Research map / Label-efficient and robust learning / Recognition and long-tail
Direct Hashing Without Pseudo-Labels
Feng Zheng, Heng Huang
Read the paperCheck 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 |
|---|---|---|---|
| Use W-shape loss instead of ITQ's squared quantization approach | ITQ · Toy set | average quantization loss: 1.599 -> 0.783 | Improved |
| Use W-shape loss instead of IsoHash | IsoHash · Toy set | average quantization loss: 1.59 -> 0.783 | Improved |
| Use WSL instead of ITQ at 16 bits | ITQ · CIFAR-10 | MAP: 0.167 -> 0.183 | Improved |
| Use WSL instead of ITQ at 64 bits | ITQ · CIFAR-10 | MAP: 0.179 -> 0.227 | Improved |
| Use WSL instead of ITQ at 16 bits | ITQ · MNIST | MAP: 0.410 -> 0.434 | Improved |
| Use WSL instead of ITQ at 64 bits | ITQ · MNIST | MAP: 0.456 -> 0.522 | Improved |
| Use WSL instead of SpH at 64 bits | SpH · MNIST | MAP: 0.356 -> 0.522 | Improved |
| Increase hash code length from 16 to 64 bits (ablation) | WSL · CIFAR-10 | MAP: 0.183 -> 0.227 | Improved |
| Increase hash code length from 16 to 64 bits | PCAH · MNIST | MAP: 0.276 -> 0.212 | Hurt |
| Set mu too small (ablation) | WSL · Embedding study | projection norm / embedded values | Hurt |
| Set mu too large (ablation) | WSL · Embedding study | retrieval performance | Hurt |
| Use linear WSL instead of nonlinear AGH | AGH · SIFT1M | recall@R | Improved |
Closest papers by meaning
- Optimizing affinity-based binary hashing using auxiliary coordinates (NeurIPS 2016 · similarity 0.88)
- MIHash: Online Hashing With Mutual Information (ICCV 2017 · similarity 0.87)
- Deep Hash Distillation for Image Retrieval (ECCV 2022 · similarity 0.87)
- Hashing with Binary Matrix Pursuit (ECCV 2018 · similarity 0.87)
- Generalized Debiased Semi-Supervised Hashing for Large-Scale Image Retrieval (AAAI 2025 · similarity 0.86)
- Dual Self-Paced Cross-Modal Hashing (AAAI 2024 · 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.
