Text-to-scene generation on 3D-FRONT Diningroom (test) benchmark leaderboard · SOTA2 Research

Text-to-scene generation on 3D-FRONT Diningroom (test)

125.32FIDMG-DiT

Evaluation Results

Method	Links
MG-DiT2025.06		125.32	7.14	10.68	55.59	75.01
MG-DiT2025.06		127.28	8.01	14.83	56.82	63.39
FreeScene2026.01		127.28	8.01	0.0148	—	63.39
InstructScene2025.06		129.13	8.24	15.27	58.93	61.47
InstructScene2026.01		129.13	8.24	0.0153	—	61.47
SceneNAT2026.01		129.65	7.51	0.0123	—	61.89
ATISS2025.06		134.75	9.82	24.25	59.83	32.56
ATISS2026.01		134.75	9.82	0.0243	—	32.56
DiffuScene2025.06		142.37	11.76	28.36	60.88	37.17
DiffuScene2026.01		142.37	11.76	0.0284	—	37.17