AI Paper Atlas

Research map / 3D vision / Face and shape

Text2Mesh: Text-Driven Neural Stylization for Meshes

Oscar Michel, Roi Bar-On, Richard Liu, Sagie Benaim, Rana Hanocka

CVPR 2022 · 3D vision › Face and shape · approach: Textured mesh generation · 289 citations (OpenAlex)

Check 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 Text2Mesh instead of the VQGAN baselineText-driven mesh stylization · User study (Q1): Overallmean opinion score: 2.83 -> 3.90Improved
Use Text2Mesh instead of the VQGAN baselineText-driven mesh stylization · User study (Q2): Contentmean opinion score: 3.60 -> 4.04Improved
Use Text2Mesh instead of the VQGAN baselineText-driven mesh stylization · User study (Q3): Stylemean opinion score: 2.59 -> 3.91Improved
Add the neural style field network rather than directly optimizing vertex colors and displacements (ablation)Text2Mesh · Text-driven mesh stylizationvisual qualityImproved
Add random 2D augmentations (ablation)Text2Mesh · Text-driven mesh stylizationalignment with target textImproved
Add Fourier feature positional encoding (ablation)Text2Mesh · Text-driven mesh stylizationfine-grained style detailImproved
Add cropping to the local image augmentation (ablation)Text2Mesh · Text-driven mesh stylizationfine-grained style detailImproved
Add the geometry-only component to the similarity loss (ablation)Text2Mesh · Text-driven mesh stylizationgeometric refinementImproved
Add the input mesh as a geometric prior (ablation)Text2Mesh · Text-driven mesh stylizationpreservation of global content structureImproved
Learn color and geometry together rather than geometry alone (ablation)Text2Mesh · Text-driven mesh stylizationgeometry qualityImproved
Learn geometry and color together rather than color alone (ablation)Text2Mesh · Text-driven mesh stylizationrealism of texture and geometryImproved
Use five or more sampled views rather than fewer sampled views (ablation)Text2Mesh · Text-driven mesh stylizationstylization qualityNo clear effect

Later papers that compared with this method

8 later accepted papers report Text2Mesh as a baseline in their results tables; 7 of them report a better number on most of the results they share with it.

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