Table 1 Comparisons between the proposed ReMamba with state-of-the-art methods on the SYSU-MM01, RegDB and LLCM datasets. The bold font and the underline denote the best and second-best performance, respectively.

From: ReMamba: a hybrid CNN-Mamba aggregation network for visible-infrared person re-identification

Methods

SYSU-MM01

RegDB

LLCM

All Search

Indoor Search

VIS to IR

IR to VIS

VIS to IR

IR to VIS

R-1

mAP

R-1

mAP

R-1

mAP

R-1

mAP

R-1

mAP

R-1

mAP

Zero-Padding45

14.8

16.0

20.6

26.9

17.8

18.9

16.6

17.8

-

-

-

-

Hi-CMD46

34.9

35.9

-

-

70.9

66.0

-

-

-

-

-

-

AlignGAN6

42.4

40.7

45.9

54.3

57.9

53.6

56.3

53.4

-

-

-

-

X-Modality5

49.9

50.7

-

-

62.2

60.2

-

-

-

-

-

-

DDAG47

54.8

53.0

61.0

68.0

69.3

63.5

68.1

61.8

48.0

52.3

40.3

48.4

LbA10

55.4

54.1

58.5

66.3

74.2

67.6

67.5

72.4

50.8

55.6

43.8

53.1

CM-NAS48

60.8

58.9

68.0

52.4

82.8

79.3

81.7

77.6

-

-

-

-

CMTR49

65.5

62.9

71.5

76.7

88.1

81.7

84.9

80.8

-

-

-

-

PMT15

67.5

65.0

71.7

76.5

84.8

76.6

84.2

75.1

-

-

-

-

CAJ50

69.9

66.9

76.3

80.4

85.0

79.1

84.8

77.8

56.5

59.8

48.8

56.6

MPANet12

70.6

68.2

76.7

81.0

82.8

80.7

83.7

80.9

-

-

-

-

MMN7

70.6

66.9

76.2

79.6

91.6

84.1

87.5

80.5

59.9

62.7

52.5

58.9

MAUM51

71.7

68.8

77.0

81.9

87.9

85.1

87.0

84.3

-

-

-

-

DEEN43

74.7

71.8

80.3

83.3

91.1

85.1

89.5

83.4

62.5

65.8

54.9

62.9

HOSNet52

75.6

74.2

84.2

86.7

94.7

90.4

93.3

89.2

64.9

67.9

56.4

63.2

SAAI53

75.9

77.0

83.2

88.0

91.1

91.5

92.1

92.0

-

-

-

-

MSCLNet54

77.0

71.6

78.5

81.2

84.1

81.0

83.9

78.3

-

-

-

-

ReMamba(ours)

80.1

78.1

85.4

87.9

91.2

84.6

90.9

84.4

64.9

68.4

57.1

63.9