List your company

Soundscape-to-image generation on SonicUrban 237K

11.553FIDSounDiT
5.4145646.8490388.2835129.71797May 19, 2025

Evaluation results

MethodLinks
SounDiT2025.0511.5530.520.7060.520.7390.759
PixArt + MHCA2025.0541.4560.5170.5920.4110.3960.796
Sound2Scene2025.0549.6810.5030.4980.3320.3240.829
GlueGen2025.0564.4340.4980.4990.4050.4810.805
CoDiPre-trained=true2025.0583.9040.5030.4970.2730.310.856
AudioTokenBackbone/Base Model=SD12025.0595.210.5050.5030.4190.4840.979
AudioTokenPre-trained=true2025.05127.3370.4970.5010.2030.1990.976
AudioTokenBackbone/Base Model=SD22025.05165.0140.5020.5040.2980.4550.98