Research map / Detection and segmentation / CNN features
Joint Learning of Semantic Alignment and Object Landmark Detection
Sangryul Jeon, Dongbo Min, Seungryong Kim, Kwanghoon Sohn
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 joint learning instead of separate learning (ablation) | semantic alignment and landmark networks · JLAD | PCK@0.1: 63.2 -> 72.7 | Improved |
| add joint learning instead of separate learning (ablation) | landmark detection network · JLAD | IOD: 7.97 -> 6.92 | Improved |
| add alternation iteration 1 to 3 (ablation) | joint learning · JLAD | PCK@0.1: 67.0 -> 72.1 | Improved |
| add a fourth alternation (ablation) | joint learning · JLAD | PCK@0.1: 72.1 -> 72.7 | Improved |
| add uncertainty prediction model (ablation) | semantic alignment network · PF-PASCAL | PCK@0.1: 68.2 -> 72.7 | Improved |
| add uncertainty prediction model (ablation) | semantic alignment network · PF-PASCAL | PCK@0.05: 49.4 -> 52.8 | Improved |
| add uncertainty prediction model (ablation) | semantic alignment network · PF-PASCAL | PCK@0.15: 76.9 -> 79.2 | Improved |
| add self-similarity correlation layer (ablation) | landmark detection network · MAFL | IOD: 3.58 -> 2.98 | Improved |
| add self-similarity correlation layer (ablation) | landmark detection network · AFLW | IOD: 7.72 -> 6.51 | Improved |
| add self-similarity correlation layer (ablation) | landmark detection network · JLAD | IOD: 8.16 -> 6.92 | Improved |
| use 30 landmarks instead of 10 (ablation) | landmark detection network · MAFL | IOD: 3.33 -> 2.98 | Improved |
| use 20 landmarks instead of 10 (ablation) | landmark detection network · JLAD | IOD: 7.54 -> 6.92 | Improved |
Closest papers by meaning
- Semantic Alignment: Finding Semantically Consistent Ground-Truth for Facial Landmark Detection (CVPR 2019 · similarity 0.90)
- Facial Landmark Detection with Tweaked Convolutional Neural Networks (TPAMI 2017 · similarity 0.90)
- End-to-End Weakly-Supervised Semantic Alignment (CVPR 2018 · similarity 0.89)
- AnchorNet: A Weakly Supervised Network to Learn Geometry-Sensitive Features for Semantic Matching (CVPR 2017 · similarity 0.88)
- Self-Supervised Learning of Geometrically Stable Features Through Probabilistic Introspection (CVPR 2018 · similarity 0.88)
- Deep Shape Matching (ECCV 2018 · similarity 0.88)
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.
