Research map / Label-efficient and robust learning / Recognition and long-tail
Consensus Style Centralizing Auto-Encoder for Weak Style Classification
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 style centralizing auto-encoder (SCAE) on top of concatenated style descriptor (ablation) | style descriptor · Hipster Wars p=0.1 | accuracy (%): 77.73 -> 84.37 | Helped |
| add multi-column late fusion (MC-SCAE) to SCAE (ablation) | SCAE · Hipster Wars p=0.1 | accuracy (%): 84.37 -> 87.42 | Helped |
| add column consensus constraint to MC-SCAE (CSCAE) (ablation) | MC-SCAE · Hipster Wars p=0.1 | accuracy (%): 87.42 -> 89.21 | Helped |
| add column consensus constraint to MC-SCAE (ablation) | MC-SCAE · Hipster Wars p=0.2 | accuracy (%): 77.00 -> 75.32 | Hurt |
| add column consensus constraint to MC-SCAE (ablation) | MC-SCAE · Online Shopping | accuracy (%): 66.15 -> 70.41 (phi=7); 53.28 -> 51.18 (phi=4) | Mixed |
| add multi-column fusion to SCAE (ablation) | SCAE · Online Shopping phi=7 | accuracy (%): 74.33 -> 66.15 | Hurt |
| add column consensus constraint to MC-SCAE (ablation) | MC-SCAE · Manga | accuracy (%): 85.38 -> 86.23 (p=0.1); 69.58 -> 68.63 (p=0.4) | Mixed |
| use CSCAE instead of [MM, 2014] features | SVM on manga features · Manga p=0.5 | accuracy (%): 60.07 -> 67.35 | Helped |
| use CSCAE instead of [ECCV, 2014] style descriptor | nearest neighbor on style descriptor · Online Shopping phi=4 | accuracy (%): 48.63 -> 51.18 | Helped |
Closest papers by meaning
- Fashion Style in 128 Floats: Joint Ranking and Classification Using Weak Data for Feature Extraction (CVPR 2016 · similarity 0.84)
- StyleBank: An Explicit Representation for Neural Image Style Transfer (CVPR 2017 · similarity 0.83)
- Towards Better Understanding the Clothing Fashion Styles: A Multimodal Deep Learning Approach (AAAI 2017 · similarity 0.82)
- Easy Regional Contrastive Learning of Expressive Fashion Representations (NeurIPS 2024 · similarity 0.81)
- CSGO: Content-Style Composition in Text-to-Image Generation (NeurIPS 2025 · similarity 0.81)
- FashionERN: Enhance-and-Refine Network for Composed Fashion Image Retrieval (AAAI 2024 · similarity 0.81)
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.