Skip to main content
Springer Nature Link
Account

Table 7 Ablation and architecture comparison of HRRM

From: AMCA-SGG: Adaptive multi-prototype learning and cross-modal alignment for unbiased scene graph generation

HRRM Configuration

mR@20

mR@50

mR@100

w/o HRRM

33.8

39.5

41.9

Mamba only

34.6

40.1

42.3

PAA only

35.8

41.2

43.1

PAA + Transformer

36.9

41.8

43.8

PAA + Mamba (Ours)

37.2

42.8

44.5

  1. The best results are highlighted in bold
Back to article page