149	70.1 149	67.3 149	71.8 149	64.4 149	66.2 149	65.9 149	76.5 150	78.1 150	72.0 150	68.2 150	64.9 150	66.5 150	52.3 150	45.1 150	70.9 150	81.8 149	81.6 149	82.3 149
MEMC-Net+ [155]	149.4	39.2 149	39.9 149	41.8 149	73.0 149	74.5 149	71.1 149	70.1 149	67.3 149	71.8 149	64.4 149	66.2 149	65.9 149	76.5 150	78.1 150	72.0 150	68.2 150	64.9 150	66.5 150	52.3 150	45.1 150	70.9 150	81.8 149	81.6 149	82.3 149
CFRF [156]	149.4	39.2 149	39.9 149	41.8 149	73.0 149	74.5 149	71.1 149	70.1 149	67.3 149	71.8 149	64.4 149	66.2 149	65.9 149	76.5 150	78.1 150	72.0 150	68.2 150	64.9 150	66.5 150	52.3 150	45.1 150	70.9 150	81.8 149	81.6 149	82.3 149
MPRN [157]	149.4	39.2 149	39.9 149	41.8 149	73.0 149	74.5 149	71.1 149	70.1 149	67.3 149	71.8 149	64.4 149	66.2 149	65.9 149	76.5 150	78.1 150	72.0 150	68.2 150	64.9 150	66.5 150	52.3 150	45.1 150	70.9 150	81.8 149	81.6 149	82.3 149
DAIN [158]	149.4	39.2 149	39.9 149	41.8 149	73.0 149	74.5 149	71.1 149	70.1 149	67.3 149	71.8 149	64.4 149	66.2 149	65.9 149	76.5 150	78.1 150	72.0 150	68.2 150	64.9 150	66.5 150	52.3 150	45.1 150	70.9 150	81.8 149	81.6 149	82.3 149
PyrWarp [159]	149.4	39.2 149	39.9 149	41.8 149	73.0 149	74.5 149	71.1 149	70.1 149	67.3 149	71.8 149	64.4 149	66.2 149	65.9 149	76.5 150	78.1 150	72.0 150	68.2 150	64.9 150	66.5 150	52.3 150	45.1 150	70.9 150	81.8 149	81.6 149	82.3 149
InterpCNN [160]	149.4	39.2 149	39.9 149	41.8 149	73.0 149	74.5 149	71.1 149	70.1 149	67.3 149	71.8 149	64.4 149	66.2 149	65.9 149	76.5 150	78.1 150	72.0 150	68.2 150	64.9 150	66.5 150	52.3 150	45.1 150	70.9 150	81.8 149	81.6 149	82.3 149
OFRI [161]	149.4	39.2 149	39.9 149	41.8 149	73.0 149	74.5 149	71.1 149	70.1 149	67.3 149	71.8 149	64.4 149	66.2 149	65.9 149	76.5 150	78.1 150	72.0 150	68.2 150	64.9 150	66.5 150	52.3 150	45.1 150	70.9 150	81.8 149	81.6 149	82.3 149
AVG_FLOW_ROB [141]	157.1	62.1 162	56.6 162	61.5 162	99.9 162	96.7 162	99.9 162	81.2 162	81.9 162	80.3 162	65.8 162	68.9 162	67.4 162	68.4 149	75.2 149	67.5 149	62.4 149	55.3 149	59.6 149	31.5 149	28.0 149	29.3 149	86.1 162	96.7 162	87.2 162

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] ROF-ND	4	2	color	S. Ali, C. Daul, E. Galbrun, and W. Blondel. Illumination invariant large displacement optical flow using robust neighbourhood descriptors. Submitted to CVIU 2015.
[108] DeepFlow2	16	2	color	J. Revaud, P. Weinzaepfel, Z. Harchaoui, and C. Schmid. Deep convolutional matching. Submitted to IJCV, 2015.
[109] HAST	2667	2	color	Anonymous. Highly accurate optical flow estimation on superpixel tree. ICCV 2015 submission 2221.
[110] FlowFields	15	2	color	C. Bailer, B. Taetz, and D. Stricker. Flow Fields: Dense unregularized correspondence fields for highly accurate large displacement optical flow estimation. ICCV 2015.
[111] SVFilterOh	1.56	2	color	Anonymous. Fast estimation of large displacement optical flow using PatchMatch and dominant motion patterns. CVPR 2016 submission 1788.
[112] FlowNetS+ft+v	0.5	2	color	Anonymous. Learning optical flow with convolutional neural networks. ICCV 2015 submission 235.
[113] CombBMOF	51	2	color	M. Brüggemann, R. Kays, P. Springer, and O. Erdler. Combined block-matching and adaptive differential motion estimation in a hierarchical multi-scale framework. ICGIP 2014. (Method improved since publication.)
[114] PMMST	182	2	color	F. Zhang, S. Xu, and X. Zhang. High accuracy correspondence field estimation via MST based patch matching. Submitted to TIP 2015.
[115] DF-Auto	70	2	color	N. Monzon, A. Salgado, and J. Sanchez. Regularization strategies for discontinuity-preserving optical flow methods. Submitted to TIP 2015.
[116] CPM-Flow	3	2	color	Anonymous. Efficient coarse-to-fine PatchMatch for large displacement optical flow. CVPR 2016 submission 241.
[117] CNN-flow-warp+ref	1.4	3	color	D. Teney and M. Hebert. Learning to extract motion from videos in convolutional neural networks. ArXiv 1601.07532, 2016.
[118] Steered-L1	804	2	color	Anonymous. Optical flow estimation via steered-L1 norm. Submitted to WSCG 2016.
[119] StereoOF-V1MT	343	2	gray	Anonymous. Visual features for action-oriented tasks: a cortical-like model for disparity and optic flow computation. BMVC 2016 submission 132.
[120] PGM-C	5	2	color	Y. Li. Pyramidal gradient matching for optical flow estimation. Submitted to PAMI 2016.
[121] RNLOD-Flow	1040	2	gray	C. Zhang, Z. Chen, M. Wang, M. Li, and S. Jiang. Robust non-local TV-L1 optical flow estimation with occlusion detection. IEEE TIP 26(8):4055-4067, 2017.
[122] FlowNet2	0.091	2	color	Anonymous. FlowNet 2.0: Evolution of optical flow estimation with deep networks. CVPR 2017 submission 900.
[123] S2F-IF	20	2	color	Anonymous. S2F-IF: Slow-to-fast interpolator flow. CVPR 2017 submission 765.
[124] BriefMatch	0.068	2	gray	G. Eilertsen, P.-E. Forssen, and J. Unger. Dense binary feature matching for real-time optical flow estimation. SCIA 2017 submission 62.
[125] OAR-Flow	60	2	color	Anonymous. Order-adaptive regularisation for variational optical flow: global, local and in between. SSVM 2017 submission 20.
[126] AdaConv-v1	2.8	2	color	S. Niklaus, L. Mai, and F. Liu. (Interpolation results only.) Video frame interpolation via adaptive convolution. CVPR 2017.
[127] SepConv-v1	0.2	2	color	S. Niklaus, L. Mai, and F. Liu. (Interpolation results only.) Video frame interpolation via adaptive separable convolution. ICCV 2017.
[128] ProbFlowFields	37	2	color	A. Wannenwetsch, M. Keuper, and S. Roth. ProbFlow: joint optical flow and uncertainty estimation. ICCV 2017.
[129] UnFlow	0.12	2	color	Anonymous. UnFlow: Unsupervised learning of optical flow with a bidirectional census loss. Submitted to AAAI 2018.
[130] FlowFields+	10.5	2	color	C. Bailer, B. Taetz, and D. Stricker. Flow fields: Dense correspondence fields for highly accurate large displacement optical flow estimation. Submitted to PAMI 2017.
[131] IIOF-NLDP	150	2	color	D.-H. Trinh, W. Blondel, and C. Daul. A general form of illumination-invariant descriptors in variational optical flow estimation. ICIP 2017.
[132] SuperSlomo	0.5	2	color	Anonymous. (Interpolation results only.) Super SloMo: High quality estimation of multiple intermediate frames for video interpolation. CVPR 2018 submission 325.
[133] EPMNet	0.061	2	color	Anonymous. EPM-convolution multilayer-network for optical flow estimation. ICME 2018 submission 1119.
[134] OFRF	90	2	color	T. Mai, M. Gouiffes, and S. Bouchafa. Optical Flow refinement using iterative propagation under color, proximity and flow reliability constraints. Submitted to Signal, Image and Video Processing 2017.
[135] 3DFlow	328	2	color	J. Chen, Z. Cai, J. Lai, and X. Xie. A filtering based framework for optical flow estimation. IEEE TCSVT 2018.
[136] CtxSyn	0.07	2	color	S. Niklaus and F. Liu. (Interpolation results only.) Context-aware synthesis for video frame interpolation. CVPR 2018.
[137] H+S_ROB	5	2	color	ROB 2018 baseline submission, based on: E. Meinhardt-Llopis, J. Sanchez, and D. Kondermann. Horn-Schunck optical flow with a multi-scale strategy. Image Processing On Line 3:151–172, 2013.
[138] TVL1_ROB	1	2	color	ROB 2018 baseline submission, based on: J. Sanchez, E. Meinhardt-Llopis, and G. Facciolo. TV-L1 optical flow estimation. Image Processing On Line 3:137-150, 2013.
[139] DMF_ROB	10	2	color	ROB 2018 baseline submission, based on: P. Weinzaepfel, J. Revaud, Z. Harchaoui, and C. Schmid. DeepFlow: large displacement optical flow with deep matching. ICCV 2013.
[140] JOF	657	2	gray	C. Zhang, L. Ge, Z. Chen, M. Li, W. Liu, and H. Chen. Refined TV-L1 optical flow estimation using joint filtering. Submitted to IEEE TMM, 2018.
[141] AVG_FLOW_ROB	N/A	2	N/A	Average flow field of ROB 2018 training set.
[142] LiteFlowNet	0.06	2	color	T.-W. Hui, X. Tang, and C. C. Loy. LiteFlowNet: A lightweight convolutional neural network for optical flow estimation. CVPR 2018.
[143] AugFNG_ROB	0.10	all	color	Anonymous. FusionNet and AugmentedFlowNet: Selective proxy ground truth for training on unlabeled images. ECCV 2018 submission 2834.
[144] ResPWCR_ROB	0.2	2	color	Anonymous. Learning optical flow with residual connections. ROB 2018 submission.
[145] FF++_ROB	17.43	2	color	R. Schuster, C. Bailer, O. Wasenmueller, D. Stricker. FlowFields++: Accurate optical flow correspondences meet robust interpolation. ICIP 2018. Submitted to ROB 2018.
[146] ProFlow_ROB	76	3	color	Anonymous. ProFlow: Learning to predict optical flow. BMVC 2018 submission 277.
[147] PWC-Net_ROB	0.069	2	color	D. Sun, X. Yang, M.-Y. Liu, and J. Kautz. PWC-Net: CNNs for optical flow using pyramid, warping, and cost volume. CVPR 2018.
[148] WOLF_ROB	0.02	2	color	Anonymous. Reversed deep neural network for optical flow. ROB 2018 submission.
[149] LFNet_ROB	0.068	2	color	Anonymous. Learning a flow network. ROB 2018 submission.
[150] WRT	9	2	color	L. Mei, J. Lai, X. Xie, J. Zhu, and J. Chen. Illumination-invariance optical flow estimation using weighted regularization transform. Submitted to IEEE TCSVT 2018.
[151] EAI-Flow	2.1	2	color	Anonymous. Hierarchical coherency sensitive hashing and interpolation with RANSAC for large displacement optical flow. CVIU 2018 submission 17-678.
[152] ContinualFlow_ROB	0.5	all	color	M Neoral, J. Sochman, and J. Matas. Continual occlusions and optical flow estimation. ACCV 2018.
[153] CyclicGen	0.088	2	color	Anonymous. (Interpolation results only.) Deep video frame interpolation using cyclic frame generation. AAAI 2019 submission 323.
[154] TOF-M	0.393	2	color	T. Xue, B. Chen, J. Wu, D. Wei, and W. Freeman. Video enhancement with task-oriented flow. arXiv 1711.09078, 2017.
[155] MEMC-Net+	0.16	2	color	W. Bao, W.-S. Lai, X. Zhang, Z. Gao, and M.-H. Yang. (Interpolation results only.) MEMC-Net: Motion estimation and motion compensation driven neural network for video interpolation and enhancement. Submitted to TPAMI 2018.
[156] CFRF	0.128	2	color	Anonymous. (Interpolation results only.) Coarse-to-fine refinement framework for video frame interpolation. CVPR 2019 submission 1992.
[157] MPRN	0.32	4	color	Anonymous. (Interpolation results only.) Multi-frame pyramid refinement network for video frame interpolation. CVPR 2019 submission 1361.
[158] DAIN	0.13	2	color	Anonymous. (Interpolation results only.) DAIN: Depth-aware video frame interpolation. CVPR 2019 submission 1769.
[159] PyrWarp	0.14	2	color	Anonymous. (Interpolation results only.) Feature pyramid warping for video frame interpolation. CVPR 2019 submission 868.
[160] InterpCNN	0.65	2	color	Anonymous. (Interpolation results only.) Video frame interpolation with a stack of synthesis networks and intermediate optical flows. CVPR 2019 submission 6533.
[161] OFRI	0.31	2	color	Anonymous. (Interpolation results only.) Efficient video frame interpolation via optical flow refinement. CVPR 2019 submission 6743.
[162] CompactFlow_ROB	0.05	2	color	Anonymous. CompactFlow: spatially shiftable window revisited. CVPR 2019 submission 1387.

* 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.