Research map / 3D vision / Face and shape
Understanding Pixel-level 2D Image Semantics with 3D Keypoint Knowledge Engine
Yang You, Chengkun Li, Yujing Lou, Zhoujun Cheng, Liangwei Li, Lizhuang Ma, Weiming Wang, Cewu Lu
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 |
|---|---|---|---|
| predict viewpoint directly from RGB image instead of from 2.5D sketches (ablation) | Ours · Pix3D Chair | PCK (alpha_img=0.1): 0.238/0.117 -> 0.560/0.323 | Improved |
| predict viewpoint directly from RGB image instead of from 2.5D sketches (ablation) | Ours · PF-PASCAL Chair | PCK (alpha_img=0.1): 0.171/0.075 -> 0.444/0.342 | Improved |
| predict viewpoint directly from RGB image instead of from 2.5D sketches (ablation) | Ours · SPair-71k Chair | PCK (alpha_img=0.1): 0.185/0.098 -> 0.565/0.346 | Improved |
| add differentiable viewpoint fine-tuning (ablation) | Ours · PF-PASCAL Chair | PCK (alpha_img=0.1): 0.497/0.316 -> 0.444/0.342 | Mixed |
| add differentiable viewpoint fine-tuning (ablation) | Ours · SPair-71k Chair | PCK (alpha_img=0.1): 0.538/0.332 -> 0.565/0.346 | Improved |
| add differentiable viewpoint fine-tuning (ablation) | Ours · Pix3D Chair | PCK (alpha_img=0.1): 0.560/0.322 -> 0.560/0.323 | No clear effect |
| replace predicted viewpoints with ground truth | Ours · Pix3D Chair | PCK (alpha_img=0.1): 0.560 -> 0.606 | Improved |
| replace predicted viewpoints with ground truth | Ours · ShapeNet Chair | PCK (alpha_img=0.1): 0.513 -> 0.545 | Improved |
| replace predicted 3D models with ground truth | Ours · Pix3D Chair | PCK (alpha_img=0.1): 0.560 -> 0.571 | Improved |
| replace predicted 2.5D sketches with ground truth | Ours · ShapeNet Chair | PCK (alpha_img=0.1): 0.513 -> 0.518 | No clear effect |
Closest papers by meaning
- ImageNet3D: Towards General-Purpose Object-Level 3D Understanding (NeurIPS 2024 · similarity 0.89)
- Unsupervised 3D Structure Inference from Category-Specific Image Collections (CVPR 2024 · similarity 0.88)
- Back to 3D: Few-Shot 3D Keypoint Detection with Back-Projected 2D Features (CVPR 2024 · similarity 0.87)
- Masked Point-Entity Contrast for Open-Vocabulary 3D Scene Understanding (CVPR 2025 · similarity 0.87)
- MESC-3D:Mining Effective Semantic Cues for 3D Reconstruction from a Single Image (CVPR 2025 · similarity 0.87)
- Tackling View-Dependent Semantics in 3D Language Gaussian Splatting (ICML 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.
