Research map / Vision-language and multimodal / Human and embodied modeling
Fashion Forward: Forecasting Visual Style in Fashion
Ziad Al-Halah, Rainer Stiefelhagen, Kristen Grauman
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 |
|---|---|---|---|
| use exponential smoothing instead of ARIMA | style forecaster · Dresses | MAPE: 13.04 -> 6.54 | Improved |
| use exponential smoothing instead of FFNN | style forecaster · Dresses | MAPE: 41.18 -> 6.54 | Improved |
| use exponential smoothing instead of ARIMA | style forecaster · Shirts | MAPE: 3.41 -> 3.16 | Improved |
| add seasonality to autoregression (AR+S) | AR · Dresses | MAPE: 9.65 -> 12.78 | Hurt |
| model styles jointly with VAR instead of independent trajectories | AR/EXP · Dresses | MAPE: 6.54 -> 20.36 | Hurt |
| vary number of styles K between 15 and 85 (ablation) | EXP forecaster · Amazon datasets | relative forecast ranking | No clear effect |
| replace tags with text description as representation | EXP forecaster · Dresses | KL divergence: 0.0261 -> 0.0185 | Improved |
| use attribute-based styles instead of text | EXP forecaster · Tops & Tees | KL divergence: 0.0075 -> 0.0049 | Improved |
| use attribute-based styles instead of ClothingNet-FC7 features | EXP forecaster · Tops & Tees | KL divergence: 0.25 -> 0.0049 | Improved |
| use attribute-based styles instead of ClothingNet-M3 features | EXP forecaster · Dresses | KL divergence: 0.0625 -> 0.0105 | Improved |
| add text to attributes (ablation) | Attributes · Dresses | KL divergence: 0.0105 -> 0.0051 | Improved |
| add tags to attributes (ablation) | Attributes · Dresses | KL divergence: 0.0105 -> 0.0336 | Hurt |
Closest papers by meaning
- From Paris to Berlin: Discovering Fashion Style Influences Around the World (CVPR 2020 · similarity 0.90)
- From Culture to Clothing: Discovering the World Events Behind a Century of Fashion Images (ICCV 2021 · similarity 0.89)
- GeoStyle: Discovering Fashion Trends and Events (ICCV 2019 · similarity 0.89)
- POP: Mining POtential Performance of New Fashion Products via Webly Cross-Modal Query Expansion (ECCV 2022 · similarity 0.89)
- Learning the Latent "Look": Unsupervised Discovery of a Style-Coherent Embedding From Fashion Images (ICCV 2017 · similarity 0.87)
- StyleTailor: Towards Personalized Fashion Styling via Hierarchical Negative Feedback (AAAI 2026 · similarity 0.83)
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.
