notesum.ai

Published at November 29

Retrieval-guided Cross-view Image Synthesis

cs.CV

cs.LG

Released Date: November 29, 2024

Authors: Hongji Yang¹, Yiru Li¹, Yingying Zhu¹

Aff.: ¹College of Computer Science and Software Engineering, Shenzhen University

Arxiv: http://arxiv.org/pdf/2411.19510v1

Refer to caption

Direction	Method	CVUSA					CVACT
Direction	Method	SSIM $\uparrow$	PSNR $\uparrow$	LPIPS $\downarrow$	FID $\downarrow$	R@1 $\uparrow$	SSIM $\uparrow$	PSNR $\uparrow$	LPIPS $\downarrow$	FID $\downarrow$	R@1 $\uparrow$
a2g	Pix2Pix	0.2849	12.14	0.5712	82.84	0.01	0.3634	13.37	0.4943	86.21	0.00
	XFork	0.3408	13.25	0.5611	79.75	6.41	0.3701	14.17	0.4919	47.98	8.72
	SelectionGAN	0.3278	13.37	0.5331	90.72	4.58	0.4705	14.31	0.5141	95.67	6.67
	PanoGAN	0.3024	13.67	0.4684	75.24	33.11	0.4631	14.18	0.4762	82.65	28.71
	CDE	0.2980	13.87	0.4752	20.63	85.04	0.4506	13.98	0.4927	43.96	65.04
	S2SP	0.3437	13.32	0.4688	44.15	10.09	0.4521	14.14	0.4718	39.64	29.39
	PPGAN	0.3516	13.91	-	-	-	-	-	-	-	-
	Sat2Density	0.3390	14.23	-	41.43	-	0.3870	14.27	-	47.09	-
	ControlNet	0.2770	11.18	-	44.63	-	0.3400	12.15	-	47.15	-
	Instruct pix2pix	0.2550	10.66	-	68.75	-	0.3920	13.12	-	57.74	-
	CrossViewDiff	0.3710	12.00	-	23.67	-	0.4120	12.41	-	41.94	-
	Ours	0.3706	14.33	0.4302	13.57	96.25	0.4945	14.55	0.4540	21.83	87.90
g2a	Pix2Pix	0.1956	15.07	0.6220	121.95	7.85	0.0870	14.24	0.6612	133.39	13.06
	CDE	0.2167	15.19	0.5706	121.98	14.73	0.0906	14.59	0.6689	160.81	14.99
	Ours	0.2461	15.77	0.5181	41.65	95.14	0.1966	16.29	0.5551	36.54	87.81