# Learning Spread-out Local Feature Descriptors

arXiv: https://arxiv.org/abs/1708.06320 · HTML: https://mlsanity.com/paper/1708.06320/

## Key facts

- Cited by 34 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 21 Aug 2017.

## Built on by (most cited first, 30 of 34)

- [EmbeddingGemma: Powerful and Lightweight Text Representations](https://mlsanity.com/paper/2509.20354/) (2025)
- [Training Vision Transformers for Image Retrieval](https://mlsanity.com/paper/2102.05644/) (2021)
- [Are All Losses Created Equal: A Neural Collapse Perspective](https://mlsanity.com/paper/2210.02192/) (2022)
- [jina-embeddings-v5-text: Task-Targeted Embedding Distillation](https://mlsanity.com/paper/2602.15547/) (2026)
- [Evaluating Multi-label Classifiers with Noisy Labels](https://mlsanity.com/paper/2102.08427/) (2021)
- [ELM: Embedding and Logit Margins for Long-Tail Learning](https://mlsanity.com/paper/2204.13208/) (2022)
- [Mixed Federated Learning: Joint Decentralized and Centralized Learning](https://mlsanity.com/paper/2205.13655/) (2022)
- [LoOp: Looking for Optimal Hard Negative Embeddings for Deep Metric Learning](https://mlsanity.com/paper/2108.09335/) (2021)
- [TurboSVM-FL: Boosting Federated Learning through SVM Aggregation for Lazy Clients](https://mlsanity.com/paper/2401.12012/) (2024)
- [Deep Learning Reforms Image Matching: A Survey and Outlook](https://mlsanity.com/paper/2506.04619/) (2025)
- [Learning Federated Representations and Recommendations with Limited Negatives](https://mlsanity.com/paper/2108.07931/) (2021)
- [SDGMNet: Statistic-based Dynamic Gradient Modulation for Local Descriptor Learning](https://mlsanity.com/paper/2106.04434/) (2021)
- [RobustMat: Neural Diffusion for Street Landmark Patch Matching under Challenging Environments](https://mlsanity.com/paper/2311.03904/) (2023)
- [Image Patch-Matching with Graph-Based Learning in Street Scenes](https://mlsanity.com/paper/2311.04617/) (2023)
- [Transforming LLMs into Cross-modal and Cross-lingual Retrieval Systems](https://mlsanity.com/paper/2404.01616/) (2024)
- [Encouraging Intra-Class Diversity Through a Reverse Contrastive Loss for Better Single-Source Domain Generalization](https://mlsanity.com/paper/2106.07916/) (2021)
- [Get the Best of Both Worlds: Improving Accuracy and Transferability by Grassmann Class Representation](https://mlsanity.com/paper/2308.01547/) (2023)
- [Loss Function Entropy Regularization for Diverse Decision Boundaries](https://mlsanity.com/paper/2205.00224/) (2022)
- [Registering Neural Radiance Fields as 3D Density Images](https://mlsanity.com/paper/2305.12843/) (2023)
- [Contrastive Learning for Conversion Rate Prediction](https://mlsanity.com/paper/2307.05974/) (2023)
- [Relational Representation Learning Network for Cross-Spectral Image Patch Matching](https://mlsanity.com/paper/2403.11751/) (2024)
- [Why and How: Knowledge-Guided Learning for Cross-Spectral Image Patch Matching](https://mlsanity.com/paper/2412.11161/) (2024)
- [MoganBert-TR: A Turkish Encoder Foundation Model Trained from Scratch with a CLM-to-MLM Curriculum](https://mlsanity.com/paper/2608.25768/) (2026)
- [Dissecting the impact of different loss functions with gradient surgery](https://mlsanity.com/paper/2201.11307/) (2022)
- [Grid-Reg: Detector-Free Gridized Feature Learning and Matching for Large-Scale SAR-Optical Image Registration](https://mlsanity.com/paper/2507.04233/) (2025)
- [SVMax: A Feature Embedding Regularizer](https://mlsanity.com/paper/2103.02770/) (2021)
- [Attention-Based Multimodal Image Matching](https://mlsanity.com/paper/2103.11247/) (2021)
- [Blind Speech Separation and Dereverberation using Neural Beamforming](https://mlsanity.com/paper/2103.13443/) (2021)
- [A Structure Feature Algorithm for Multi-modal Forearm Registration](https://mlsanity.com/paper/2111.05485/) (2021)
- [GRM: Gradient Rectification Module for Visual Place Retrieval](https://mlsanity.com/paper/2204.10972/) (2022)

## Abstract (arXiv)

We propose a simple, yet powerful regularization technique that can be used to significantly improve both the pairwise and triplet losses in learning local feature descriptors. The idea is that in order to fully utilize the expressive power of the descriptor space, good local feature descriptors should be sufficiently "spread-out" over the space. In this work, we propose a regularization term to maximize the spread in feature descriptor inspired by the property of uniform distribution. We show that the proposed regularization with triplet loss outperforms existing Euclidean distance based descriptor learning techniques by a large margin. As an extension, the proposed regularization technique can also be used to improve image-level deep feature embedding.