HMFT: Hyperspectral and Multispectral Image Fusion Super-Resolution Method Based on Efficient Transformer and Spatial-Spectral Attention Mechanism

Table 9

Average QIs of the results and running time and memory usage on the CAVE dataset using the proposed method with and without the feature split module (FSM).

Method	PSNR	SAM	ERGAS	RMSE	Time	Memory
w/o	50.56	2.15	1.08	0.9959	1598.78	16.10
w	51.31	1.93	1.08	0.9959	152.50	4.03

The best values are underlined.