cross-domain-logo-recognition
https://github.com/wacv2025-image-quality-workshop2/cross-domain-logo-recognition
Science Score: 44.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
✓CITATION.cff file
Found CITATION.cff file -
✓codemeta.json file
Found codemeta.json file -
✓.zenodo.json file
Found .zenodo.json file -
○DOI references
-
○Academic publication links
-
○Academic email domains
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (8.6%) to scientific vocabulary
Repository
Basic Info
- Host: GitHub
- Owner: wacv2025-image-quality-workshop2
- Default Branch: main
- Size: 24.1 MB
Statistics
- Stars: 5
- Watchers: 2
- Forks: 1
- Open Issues: 0
- Releases: 0
Metadata Files
README.md
Cross-Domain Logo Recognition Dataset
Overview
The Cross-Domain Logo Recognition Dataset is designed to foster research in cross-domain logo recognition. This dataset provides images of logos collected from two distinct domains—Product and Registration—allowing researchers to explore domain adaptation, domain generalization, and other challenges in logo recognition.
This dataset is suitable for tasks such as: - Few-shot learning for logo classification. - Domain adaptation from one logo distribution (Product) to another (Registration). - Cross-domain retrieval and representation learning for brand logo recognition.
Dataset Structure
The dataset consists of images from two distinct domains:
- Product domain: Logos appearing on real-world product packaging, advertisements, and marketing materials.
- Registration domain: Logos extracted from official trademark registration records.
Training Split
| Domain | #Classes | #Images | #Images per Logo | |--------------|---------|----------|------------------| | Product | 9,282 | 212,381 | 22.8 | | Registration | 9,205 | 22,196 | 2.4 |
Validation Split
| Domain | #Classes | #Images | #Images per Logo | |--------------|---------|----------|------------------| | Product | 1,040 | 23,246 | 22.3 | | Registration | 1,022 | 2,352 | 2.3 |
Key Characteristics
- Large class diversity: Over 9,000 unique logos across both domains.
- Significant domain shift: Logos in the Product domain are more varied in terms of color, distortions, and real-world noise, whereas the Registration domain contains cleaner, official logo designs.
- Few-shot setting for Registration domain: The Registration domain has an average of 2.3 images per class, making it a challenging test bed for low-data learning.
Usage
Data Format
- The dataset is organized by domain and class labels.
- Images are provided as PNG/JPEG files.
Benchmarking
- Baseline task: Train/Indexing on the Registration domain and evaluate on the Product domain.
Visualization of cross domain pairs
Citation
If you use this dataset in your research, please cite:
@software{Zhao_Open_set_cross_2025,
author = {Zhao, Xiaonan and Li, Chenge and Liu, Zongyi and Feng, Yarong and Chen, Qipin},
month = feb,
title = {{Open set cross domain few shot logo recognition}},
url = {https://github.com/wacv2025-image-quality-workshop2/cross-domain-logo-recognition},
version = {1.0.4},
year = {2025}
}
Owner
- Login: wacv2025-image-quality-workshop2
- Kind: user
- Repositories: 1
- Profile: https://github.com/wacv2025-image-quality-workshop2
Citation (CITATION.cff)
cff-version: 1.2.0 message: "If you use this software, please cite it as below." authors: - family-names: "Zhao" given-names: "Xiaonan" orcid: "https://orcid.org/0009-0002-6522-0277" - family-names: "Li" given-names: "Chenge" orcid: "https://orcid.org/0000-0000-0000-0000" - family-names: "Liu" given-names: "Zongyi" orcid: "https://orcid.org/0000-0000-0000-0000" - family-names: "Feng" given-names: "Yarong" orcid: "https://orcid.org/0000-0000-0000-0000" - family-names: "Chen" given-names: "Qipin" orcid: "https://orcid.org/0000-0000-0000-0000" title: "Open set cross domain few shot logo recognition" version: 1.0.4 date-released: 2025-02-27 url: "https://github.com/wacv2025-image-quality-workshop2/cross-domain-logo-recognition"
GitHub Events
Total
- Watch event: 5
- Push event: 9
- Pull request event: 3
- Fork event: 1
- Create event: 1
Last Year
- Watch event: 5
- Push event: 9
- Pull request event: 3
- Fork event: 1
- Create event: 1
Issues and Pull Requests
Last synced: 8 months ago
All Time
- Total issues: 0
- Total pull requests: 2
- Average time to close issues: N/A
- Average time to close pull requests: about 3 hours
- Total issue authors: 0
- Total pull request authors: 1
- Average comments per issue: 0
- Average comments per pull request: 0.0
- Merged pull requests: 2
- Bot issues: 0
- Bot pull requests: 0
Past Year
- Issues: 0
- Pull requests: 2
- Average time to close issues: N/A
- Average time to close pull requests: about 3 hours
- Issue authors: 0
- Pull request authors: 1
- Average comments per issue: 0
- Average comments per pull request: 0.0
- Merged pull requests: 2
- Bot issues: 0
- Bot pull requests: 0
Top Authors
Issue Authors
Pull Request Authors
- xiaonanzzz (2)