# Unifying Deep Local and Global Features for Image Search

arXiv: https://arxiv.org/abs/2001.05027 · HTML: https://mlsanity.com/paper/2001.05027/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 14 Jan 2020.

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

- [Where is your place, Visual Place Recognition?](https://mlsanity.com/paper/2103.06443/) (2021)
- [Deep Learning for Instance Retrieval: A Survey](https://mlsanity.com/paper/2101.11282/) (2021)
- [Large-scale Localization Datasets in Crowded Indoor Spaces](https://mlsanity.com/paper/2105.08941/) (2021)
- [Deep Multi-Task Learning for Joint Localization, Perception, and Prediction](https://mlsanity.com/paper/2101.06720/) (2021)
- [Inside Out Visual Place Recognition](https://mlsanity.com/paper/2111.13546/) (2021)
- [Re-rank Coarse Classification with Local Region Enhanced Features for Fine-Grained Image Recognition](https://mlsanity.com/paper/2102.09875/) (2021)
- [Efficient large-scale image retrieval with deep feature orthogonality and Hybrid-Swin-Transformers](https://mlsanity.com/paper/2110.03786/) (2021)
- [Unifying Remote Sensing Image Retrieval and Classification with Robust Fine-tuning](https://mlsanity.com/paper/2102.13392/) (2021)
- [Instance Image Retrieval by Learning Purely From Within the Dataset](https://mlsanity.com/paper/2208.06119/) (2022)
- [Connecting Images through Time and Sources: Introducing Low-data, Heterogeneous Instance Retrieval](https://mlsanity.com/paper/2103.10729/) (2021)
- [Supervised Fine-tuning Evaluation for Long-term Visual Place Recognition](https://mlsanity.com/paper/2211.07696/) (2022)
- [Region-Aware CLS Token Augmentation for Fine-Grained Image Retrieval](https://mlsanity.com/paper/2610.10991/) (2026)

## Abstract (arXiv)

Image retrieval is the problem of searching an image database for items that are similar to a query image. To address this task, two main types of image representations have been studied: global and local image features. In this work, our key contribution is to unify global and local features into a single deep model, enabling accurate retrieval with efficient feature extraction. We refer to the new model as DELG, standing for DEep Local and Global features. We leverage lessons from recent feature learning work and propose a model that combines generalized mean pooling for global features and attentive selection for local features. The entire network can be learned end-to-end by carefully balancing the gradient flow between two heads -- requiring only image-level labels. We also introduce an autoencoder-based dimensionality reduction technique for local features, which is integrated into the model, improving training efficiency and matching performance. Comprehensive experiments show that our model achieves state-of-the-art image retrieval on the Revisited Oxford and Paris datasets, and state-of-the-art single-model instance-level recognition on the Google Landmarks dataset v2. Code and models are available at https://github.com/tensorflow/models/tree/master/research/delf .