Arbitrary‑Scale Spatial–Spectral Fusion using Kernel Integral and Progressive Resampling

Note: This manuscript is currently under review. Full training‐code details will be released immediately upon acceptance.

---

🍗 Model

• You can find detailed model architecture information here:

cd Assets/Model

---

📂 Dataset

• Training data & Testing data: CAVE Dataset Extraction code: dju8

---

▶️ Quick Test

Clone or navigate into the CAVE directory and run the demo:

cd CAVE
python Test.py

---

📊 Quantitative Results on CAVE ×4

The numerical results under the ×4 CAVE dataset.

Methods	PSNR	P-Value	SAM	P-Value	ERGAS	P-Value
Bicubic	34.96±3.69	<0.001	4.15±1.32	<0.001	5.16±1.08	<0.001
DHIF-Net	50.79±3.68	<0.001	1.94±0.57	<0.001	0.86±0.51	<0.001
PSRT	51.76±3.45	<0.001	2.05±0.65	<0.001	0.79±0.49	<0.001
DSPNet	50.93±3.53	<0.001	2.17±0.62	<0.001	0.86±0.46	<0.001
BDT	52.41±3.37	<0.001	1.98±0.64	<0.001	0.85±0.49	<0.001
MIMO-SST	51.33±3.55	<0.001	2.22±0.61	<0.001	0.85±0.48	<0.001
DCINN	52.66±3.47	≈0.001	1.87±0.59	<0.001	0.75±0.48	<0.001
SFNO	52.71±3.49	–	1.85±0.58	–	0.72±0.46	–
F-statistic	28.959	<0.001	10.859	<0.001	67.844	<0.001

---

📊 Quantitative Results on CAVE ×8

The numerical results under the ×8 CAVE dataset.

Methods	PSNR	P-Value	SAM	P-Value	ERGAS	P-Value
Bicubic	30.31±3.47	<0.001	5.67±1.75	<0.001	8.36±1.62	<0.001
DHIF-Net	47.59±4.20	<0.001	2.55±0.67	<0.001	1.33±0.35	<0.001
PSRT	48.84±3.78	<0.001	2.41±0.77	<0.001	1.22±0.34	<0.001
DSPNet	48.73±3.95	<0.001	2.42±0.77	≈0.001	1.12±0.34	<0.001
BDT	49.15±3.67	<0.001	2.41±0.79	<0.001	1.19±0.33	<0.001
MIMO-SST	48.83±3.92	<0.001	2.78±0.73	<0.001	1.14±0.34	<0.001
DCINN	49.48±4.01	<0.001	2.49±0.71	<0.001	1.05±0.34	<0.001
SFNO	50.04±4.12	–	2.29±0.73	–	1.00±0.34	–
F-statistic	29.412	<0.001	15.185	<0.001	152.279	<0.001

---

📊 Quantitative Results on Chikusei ×4

The numerical results under the ×4 Chikusei dataset.

Methods	PSNR	P-Value	SAM	P-Value	ERGAS	P-Value
Bicubic	39.23±4.62	<0.001	2.78±0.94	<0.001	5.86±2.05	<0.001
DHIF-Net	42.61±3.12	<0.001	2.29±0.77	<0.001	3.78±0.65	<0.001
PSRT	43.78±3.64	<0.001	2.23±0.80	<0.001	3.49±1.30	<0.001
DSPNet	43.88±3.04	<0.001	2.02±0.80	<0.001	3.27±0.71	<0.001
BDT	44.32±2.95	<0.001	1.84±0.82	<0.001	2.97±0.68	≈0.001
MIMO-SST	44.23±3.01	<0.001	2.04±0.82	<0.001	3.11±0.70	<0.001
DCINN	44.85±3.29	<0.001	1.92±0.77	<0.001	3.15±0.78	<0.001
SFNO	45.51±2.87	–	1.72±0.70	–	2.80±0.58	–
F-statistic	3.324	<0.001	3.696	<0.001	8.815	<0.001

---

📊 Quantitative Results on Chikusei ×8

The numerical results under the ×8 Chikusei dataset.

Methods	PSNR	P-Value	SAM	P-Value	ERGAS	P-Value
Bicubic	35.56±5.02	<0.001	4.05±1.02	<0.001	8.98±1.50	<0.001
DHIF-Net	41.25±3.27	<0.001	3.74±0.77	<0.001	7.37±0.44	<0.001
PSRT	43.24±4.19	<0.001	3.67±0.79	<0.001	4.75±0.92	≈0.001
DSPNet	42.47±3.30	<0.001	3.20±0.78	<0.001	6.18±0.47	<0.001
BDT	43.10±3.09	<0.001	2.99±0.84	<0.001	5.86±0.41	<0.001
MIMO-SST	42.91±3.29	<0.001	2.91±0.80	<0.001	5.18±0.54	<0.001
DCINN	43.27±3.61	≈0.001	2.55±0.77	<0.001	5.15±0.50	<0.001
SFNO	44.32±3.00	–	2.45±0.79	–	4.36±0.38	–
F-statistic	5.655	<0.001	4.943	<0.001	43.076	<0.001

---

📊 Quantitative Results on Pavia Centre ×4

The numerical results under the ×4 Pavia Centre dataset.

Methods	PSNR	P-Value	SAM	P-Value	ERGAS	P-Value
Bicubic	28.04±5.55	<0.001	6.34±0.56	<0.001	7.43±3.02	<0.001
DHIF-Net	41.64±2.56	<0.001	3.43±0.97	<0.001	2.14±0.79	<0.001
PSRT	40.58±3.83	<0.001	3.96±0.95	<0.001	2.36±2.11	<0.001
DSPNet	41.62±2.55	<0.001	3.46±0.98	<0.001	2.29±0.96	<0.001
BDT	43.17±2.53	<0.001	3.25±1.00	<0.001	1.97±0.87	<0.001
MIMO-SST	42.43±2.47	<0.001	3.35±1.03	<0.001	2.06±0.92	<0.001
DCINN	43.30±3.11	<0.001	3.19±0.95	<0.001	1.91±1.08	<0.001
SFNO	43.81±2.25	–	3.10±0.82	–	1.85±0.70	–
F-statistic	24.974	<0.001	13.680	<0.001	15.694	<0.001

---

📊 Quantitative Results on Pavia Centre ×8

The numerical results under the ×8 Pavia Centre dataset.

Methods	PSNR	P-Value	SAM	P-Value	ERGAS	P-Value
Bicubic	25.08±6.57	<0.001	8.39±0.29	<0.001	10.35±1.38	<0.001
DHIF-Net	40.65±2.34	<0.001	3.70±0.87	<0.001	2.22±0.53	<0.001
PSRT	40.08±4.60	<0.001	4.08±0.81	<0.001	2.56±1.50	<0.001
DSPNet	37.51±2.65	<0.001	4.62±0.79	<0.001	3.16±0.60	<0.001
BDT	40.22±2.51	<0.001	3.79±0.89	<0.001	2.45±0.49	<0.001
MIMO-SST	40.65±2.66	<0.001	4.00±0.87	<0.001	2.56±0.74	<0.001
DCINN	41.89±3.21	<0.001	3.60±0.83	<0.001	2.15±0.66	<0.001
SFNO	42.61±1.88	–	3.41±0.85	–	2.11±0.42	–
F-statistic	24.617	<0.001	42.043	<0.001	101.974	<0.001