Lost in quantization: Improving particular object retrieval in large scale image databases

The state of the art in visual object retrieval from large databases is achieved by systems that are inspired by text retrieval. A key component of these approaches is that local regions of images are characterized using high-dimensional descriptors which are then mapped to ldquovisual wordsrdquo selected from a discrete vocabulary.This paper explores techniques to map each visual region to a weighted set of words, allowing the inclusion of features which were lost in the quantization stage of previous systems. The set of visual words is obtained by selecting words based on proximity in descriptor space. We describe how this representation may be incorporated into a standard tf-idf architecture, and how spatial verification is modified in the case of this soft-assignment. We evaluate our method on the standard Oxford Buildings dataset, and introduce a new dataset for evaluation. Our results exceed the current state of the art retrieval performance on these datasets, particularly on queries with poor initial recall where techniques like query expansion suffer. Overall we show that soft-assignment is always beneficial for retrieval with large vocabularies, at a cost of increased storage requirements for the index.

• Publication year

  2008

• Venue

  2008 IEEE Conference on Computer Vision and Pattern Recognition

• Publication date

  2008-06-01

• Fields of study

  Computer Science

• Identifiers
  DOI 10.1109/CVPR.2008.4587635
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Object retrieval with large vocabularies and fast spatial matching

  2007influential reference
• A contextual dissimilarity measure for accurate and efficient image search

  2007cited by this paper
• Total Recall: Automatic Query Expansion with a Generative Feature Model for Object Retrieval

  2007influential reference
• Vector Quantizing Feature Space with a Regular Lattice

  2007influential reference
• Scalable Recognition with a Vocabulary Tree

  2006influential reference
• Pattern Recognition and Machine Learning

  2006cited by this paper
• Distinctive Image Features from Scale-Invariant Keypoints

  2004influential reference
• Scale & Affine Invariant Interest Point Detectors

  2004cited by this paper
• Robust wide-baseline stereo from maximally stable extremal regions

  2004cited by this paper
• Point matching as a classification problem for fast and robust object pose estimation

  2004cited by this paper
• Localisation using an image-map

  2004cited by this paper
• Web Search for a Planet: The Google Cluster Architecture

  2003cited by this paper
• Video Google: a text retrieval approach to object matching in videos

  2003influential reference
• Multi-tier architecture for Web search engines

  2003cited by this paper
• Vector quantization and signal compression

  1991influential reference

• A comprehensive survey of content based image retrieval schemes: advancements, challenges, and future directions

  2026cites this paper
• Particle Diffusion Matching: Random Walk Correspondence Search for the Alignment of Standard and Ultra-Widefield Fundus Images

  2026cites this paper
• Unsupervised Cross-Domain Deep-Fused Feature Descriptor for Efficient Image Retrieval

  2025cites this paper
• Global-to-Local or Local-to-Global? Enhancing Image Retrieval with Efficient Local Search and Effective Global Re-Ranking

  2025cites this paper
• LoCoRe: Image Re-Ranking with Long-Context Sequence Modeling

  2025influential citation
• MegaLoc: One Retrieval to Place Them All

  2025cites this paper
• UAVPairs: A benchmark for match pair retrieval of large-scale UAV images

  2025cites this paper
• ILIAS: Instance-Level Image retrieval At Scale

  2025cites this paper
• Act-retrieval: a framework for action-based image retrieval

  2025cites this paper
• Exploiting Hu invariant moments and deep features for image retrieval

  2025cites this paper
• Auto-regressive transformation for image alignment

  2025cites this paper
• Hypergraph Vision Transformers: Images are More than Nodes, More than Edges

  2025cites this paper
• UAVPairs: A Challenging Benchmark for Match Pair Retrieval of Large-scale UAV Images

  2025cites this paper
• Adversarially Robust CLIP Models Can Induce Better (Robust) Perceptual Metrics

  2025cites this paper
• Exploiting Deep Contrast Feature for Image Retrieval

  2025cites this paper
• Improved Bag-of-Words Image Retrieval with Geometric Constraints for Ground Texture Localization

  2025influential citation
• Locality Preserving Markovian Transition for Instance Retrieval

  2025influential citation
• Instance-Level Generation for Representation Learning

  2025cites this paper
• Automatic Monument Detection: Leveraging Deep Learning for Cultural Heritage Applications

  2025cites this paper
• Patch Based Unsupervised Deep Learning for Retrieval and Part Matching in Large Image Datasets

  2025cites this paper
• Class Agnostic Instance-level Descriptor for Visual Instance Search

  2025cites this paper
• MCDINO: Self-supervised learning of masks based on combination of multi-path channel attention and local feature weighting

  2025cites this paper
• Image retrieval using deep saliency edge feature

  2025cites this paper
• A Novel Deep Learning-Based Framework for Semantic Image Retrieval Using Multi-Scale Attention and Contrastive Learning

  2024cites this paper
• Survey of Semantic-Based Image-to-Image Retrieval

  2024cites this paper
• ULTRON: Unifying Local Transformer and Convolution for Large-Scale Image Retrieval

  2024cites this paper
• Image retrieval using compact deep semantic correlation descriptors

  2024cites this paper
• Object Retrieval for Visual Question Answering with Outside Knowledge

  2024cites this paper
• Anomaly Detection on the Edge Using Smart Cameras under Low-Light Conditions

  2024cites this paper
• Efficient Approximation of Earth Mover’s Distance Based on Nearest Neighbor Search

  2024cites this paper
• INQUIRE: A Natural World Text-to-Image Retrieval Benchmark

  2024cites this paper
• AMES: Asymmetric and Memory-Efficient Similarity Estimation for Instance-level Retrieval

  2024cites this paper
• Image Retrieval Using Multilayer Feature Aggregation Histogram

  2024cites this paper
• Towards Unified Benchmark and Models for Multi-Modal Perceptual Metrics

  2024cites this paper
• Content-based image retrieval through fusion of deep features extracted from segmented neutrosophic using depth map

  2024cites this paper
• Adaptive Weighting Feature Aggregation using Particle Swarm Optimization for Image Retrieval

  2024cites this paper
• DINO as a von Mises-Fisher mixture model

  2024cites this paper
• Cluster-Aware Similarity Diffusion for Instance Retrieval

  2024cites this paper
• Correlation Verification for Image Retrieval and Its Memory Footprint Optimization

  2024cites this paper
• MeshVPR: Citywide Visual Place Recognition Using 3D Meshes

  2024cites this paper
• Component-Level Oracle Bone Inscription Retrieval

  2024cites this paper
• Adaptive Feature Refinement and Weighted Similarity for Deep Loop Closure Detection in Appearance Variation

  2024cites this paper
• Automated Near-Duplicate Image Detection using Sparrow Search Algorithm with Deep Learning Model

  2024cites this paper
• Automatic Data Curation for Self-Supervised Learning: A Clustering-Based Approach

  2024cites this paper
• Image retrieval by aggregating deep orientation structure features

  2024cites this paper
• Low-Light Outdoor Smoke Detection Based on Unsupervised Pre-Training

  2024cites this paper
• Efficient Image Retrieval Using Hierarchical K-Means Clustering

  2024cites this paper
• Structure Similarity Preservation Learning for Asymmetric Image Retrieval

  2024cites this paper
• MoReSo: A DNN Framework Expediting Content-Based Video Image Retrieval (CBVIR)

  2024cites this paper
• Evolved Hierarchical Masking for Self-Supervised Learning

  2024cites this paper
• Exploiting deep cross-semantic features for image retrieval

  2024cites this paper
• On Train-Test Class Overlap and Detection for Image Retrieval

  2024cites this paper
• Gen-SIS: Generative Self-augmentation Improves Self-supervised Learning

  2024cites this paper
• Locating Target Regions for Image Retrieval in an Unsupervised Manner

  2024cites this paper
• Where's Waldo: Diffusion Features For Personalized Segmentation and Retrieval

  2024cites this paper
• Attention-Guided Masked Autoencoders for Learning Image Representations

  2024influential citation
• Multi-scale pyramidal hash learning for traditional building facade image retrieval

  2024cites this paper
• Image retrieval using underlying importance feature histogram

  2024cites this paper
• Semantic-aware Representation Learning for Homography Estimation

  2024cites this paper
• Unsupervised affinity learning based on manifold analysis for image retrieval: A survey

  2024cites this paper
• You Only Demanufacture Once (Yodo): WEEE Retrieval Using Unsupervised Learning

  2023cites this paper
• Module-based multiple feature integration descriptor for image retrieval

  2023cites this paper
• Let All be Whitened: Multi-teacher Distillation for Efficient Visual Retrieval

  2023cites this paper
• Unsupervised Deep-Embedding Global Feature Descriptor for Image Retrieval

  2023cites this paper
• FoundPose: Unseen Object Pose Estimation with Foundation Features

  2023cites this paper
• BEDOI: BENCHMARKS FOR DETERMINING OVERLAPPING IMAGES WITH PHOTOGRAMMETRIC INFORMATION

  2023cites this paper
• Patch-wise self-supervised visual representation learning: a fine-grained approach

  2023cites this paper
• Adapting Video Instance Segmentation for Instance Search

  2023cites this paper
• Battle of the Backbones: A Large-Scale Comparison of Pretrained Models across Computer Vision Tasks

  2023cites this paper
• Modelling of Firefly Algorithm with Densely Connected Networks for Near-Duplicate Image Detection System

  2023cites this paper
• ClusVPR: Efficient Visual Place Recognition With Clustering-Based Weighted Transformer

  2023cites this paper
• The FASHION Visual Search using Deep Learning Approach

  2023cites this paper
• Topological RANSAC for instance verification and retrieval without fine-tuning

  2023cites this paper
• Visual Passage Score Aggregation for Image Retrieval

  2023cites this paper
• Enhancing Landmark Detection in Cluttered Real-World Scenarios with Vision Transformers

  2023influential citation
• Towards Content-based Pixel Retrieval in Revisited Oxford and Paris

  2023cites this paper
• FORB: A Flat Object Retrieval Benchmark for Universal Image Embedding

  2023cites this paper
• Multi-View Diffusion Process for Spectral Clustering and Image Retrieval

  2023cites this paper
• Revisiting Self-Similarity: Structural Embedding for Image Retrieval

  2023cites this paper
• Coarse-to-Fine: Learning Compact Discriminative Representation for Single-Stage Image Retrieval

  2023cites this paper
• Exploring Last Mile Drone Logistics in Urban/Rural Areas: A Viability Study

  2023cites this paper
• Regional attention based multi-scale feature fusion for image retrieval

  2023cites this paper
• Recent advance in image search

  2023cites this paper
• A Multi-feature Integration Descriptor for Instance Image Retrieval

  2023influential citation
• Data-efficient representation learning for visual place recognition

  2023cites this paper
• AST-MHSA : Code Summarization using Multi-Head Self-Attention

  2023cites this paper
• Content-based image retrieval with fuzzy clustering for feature vector normalization

  2023cites this paper
• Dartmouth Digital

  2023cites this paper
• Keep It SimPool:Who Said Supervised Transformers Suffer from Attention Deficit?

  2023cites this paper
• Survey of Action Recognition, Spotting and Spatio-Temporal Localization in Soccer - Current Trends and Research Perspectives

  2023cites this paper
• Image Retrieval Using Object Semantic Aggregation Histogram

  2023cites this paper
• CityUPlaces: a new dataset for efficient vision-based recognition

  2023cites this paper
• Aggregating Deep Features of Multi-CNN Models for Image Retrieval

  2023cites this paper
• Unsupervised Hashing via Similarity Distribution Calibration

  2023cites this paper
• Global Features are All You Need for Image Retrieval and Reranking

  2023cites this paper
• Learning Visual Representation Clusters for Cross-View Geo-Location

  2023cites this paper
• SoftCLIP: Softer Cross-modal Alignment Makes CLIP Stronger

  2023cites this paper
• Are Local Features All You Need for Cross-Domain Visual Place Recognition?

  2023cites this paper
• Universal Image Embedding: Retaining and Expanding Knowledge With Multi-Domain Fine-Tuning

  2023cites this paper
• Deep Learning for Camera Calibration and Beyond: A Survey

  2023cites this paper