Research map / 3D vision / Face and shape
Shapeglot: Learning Language for Shape Differentiation
Panos Achlioptas, Judy Fan, Robert Hawkins, Noah Goodman, Leonidas J. Guibas
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add image input to point-cloud input | baseline listener · CiC object generalization | accuracy: 66.4 -> 77.4 | Improved |
| use both images and point clouds instead of images only | baseline listener · CiC object generalization | accuracy: 77.4 -> 78.9 | Improved |
| add word attention (ablation) | baseline listener with both modalities · CiC object generalization | accuracy: 78.9 -> 79.6 | Improved |
| add word attention (ablation) | baseline listener with point clouds only · CiC object generalization | accuracy: 66.4 -> 65.6 | No clear effect |
| share weights across objects (Baseline) instead of Combined-Interpretation (ablation) | listener · CiC object generalization | accuracy: 75.9 -> 79.6 | Improved |
| add explicit context encoding (Early-Context) (ablation) | Baseline listener · CiC object generalization | accuracy: 79.6 -> 79.4 | No clear effect |
| ground LSTM with images and add point clouds late rather than the reverse (ablation) | listener · CiC | accuracy: +4.8% on average | Improved |
| add dropout and L2 regularization (ablation) | listener · CiC | accuracy: ~10% | Improved |
| add label smoothing (ablation) | listener · CiC | accuracy: ~2% | Improved |
| lesion the mentioned part in the input (ablation) | image-only listener · CiC part-lesion subset (image-only listener) | accuracy: 77.4 -> 42.8 | Hurt |
| lesion a random part in the input (ablation) | image-only listener · CiC part-lesion subset (image-only listener) | accuracy: 77.4 -> 67.0 | Hurt |
| make the speaker pragmatic (rerank with internal listener) | literal image speaker · CiC human listeners | accuracy: 68.3 -> 78.7 | Improved |
Closest papers by meaning
- Hidden in Plain Sight: Evaluating Abstract Shape Recognition in Vision-Language Models (NeurIPS 2024 · similarity 0.89)
- 3D-PLOT-LLM: Part-Level Object Tokens for 3D Large Language Models (NeurIPS 2026 · similarity 0.88)
- Unifying 2D and 3D Vision-Language Understanding (ICML 2025 · similarity 0.87)
- Synthesize Diagnose and Optimize: Towards Fine-Grained Vision-Language Understanding (CVPR 2024 · similarity 0.87)
- GroundingME: Exposing the Visual Grounding Gap in MLLMs through Multi-Dimensional Evaluation (CVPR 2026 · similarity 0.87)
- ShapeWalk: Compositional Shape Editing Through Language-Guided Chains (CVPR 2024 · similarity 0.87)
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.
