Research map / Vision-language and multimodal / Cross-modal learning
StyleMeUp: Towards Style-Agnostic Sketch-Based Image Retrieval
Aneeshan Sain, Ayan Kumar Bhunia, Yongxin Yang, Tao Xiang, Yi-Zhe Song
Check 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 |
|---|---|---|---|
| disentangle shared semantic content from sketch-specific style and meta-train adaptation components | StyleMeUp · Chair-V2 | acc.@1: 56.54 -> 62.86 | Improved |
| disentangle shared semantic content from sketch-specific style and meta-train adaptation components | StyleMeUp · Shoe-V2 | acc.@1: 34.10 -> 36.47 | Improved |
| disentangle shared semantic content from sketch-specific style and meta-train adaptation components | StyleMeUp · Sketchy (ext) | mAP: 0.810 -> 0.905 | Improved |
| disentangle shared semantic content from sketch-specific style and meta-train adaptation components | StyleMeUp · TU Berlin (ext) | mAP: 0.690 -> 0.778 | Improved |
| add explicit sketch-style diversity modeling (ablation) | StyleMeUp · Shoe-V2 | acc.@1: 27.12 -> 36.47 | Improved |
| add meta-learned feature-transformation layers (ablation) | StyleMeUp · Shoe-V2 | acc.@1: 33.28 -> 36.47 | Improved |
| meta-learn FT parameters instead of fixing them (ablation) | StyleMeUp · Shoe-V2 | acc.@1: 34.18 -> 36.47 | Improved |
| add the adaptive disentanglement regulariser (ablation) | StyleMeUp · Shoe-V2 | acc.@1: 32.57 -> 36.47 | Improved |
| use one inner-loop update rather than additional updates (ablation) | StyleMeUp · QMUL Shoe-V2 | accuracy | Hurt |
| increase embedding dimension beyond 64 (ablation) | StyleMeUp · QMUL Shoe-V2 | accuracy | No clear effect |
| use the style-agnostic encoder as a feature extractor for sketch classification | StyleMeUp · QuickDraw | classification accuracy: 68.71% -> 75.81% | Improved |
| use meta-learned disentangled sketch representation for sketch-to-sketch retrieval | StyleMeUp · QuickDraw | mAP: +0.096 | Improved |
Later papers that compared with this method
1 later accepted paper reports StyleMeUp as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Zero-Shot Everything Sketch-Based Image Retrieval, and in Explainable Style (CVPR 2023 · better on 1 of 1 shared result)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Adaptive Fine-Grained Sketch-Based Image Retrieval (ECCV 2022 · similarity 0.92)
- Generalising Fine-Grained Sketch-Based Image Retrieval (CVPR 2019 · similarity 0.91)
- Sketch Me That Shoe (CVPR 2016 · similarity 0.91)
- A Zero-Shot Framework for Sketch based Image Retrieval (ECCV 2018 · similarity 0.91)
- ARNet: Self-Supervised FG-SBIR with Unified Sample Feature Alignment and Multi-Scale Token Recycling (AAAI 2025 · similarity 0.89)
- How to Handle Sketch-Abstraction in Sketch-Based Image Retrieval? (CVPR 2024 · similarity 0.89)
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.
