Research map / Vision-language and multimodal / Cross-modal learning
Dual Deep Neural Networks Cross-Modal Hashing
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 Stage 3 text network trained with image-network codes (ablation) | DDCMH-1 · MIRFlickr-25K T→I 64 bits | MAP: 0.7336 -> 0.7905 | Helped |
| add Stage 3 text network trained with image-network codes (ablation) | DDCMH-1 · NUS-WIDE T→I 16 bits | MAP: 0.6839 -> 0.6867 | Helped |
| add code reconstruction in Stage 3 (ablation) | DDCMH-2 · MIRFlickr-25K T→I 64 bits | MAP: 0.7295 -> 0.7905 | Helped |
| add code reconstruction in Stage 3 (ablation) | DDCMH-2 · NUS-WIDE T→I 64 bits | MAP: 0.6849 -> 0.7412 | Helped |
| use COSDISH instead of ITQ as Stage 1 single-modal hashing (ablation) | DDCMH · MIRFlickr-25K | MAP | Mixed |
| use DDCMH instead of PRDH (best deep baseline) | PRDH · MIRFlickr-25K I→T 64 bits | MAP: 0.7201 -> 0.8551 | Helped |
| use DDCMH instead of PRDH | PRDH · NUS-WIDE I→T 64 bits | MAP: 0.6506 -> 0.8316 | Helped |
| use DDCMH instead of PRDH | PRDH · NUS-WIDE T→I 16 bits | MAP: 0.6808 -> 0.6867 | Helped |
| use deep hashing instead of shallow hashing | shallow methods (SCMseq) · MIRFlickr-25K I→T 16 bits | MAP: 0.6405 -> 0.7056 (DCMH) | Helped |
Closest papers by meaning
- Deep Cross-Modal Hashing (CVPR 2017 · similarity 0.92)
- Pairwise Relationship Guided Deep Hashing for Cross-Modal Retrieval (AAAI 2017 · similarity 0.91)
- Transitive Hashing Network for Heterogeneous Multimedia Retrieval (AAAI 2017 · similarity 0.91)
- Lightweight Contrastive Distilled Hashing for Online Cross-modal Retrieval (AAAI 2025 · similarity 0.89)
- Dual Self-Paced Cross-Modal Hashing (AAAI 2024 · similarity 0.89)
- An End-To-End Graph Attention Network Hashing for Cross-Modal Retrieval (NeurIPS 2024 · similarity 0.87)
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.