Research map / Vision-language and multimodal / Human and embodied modeling
Sketch Me That Shoe
Check 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 |
|---|---|---|---|
| use the proposed triplet-ranking model instead of Dense-HOG + RankSVM | fine-grained SBIR · Shoe dataset | acc.@1: 24.35% -> 39.13% | Helped |
| use the proposed triplet-ranking model instead of Dense-HOG + RankSVM | fine-grained SBIR · Chair dataset | acc.@1: 52.57% -> 69.07% | Helped |
| add sketch-specific data augmentation (ablation) | Sketch-a-Net and the shoe retrieval pipeline · Shoe dataset | acc.@1: 33.04% -> 36.52% | Helped |
| add ImageNet-1K edge-map pre-training (Step 1) (ablation) | shoe retrieval pipeline · Shoe dataset | acc.@1: 36.52% -> 38.26% | Helped |
| add category-level sketch-photo ranking pre-training (Step 3) (ablation) | shoe retrieval pipeline · Shoe dataset | acc.@1: 38.26% -> 39.13% | Helped |
| replace triplet ranking with pairwise verification (ablation) | shoe retrieval model · Shoe dataset | acc.@1: 39.13% -> 28.70% | Hurt |
| replace tied Siamese branches with heterogeneous branches (ablation) | shoe triplet-ranking model · Shoe dataset | acc.@1: 39.13% -> 21.74% | Hurt |
| use heterogeneous branches with pairwise verification (ablation) | shoe retrieval model · Shoe dataset | acc.@1: 39.13% -> 16.52% | Hurt |
| omit data augmentation while retaining Steps 1–4 (ablation) | shoe retrieval pipeline · Shoe dataset | acc.@1: 39.13% -> 37.39% | Hurt |
| use only target-data fine-tuning (Step 4) (ablation) | shoe retrieval pipeline · Shoe dataset | acc.@1: 39.13% -> 27.83% | Hurt |
Closest papers by meaning
- Generalising Fine-Grained Sketch-Based Image Retrieval (CVPR 2019 · similarity 0.94)
- Deep Sketch Hashing: Fast Free-Hand Sketch-Based Image Retrieval (CVPR 2017 · similarity 0.93)
- ARNet: Self-Supervised FG-SBIR with Unified Sample Feature Alignment and Multi-Scale Token Recycling (AAAI 2025 · similarity 0.91)
- You'll Never Walk Alone: A Sketch and Text Duet for Fine-Grained Image Retrieval (CVPR 2024 · similarity 0.91)
- Sketch Down the FLOPs: Towards Efficient Networks for Human Sketch (CVPR 2025 · similarity 0.91)
- How to Handle Sketch-Abstraction in Sketch-Based Image Retrieval? (CVPR 2024 · similarity 0.90)
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.