Research map / Label-efficient and robust learning / Domain adaptation and generalization
Transfer Adversarial Hashing for Hamming Space Retrieval
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 |
|---|---|---|---|
| Use TAH rather than HashNet or THN | NUS-WIDE retrieval · NUS-WIDE | MAP within Hamming Radius 2: 0.031 and 0.053 in average MAP | Helped |
| Use TAH rather than the strongest deep hashing baseline | VisDA2017 transfer retrieval · synthetic → real | MAP within Hamming Radius 2: 0.073 | Helped |
| Use TAH rather than the strongest deep hashing baseline | VisDA2017 transfer retrieval · real → synthetic | MAP within Hamming Radius 2: 0.090 | Helped |
| Use adversarial distribution alignment rather than THN's MMD alignment | TAH · VisDA2017 synthetic → real | MAP within Hamming Radius 2: 0.140 | Helped |
| Use adversarial distribution alignment rather than THN's MMD alignment | TAH · VisDA2017 real → synthetic | MAP within Hamming Radius 2: 0.096 | Helped |
| Replace pairwise t-distribution cross-entropy with DHN pairwise cross-entropy (ablation) | TAH · VisDA2017 synthetic → real and real → synthetic | average MAP within Hamming Radius 2: 0.031 / 0.060 | Hurt |
| Remove adversarial learning and target-domain unsupervised training data (ablation) | TAH · VisDA2017 synthetic → real | average MAP within Hamming Radius 2: 0.078 | Hurt |
| Remove adversarial learning and target-domain unsupervised training data (ablation) | TAH · VisDA2017 real → synthetic | average MAP within Hamming Radius 2: 0.044 | Hurt |
| Use t-transfer similarity in place of DHN similarity | DHN on VisDA2017 · real → synthetic | average similar points within Hamming Radius 2: 58 → 620 | Helped |
Closest papers by meaning
- You See What I Want You To See: Exploring Targeted Black-Box Transferability Attack for Hash-Based Image Retrieval Systems (CVPR 2021 · similarity 0.89)
- Optimal Projection Guided Transfer Hashing for Image Retrieval (AAAI 2019 · similarity 0.89)
- Deep Supervised Hashing for Fast Image Retrieval (CVPR 2016 · similarity 0.89)
- Once and for All: Universal Transferable Adversarial Perturbation against Deep Hashing-Based Facial Image Retrieval (AAAI 2024 · similarity 0.88)
- Effective Comparative Prototype Hashing for Unsupervised Domain Adaptation (AAAI 2024 · similarity 0.86)
- HUANG: A Robust Diffusion Model-based Targeted Adversarial Attack Against Deep Hashing Retrieval (AAAI 2025 · similarity 0.86)
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.