Research map / Detection and segmentation / Open-set and few-shot detection
Improving Facial Attribute Prediction Using Semantic Segmentation
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 semantic segmentation-based pooling (SSP) (ablation) | Global average pooling attribute model · CelebA original | classification error (%): 9.83 -> 8.98 | Helped |
| Add semantic segmentation-based gating (SSG) (ablation) | Global average pooling attribute model · CelebA original | classification error (%): 9.83 -> 9.13 | Helped |
| Combine SSP and SSG predictions by averaging (ablation) | SSP and SSG attribute models · CelebA original | classification error (%): 8.98 -> 8.84 | Helped |
| Replace global average pooling with two-level spatial pyramid pooling (ablation) | Basic attribute prediction model · CelebA pre-cropped | average precision (%): 79.74 -> 77.69 | Hurt |
| Feed segmentation maps as additional input channels (ablation) | Basic average-pooling attribute model · CelebA pre-cropped | average precision (%): 79.74 -> 79.74 | No clear effect |
| Replace semantic masks with landmark bounding boxes (ablation) | SSP-style attribute model · CelebA pre-cropped | average precision (%): 81.02 -> 79.95 | Hurt |
| Add semantic segmentation-based pooling (SSP) (ablation) | Global average pooling attribute model · LFWA | average precision (%): 82.69 -> 84.53 | Helped |
| Add semantic segmentation-based gating (SSG) (ablation) | Global average pooling attribute model · LFWA | classification error (%): 14.73 -> 13.87 | Helped |
| Combine SSP and SSG predictions by averaging (ablation) | SSP and SSG attribute models · LFWA | average precision (%): 84.53 -> 85.28 | Helped |
| Add facial attribute prediction as a joint training task (ablation) | Semantic face parsing network · Helen face parsing | average IoU (%): 61.84 -> 65.66 | Helped |
Closest papers by meaning
- Attributes for Improved Attributes: A Multi-Task Network Utilizing Implicit and Explicit Relationships for Facial Attribute Classification (AAAI 2017 · similarity 0.89)
- Attributes Shape the Embedding Space of Face Recognition Models (ICML 2025 · similarity 0.85)
- SegFace: Face Segmentation of Long-Tail Classes (AAAI 2025 · similarity 0.85)
- Self-Supervised Facial Representation Learning with Facial Region Awareness (CVPR 2024 · similarity 0.85)
- Benchmarking Segmentation Models with Mask-Preserved Attribute Editing (CVPR 2024 · similarity 0.85)
- Hunting Attributes: Context Prototype-Aware Learning for Weakly Supervised Semantic Segmentation (CVPR 2024 · similarity 0.84)
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.