Research map / Point clouds / Shape and surface completion
Interactive Annotation of 3D Object Geometry using 2D Scribbles
Tianchang Shen, Jun Gao, Amlan Kar, Sanja Fidler
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 scribble annotation (SIM) on top of automatic prediction | GenRe · ShapeNet seen classes | Chamfer distance: 0.1093 -> 0.0503 | Improved |
| add scribble annotation (SIM) on top of automatic prediction | GenRe · ShapeNet unseen classes (average) | Chamfer distance: 0.1640 -> 0.0541 | Improved |
| add scribble annotation (SIM) on top of automatic prediction | GenRe · ShapeNet unseen classes (average) | F1 score: 0.6963 -> 0.9578 | Improved |
| add scribble annotation (SIM) on top of automatic prediction | GenRe · ShapeNet unseen classes (average) | Normal consistency: 0.5764 -> 0.8002 | Improved |
| add Vox2Mesh mesh conversion after SIM | GenRe + SIM · ShapeNet seen classes | Normal consistency: 0.7954 -> 0.8128 | Improved |
| add Vox2Mesh mesh conversion after SIM | GenRe + SIM · ShapeNet unseen classes (average) | Chamfer distance: 0.0541 -> 0.0543 | No clear effect |
| add scribble annotation (SIM) on top of automatic prediction | GenRe · Pix3D Table | Chamfer distance: 0.1735 -> 0.0968 | Improved |
| add Vox2Mesh mesh conversion after SIM | GenRe + SIM · Pix3D Wardrobe | Chamfer distance: 0.0413 -> 0.0458 | Hurt |
| add multi-view scribble propagation instead of direct 3D scribble extrusion (ablation) | SIM · ShapeNet unseen classes | Chamfer distance: 0.0804 -> 0.0589 | Improved |
| add multi-view scribble propagation instead of direct 3D scribble extrusion (ablation) | SIM · ShapeNet unseen classes | F1 score: 0.9109 -> 0.9538 | Improved |
| add multi-view side-view prediction over single-view feature extrusion (ablation) | SIM · ShapeNet seen classes | Chamfer distance: 0.0555 -> 0.0535 | Improved |
| add multi-view side-view prediction over single-view feature extrusion (ablation) | SIM · ShapeNet unseen classes | Chamfer distance: 0.0609 -> 0.0589 | Improved |
Closest papers by meaning
- Learning 3D Object Shape and Layout Without 3D Supervision (CVPR 2022 · similarity 0.87)
- Interactive 3D Object Detection with Prompts (ECCV 2024 · similarity 0.87)
- Pixels, Voxels, and Views: A Study of Shape Representations for Single View 3D Object Shape Prediction (CVPR 2018 · similarity 0.87)
- ShapeTalk: A Language Dataset and Framework for 3D Shape Edits and Deformations (CVPR 2023 · similarity 0.87)
- ImageNet3D: Towards General-Purpose Object-Level 3D Understanding (NeurIPS 2024 · similarity 0.86)
- AGILE3D: Attention Guided Interactive Multi-object 3D Segmentation (ICLR 2024 · similarity 0.85)
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.
