# SUBIC: A supervised, structured binary code for image search

arXiv: https://arxiv.org/abs/1708.02932 · HTML: https://mlsanity.com/paper/1708.02932/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 9 Aug 2017.

## Built on by (most cited first, 12 of 12)

- [FSDR: Frequency Space Domain Randomization for Domain Generalization](https://mlsanity.com/paper/2103.02370/) (2021)
- [A Comprehensive Survey on Source-free Domain Adaptation](https://mlsanity.com/paper/2302.11803/) (2023)
- [Self-supervised Product Quantization for Deep Unsupervised Image Retrieval](https://mlsanity.com/paper/2109.02244/) (2021)
- [Source-free unsupervised domain adaptation for cross-modality abdominal multi-organ segmentation](https://mlsanity.com/paper/2111.12221/) (2021)
- [LUT-NN: Empower Efficient Neural Network Inference with Centroid Learning and Table Lookup](https://mlsanity.com/paper/2302.03213/) (2023)
- [BTR: Binary Token Representations for Efficient Retrieval Augmented Language Models](https://mlsanity.com/paper/2310.01329/) (2023)
- [Towards Efficient Cross-Modal Visual Textual Retrieval using Transformer-Encoder Deep Features](https://mlsanity.com/paper/2106.00358/) (2021)
- [Composite Code Sparse Autoencoders for first stage retrieval](https://mlsanity.com/paper/2204.07023/) (2022)
- [Loss Function Entropy Regularization for Diverse Decision Boundaries](https://mlsanity.com/paper/2205.00224/) (2022)
- [LNPT: Label-free Network Pruning and Training](https://mlsanity.com/paper/2403.12690/) (2024)
- [Uncertainty-Aware Spatial Color Correlation for Low-Light Image Enhancement](https://mlsanity.com/paper/2508.04176/) (2025)
- [Model Inversion Attack Against Deep Hashing](https://mlsanity.com/paper/2511.12233/) (2025)

## Abstract (arXiv)

For large-scale visual search, highly compressed yet meaningful representations of images are essential. Structured vector quantizers based on product quantization and its variants are usually employed to achieve such compression while minimizing the loss of accuracy. Yet, unlike binary hashing schemes, these unsupervised methods have not yet benefited from the supervision, end-to-end learning and novel architectures ushered in by the deep learning revolution. We hence propose herein a novel method to make deep convolutional neural networks produce supervised, compact, structured binary codes for visual search. Our method makes use of a novel block-softmax non-linearity and of batch-based entropy losses that together induce structure in the learned encodings. We show that our method outperforms state-of-the-art compact representations based on deep hashing or structured quantization in single and cross-domain category retrieval, instance retrieval and classification. We make our code and models publicly available online.