AI Paper Atlas

Research map / 3D vision / Face and shape

ISS: Image as Stepping Stone for Text-Guided 3D Shape Generation

Zhengzhe Liu, Peng Dai, Ruihui Li, XIAOJUAN QI, Chi-Wing Fu

ICLR 2023 · 3D vision › Face and shape · approach: 3D shape generative modeling · 4 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
Add stage-1 CLIP image-to-shape alignment (ablation)ISS · ShapeNet text setFID: 222.96 -> 124.42 ± 5.11Improved
Add stage-1 CLIP image-to-shape alignment (ablation)ISS · ShapeNet text setConsistency Score: 1.92 ± 2.22 -> 60.0 ± 10.94Improved
Add test-time text-to-shape alignment (ablation)ISS · ShapeNet text setConsistency Score: 29.52 ± 14.86 -> 60.0 ± 10.94Improved
Add test-time text-to-shape alignment (ablation)ISS · ShapeNet text setFID: 202.33 -> 124.42 ± 5.11Improved
Add the background loss in stage 1 (ablation)ISS · ShapeNet text setConsistency Score: 29.45 ± 14.67 -> 60.0 ± 10.94Improved
Add the background loss in stage 2 (ablation)ISS · ShapeNet text setConsistency Score: 31.55 ± 8.87 -> 60.0 ± 10.94Improved
Add background loss in both stages (ablation)ISS · ShapeNet text setConsistency Score: 30.96 ± 15.49 -> 60.0 ± 10.94Improved
Use the two-stage image-to-shape and text-to-shape alignment instead of the direct CLIP image encoder (ablation)ISS · ShapeNet text setFID: 181.88 -> 124.42 ± 5.11Improved
Use the two-stage image-to-shape and text-to-shape alignment instead of the direct CLIP image encoder (ablation)ISS · ShapeNet text setConsistency Score: 20.97 ± 13.59 -> 60.0 ± 10.94Improved
Generate a shape directly from text with ISS instead of generating an image with LAFITE and reconstructing it with DVRISS · ShapeNet text setFID: 135.01 -> 124.42 ± 5.11Improved
Generate a shape directly from text with ISS instead of generating an image with LAFITE and reconstructing it with DVRISS · ShapeNet text setConsistency Score: 52.12 ± 11.05 -> 60.0 ± 10.94Improved
Generate a shape directly from text with ISS instead of using CLIP-ForgeISS · ShapeNet text setFID: 162.87 -> 124.42 ± 5.11Improved

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