Research map / Label-efficient and robust learning / Recognition and long-tail
Learning Deep Binary Descriptor With Multi-Quantization
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 |
|---|---|---|---|
| replace rigid sign binarization with KAEs multi-quantization (ablation) | DBD-MQ · CIFAR-10 test, 16 bits | mAP: 19.16 -> 21.53 | Helped |
| replace rigid sign binarization with KAEs multi-quantization (ablation) | DBD-MQ · CIFAR-10 test, 32 bits | mAP: 23.89 -> 26.50 | Helped |
| replace rigid sign binarization with KAEs multi-quantization (ablation) | DBD-MQ · CIFAR-10 test, 64 bits | mAP: 26.90 -> 31.85 | Helped |
| replace rigid sign binarization with KAEs multi-quantization (ablation) | DBD-MQ · Brown average | 95% error rate (ERR): 41.15 -> 38.59 | Helped |
| use DBD-MQ rather than DeepBit | unsupervised hashing on CIFAR-10 · CIFAR-10 test, 16 bits | mAP: 19.43 -> 21.53 | Helped |
| use DBD-MQ rather than DeepBit | unsupervised hashing on CIFAR-10 · CIFAR-10 test, 32 bits | mAP: 24.86 -> 26.50 | Helped |
| use DBD-MQ rather than DeepBit | unsupervised hashing on CIFAR-10 · CIFAR-10 test, 64 bits | mAP: 27.73 -> 31.85 | Helped |
| use DBD-MQ rather than DeepBit | binary descriptor matching · Brown average | 95% error rate (ERR): 40.67 -> 38.59 | Helped |
| use DBD-MQ rather than SIFT | image retrieval · Oxford | mAP: 43.7 -> 38.9 | Hurt |
| use DBD-MQ rather than CKN-raw | image retrieval · Oxford | mAP: 23.0 -> 38.9 | Helped |
Closest papers by meaning
- Learning Compact Binary Descriptors With Unsupervised Deep Neural Networks (CVPR 2016 · similarity 0.90)
- DLBD: A Self-Supervised Direct-Learned Binary Descriptor (CVPR 2023 · similarity 0.86)
- BinGAN: Learning Compact Binary Descriptors with a Regularized GAN (NeurIPS 2018 · similarity 0.84)
- Contrastive Quantization with Code Memory for Unsupervised Image Retrieval (AAAI 2022 · similarity 0.84)
- Quantization Networks (CVPR 2019 · similarity 0.84)
- Deep Visual-Semantic Quantization for Efficient Image Retrieval (CVPR 2017 · similarity 0.84)
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.