Research map / Vision-language and multimodal / Human and embodied modeling
Towards Better Understanding the Clothing Fashion Styles: A Multimodal Deep Learning Approach
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 autoencoder feature learning (DA) before SVM (ablation) | SVM regression · Fashion show images, mapping to FSS (5-fold CV) | MSE: 0.2083 -> 0.1927 | Helped |
| add bimodal strategy (BDA vs DA) (ablation) | autoencoder + SVM · Fashion show images, mapping to FSS (5-fold CV) | MSE: 0.1927 -> 0.1881 | Helped |
| add correlative category labels (CDA vs DA) (ablation) | autoencoder + SVM · Fashion show images, mapping to FSS (5-fold CV) | MSE: 0.1927 -> 0.1860 | Helped |
| combine bimodal and correlative labels (BCDA vs BDA) (ablation) | autoencoder + SVM · Fashion show images, mapping to FSS (5-fold CV) | MSE: 0.1881 -> 0.1841 | Helped |
| use categories as correlative labels instead of input features (ablation) | BCDA + SVM · Fashion show images, mapping to FSS (5-fold CV) | MSE: 0.1876 -> 0.1841 | Helped |
| use SVM instead of Decision Tree as regressor (ablation) | BCDA · Fashion show images, mapping to FSS (5-fold CV) | MSE: 0.3378 -> 0.1841 | Helped |
| use SVM instead of linear regression (ablation) | BCDA · Fashion show images, mapping to FSS (5-fold CV) | MSE: 0.1854 -> 0.1841 | No clear effect |
| use SVM instead of DNN regressor (ablation) | BCDA · Fashion show images, mapping to FSS (5-fold CV) | MSE: 0.2591 -> 0.1841 | Helped |
| increase hidden layers beyond 5 (ablation) | BCDA · Fashion show images, mapping to FSS (5-fold CV) | MSE | Hurt |
| increase training data size up to about 25000 (ablation) | BCDA · Fashion show images, mapping to FSS (5-fold CV) | MSE | Helped |
| use top features instead of bottom features only (ablation) | BCDA · Fashion show images, mapping to FSS (5-fold CV) | MSE | Helped |
Closest papers by meaning
- Dress Fashionably: Learn Fashion Collocation With Deep Mixed-Category Metric Learning (AAAI 2018 · similarity 0.90)
- Attentive Fashion Grammar Network for Fashion Landmark Detection and Clothing Category Classification (CVPR 2018 · similarity 0.88)
- Easy Regional Contrastive Learning of Expressive Fashion Representations (NeurIPS 2024 · similarity 0.87)
- TexFit: Text-Driven Fashion Image Editing with Diffusion Models (AAAI 2024 · similarity 0.85)
- AIpparel: A Multimodal Foundation Model for Digital Garments (CVPR 2025 · similarity 0.85)
- AnyDressing: Customizable Multi-Garment Virtual Dressing via Latent Diffusion Models (CVPR 2025 · 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.