Research map / 3D vision / 3D generation
Dragging with Geometry: From Pixels to Geometry-Guided Image Editing
Xinyu Pu, Hongsong Wang, Jie Gui, Pan Zhou
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 LoRA finetuning to GeoDrag (ablation) | GeoDrag · DragBench | MD / DAI_1: MD 29.24 -> 30.91; DAI_1 0.128 -> 0.186 | Hurt |
| add LoRA finetuning to GeoDrag (ablation) | GeoDrag · DragBench | IF: 0.848 -> 0.851 | Improved |
| use GeoDrag instead of FastDrag (2D-only one-step) | GeoDrag · DragBench | MD: 32.10 -> 29.24 | Improved |
| use GeoDrag instead of FastDrag | GeoDrag · DragBench | editing time per point: 3.23 s -> 3.95 s | Hurt |
| use GeoDrag instead of FastDrag | GeoDrag · DragBench | IF: 0.850 -> 0.847 | No clear effect |
| remove depth-aware field (w/o Depth) (ablation) | GeoDrag · DragBench | MD, DAI, IF | Hurt |
| remove plane-aware field (w/o Plane) (ablation) | GeoDrag · DragBench | MD, DAI, IF | Hurt |
| use conflict-free partitioning instead of Directly Add / Pixel Distance / Drag Magnitude (ablation) | GeoDrag · DragBench | MD, DAI, IF | Improved |
| increase depth noise sigma from 0.01 to 0.5 in predicted depth (ablation) | GeoDrag · DragBench | MD: +0.66 -> +9.81 | Hurt |
| use more than 10 inversion steps (ablation) | GeoDrag · DragBench | edit quality | Hurt |
| increase number of drag points from 1 to 8 (ablation) | GeoDrag · DragBench subset (80 samples) | editing time: 12.71 s (1 point), 12.93 s (8 points) | No clear effect |
Closest papers by meaning
- 3DGS-Drag: Dragging Gaussians for Intuitive Point-Based 3D Editing (ICLR 2025 · similarity 0.91)
- FlowDrag: 3D-aware Drag-based Image Editing with Mesh-guided Deformation Vector Flow Fields (ICML 2025 · similarity 0.90)
- EasyDrag: Efficient Point-based Manipulation on Diffusion Models (CVPR 2024 · similarity 0.88)
- CLIPDrag: Combining Text-based and Drag-based Instructions for Image Editing (ICLR 2025 · similarity 0.87)
- Dragin3D: Image Editing by Dragging in 3D Space (CVPR 2025 · similarity 0.87)
- FastDrag: Manipulate Anything in One Step (NeurIPS 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.
