AI Paper Atlas

Research map / 3D vision / 3D generation

MV-Adapter: Multi-View Consistent Image Generation Made Easy

Zehuan Huang, Yuan-Chen Guo, Haoran Wang, Ran Yi, Lizhuang Ma, Yan-Pei Cao, Lu Sheng

ICCV 2025 · 3D vision › 3D generation · approach: Diffusion-based novel view synthesis · 102 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use parallel instead of serial attention architecture (ablation)MV-Adapter (SDXL) · GSO image-to-multiviewPSNR: 20.687 -> 22.131Improved
use parallel instead of serial attention architecture (ablation)MV-Adapter (SDXL) · GSO image-to-multiviewSSIM: 0.8681 -> 0.8816Improved
use parallel instead of serial attention architecture (ablation)MV-Adapter (SDXL) · GSO image-to-multiviewLPIPS: 0.1149 -> 0.1002Improved
scale base model from SD2.1 to SDXLMV-Adapter · Objaverse text-to-multiviewFID: 31.24 -> 29.71Improved
scale base model from SD2.1 to SDXLMV-Adapter · GSO image-to-multiviewPSNR: 20.867 -> 22.131Improved
scale base model from SD2.1 to SDXLMV-Adapter · 3D reconstructionChamfer Distance: 0.0317 -> 0.0206Improved
condition on image instead of text for texture generationMV-Adapter (SDXL) · 3D texture generationFID: 32.75 -> 27.28Improved
adapter training instead of full fine-tuning (vs Era3D)MV-Adapter (SD2.1) · training costtrainable parameters: 993M -> 127MImproved
adapter training instead of full fine-tuning (vs Era3D)MV-Adapter (SDXL) · training costtrainable parameters: 3.1B -> 490MImproved
adapter training instead of full fine-tuning (vs Era3D)MV-Adapter (SD2.1) · training speediter/s: 2.2iter/s -> 3.1iter/sImproved
decoupled adapter instead of full-tuned MVDream with LoRA (ablation)MV-Adapter · community T2I derivativesalignment with text and styleImproved
multi-view LoRA instead of decoupled attention (ablation)multi-view LoRA · switching base model or adding LoRAmulti-view consistencyHurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy