Research map / 3D vision / 3D generation
Learning Geometry-Aware Representations by Sketching
Hyundo Lee, Inwoo Hwang, Hyunsung Go, Won-Seok Choi, Kibeom Kim, Byoung-Tak Zhang
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 |
|---|---|---|---|
| Use LBS (CE) rather than E(2)-CNN | rotMNIST · 0°, 90° test digits | accuracy: 85.12\stdv 0.17 -> 65.00\stdv 3.32 | Hurt |
| Use LBS (CE) rather than E(2)-CNN | Geoclidean-Constraints · Geoclidean-Constraints | accuracy: 71.03\stdv 1.94 -> 50.01\stdv 1.58 | Hurt |
| Train without labels instead of using CE supervision (ablation) | LBS on Geoclidean-Elements · Geoclidean-Elements | accuracy: 81.06\stdv 3.14 -> 81.34\stdv 0.16 | No clear effect |
| Use LBS (CE) rather than SupCon | CLEVR labeled · leftmost object color | accuracy: 63.82\stdv{2.36} -> 81.79\stdv{0.27} | Improved |
| Use LBS rather than DefGrid | STL-10-trained representations transferred to CLEVR · CLEVR rightmost object size | accuracy: 78.97\stdv{0.29} -> 92.46\stdv{0.62} | Improved |
| Add the sketch-based representation to the image encoder (ablation) | LBS on QMUL-ShoeV2 · QMUL-ShoeV2 retrieval | Top-1 accuracy: 30.8\stdv{3.89} -> 40.8\stdv{4.19} | Improved |
| Replace perceptual loss with L1 loss (ablation) | LBS stroke representation · CLEVR rightmost color (RC) | accuracy: 55.15 -> 34.43 | Hurt |
| Replace perceptual loss with L1 loss (ablation) | LBS stroke representation · CLEVR rightmost object size | accuracy: 90.14 -> 78.32 | Hurt |
Closest papers by meaning
- Sketching based Representations for Robust Image Classification with Provable Guarantees (NeurIPS 2022 · similarity 0.88)
- Unsupervised Sketch to Photo Synthesis (ECCV 2020 · similarity 0.88)
- Open Vocabulary Semantic Scene Sketch Understanding (CVPR 2024 · similarity 0.88)
- Learning Dense Correspondences between Photos and Sketches (ICML 2023 · similarity 0.88)
- SketchINR: A First Look into Sketches as Implicit Neural Representations (CVPR 2024 · similarity 0.88)
- SketchFusion: Learning Universal Sketch Features through Fusing Foundation Models (CVPR 2025 · 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.
