Contents Menu Expand Light mode Dark mode Auto light/dark, in light mode Auto light/dark, in dark mode Skip to content
pyvisim 0.10.0
Light Logo Dark Logo
pyvisim 0.10.0

Getting started

  • Introduction

API reference

  • Classic embedders
    • VLADEmbedder
    • FisherVectorEmbedder
    • Pipeline
  • Features
    • Handcrafted features
      • SIFT
      • RootSIFT
      • Lambda
    • Deep learning based features
      • DeepConvFeature
  • Image Similarity Retrieval
    • Image store
      • In Memory Image Embedding Store
      • External Search Index
    • Reranking
      • K-Reciprocal Reranking
  • Dataset
    • OxfordFlowerDataset
  • Neural networks
    • ContrastiveSiameseNetwork
    • BCESiameseNetwork
    • TripletNeuralNetwork
    • ClipEmbedder
    • Backbones
  • Distance metrics
  • Dense
    • Structural
      • SSIM
      • MSSSIM
    • Pixelwise
      • PSNR
  • Typing
  • Evaluation functions

Tutorials

  • 1 Introduction
    • 1.1 pyvisim Introduction
  • 2 Classical methods
    • 2.1 Oxford Flower VLAD and Fisher Vector Retrieval Demo
    • 2.2 Pipeline with Deep Features
    • 2.3 Custom Feature Extractor with ORB
  • 3 Metric learning methods
    • 3.1 Siamese Neural Network
    • 3.2 Triplet Neural Network
  • 4 Image similarity search
    • 4.1 Image Search
    • 4.2 Computing Mean Average Precision (mAP) and Top-k Accuracy for our Retrieval System
    • 4.3 Oxford Flowers Clustering Notebook

Release notes

  • Release notes
    • Unreleased
    • v0.10.0
    • v0.9.5
    • v0.9.4
    • Changelog
Back to top
View this page

4 Image similarity search¶

Tutorial

Description

Link

Image search

This is the most interesting tutorial of this library, so I recommend that you definitely check this out 😊

Build an image gallery, index the embeddings and retrieve the most similar images for a query using different enhancement tricks such as reranking and query expansion.

Image Search

Performance accuracy of retrieval pipeline

Evaluation of the mean Average Precision (mAP) of the retrieval pipeline, composing the Clip Embedder as the embedding model.

Retrieval Evaluation

Image Clustering

Show the clustering performance via metrics like RI, ARI, NMI using embeddings generated by the Clip Embedder.

Clustering Images

Next
4.1 Image Search
Previous
3.2 Triplet Neural Network
Copyright © 2026, Nhật Huy Vũ
Made with Sphinx and @pradyunsg's Furo