Research map / Graph learning / Knowledge and temporal graphs
Facial Action Unit Intensity Estimation via Semantic Correspondence Learning with Dynamic Graph Convolution
Read the paperCheck 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 dynamic semantic correspondence convolutions to the heatmap regression network | ResNet-Deconv · BP4D | ICC: .67 -> .72 | Helped |
| add dynamic semantic correspondence convolutions to the heatmap regression network | ResNet-Deconv · DISFA | ICC: .40 -> .47 | Helped |
| add SCC modules (ablation) | SCC-Heatmap with 3 deconvolutional layers and 64×64 heatmap · DISFA | ICC: .40 -> .47 | Helped |
| add SCC modules (ablation) | SCC-Heatmap with 3 deconvolutional layers and 64×64 heatmap · DISFA | MAE: .21 -> .20 | Helped |
| use 3 deconvolutional layers and a 64×64 heatmap rather than 2 layers and a 32×32 heatmap (ablation) | SCC-Heatmap · DISFA | ICC: .44 -> .47 | Helped |
| use 3 deconvolutional layers and a 64×64 heatmap rather than 2 layers and a 32×32 heatmap (ablation) | SCC-Heatmap · DISFA | MAE: .23 -> .20 | Helped |
| include both the center feature fᵢ and local difference fⱼ−fᵢ in edge features (ablation) | SCC-Heatmap · DISFA | ICC: .43 -> .47; .38 -> .47 | Helped |
| include both the center feature fᵢ and local difference fⱼ−fᵢ in edge features (ablation) | SCC-Heatmap · DISFA | MAE: .22 -> .20; .21 -> .20 | Mixed |
| increase the number of nearest neighbors k (ablation) | SCC-Heatmap · BP4D and DISFA | ICC | Hurt |
Closest papers by meaning
- Dynamic Probabilistic Graph Convolution for Facial Action Unit Intensity Estimation (CVPR 2021 · similarity 0.92)
- Weakly-Supervised Deep Convolutional Neural Network Learning for Facial Action Unit Intensity Estimation (CVPR 2018 · similarity 0.92)
- Joint Representation and Estimator Learning for Facial Action Unit Intensity Estimation (CVPR 2019 · similarity 0.90)
- Exploiting Semantic Embedding and Visual Feature for Facial Action Unit Detection (CVPR 2021 · similarity 0.89)
- Trend-Aware Supervision: On Learning Invariance for Semi-supervised Facial Action Unit Intensity Estimation (AAAI 2024 · similarity 0.88)
- AUEditNet: Dual-Branch Facial Action Unit Intensity Manipulation with Implicit Disentanglement (CVPR 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.