Spaces:
Runtime error
Runtime error
| <html xmlns="http://www.w3.org/1999/xhtml"> | |
| <link href="https://fonts.cdnfonts.com/css/chalkduster" rel="stylesheet"> | |
| <style> | |
| @import url('https://fonts.cdnfonts.com/css/chalkduster'); | |
| </style> | |
| <!-- <script src="./sm/assets/teaser-data.js"></script> --> | |
| <head> | |
| <meta http-equiv="Content-Type" content="text/html; charset=UTF-8"> | |
| <title>Image Retrieval</title> | |
| <link href="style.css" rel="stylesheet" type="text/css"> | |
| <meta name="description" | |
| content="Project page for 'Composite Sketch+Text Queries for Retrieving Objects with Elusive Names and Complex Interactions.'"> | |
| <!-- <link rel="icon" href="./pics/wis_logo.jpg"> --> | |
| <link rel="stylesheet" href="https://cdn.jsdelivr.net/npm/bootstrap@4.0.0/dist/css/bootstrap.min.css" | |
| integrity="sha384-Gn5384xqQ1aoWXA+058RXPxPg6fy4IWvTNh0E263XmFcJlSAwiGgFAW/dAiS6JXm" crossorigin="anonymous"> | |
| <meta http-equiv="Content-Type" content="text/html; charset=UTF-8"> | |
| <title>Image Retrieval</title> | |
| <link href="styles/style.css" rel="stylesheet" type="text/css"> | |
| <meta name="description" | |
| content="Project page for 'Composite Sketch+Text Queries for Retrieving Objects with Elusive Names and Complex Interactions.'"> | |
| <meta charset="utf-8"> | |
| <meta name="description" content="Composite Sketch+Text Queries for Retrieving Objects with Elusive Names and Complex Interactions"> | |
| <meta name="keywords" content="sketch+text-based image retrieval, cross-modal retrieval, image retrieval, SBIR, CSTBIR;"> | |
| <meta name="author" content="Prajwal Gatti"> | |
| <title>Image Retrieval</title> | |
| <meta name="twitter:card" content="summary_large_image" /> | |
| <meta name="twitter:title" content="[AAAI 2024] Composite Sketch+Text Queries for Retrieving Objects with Elusive Names and Complex Interactions" /> | |
| <meta name="twitter:description" content="[AAAI 2024] Composite Sketch+Text Queries for Retrieving Objects with Elusive Names and Complex Interactions" /> | |
| <meta name="twitter:image:alt" content="CSTBIR (AAAI 2024)" /> | |
| <link rel="stylesheet" href="https://maxcdn.bootstrapcdn.com/font-awesome/4.7.0/css/font-awesome.min.css"> | |
| </head> | |
| <body> | |
| <p class="title">Image Retrieval</p> | |
| <div style="text-align: center; font-size: 15pt; margin-bottom: 15px"> | |
| <span><a href="https://aaai.org/aaai-conference/">CSL2050 Course Project 2024 remianing</a></span> | |
| </div> | |
| <p class="author"> | |
| <span class="author"><a href="https://github.com/Jay18Mehta"> Jay Mehta</a></span> | |
| <span class="author"><a href="https://github.com/akshatb22me007">Akshat Jain</a></span> | |
| <span class="author"><a href="https://github.com/harshivshah2504">Harshiv Shah</a></span> | |
| <span class="author"><a href="https://github.com/gjyotin305">Jyotin Goel</a></span> | |
| <span class="author"><a href="https://github.com/RHYTHM2405">Rhythm Baghel</a></span> | |
| </p> | |
| <table border="0" align="center" class="affiliations" width="700px"> | |
| <tbody align="center"> | |
| <tc> | |
| <td style="text-align: center;">IIT Jodhpur</td> | |
| </tc> | |
| </tbody> | |
| </table> | |
| <table width="999" border="0" align="center" class="menu" style="margin-bottom: 8px;"> | |
| <tbody> | |
| <tr> | |
| <td align="center">| <a href="./resources/paper.pdf">Report link remaining</a> | <a href="https://github.com/gjyotin305/CSL2050_CourseProject">Github Link</a> | <a href="https://www.cs.toronto.edu/~kriz/cifar.html">Cifar-10 Dataset</a> | <a href="./resources/CSTBIR-AAAI24Poster.pdf">Youtube link remaining</a> </td> | |
| </tr> | |
| </tbody> | |
| </table> | |
| <div class="container"> | |
| <table width="1000" border="0" align="center"> | |
| <tbody> | |
| <tr> | |
| <div id="imageCarousel" class="carousel slide teaser-carousel" data-ride="carousel" data-interval="5000"> | |
| <div class="carousel-inner"> | |
| <div class="carousel-item active"> | |
| <div> | |
| <img class="teaser-img" src="images/harshiv.jpg" height="390" /> | |
| </div> | |
| </div> | |
| <div class="carousel-item"> | |
| <div> | |
| <img class="teaser-img" src="images/truck.jpg" height="390" /> | |
| </div> | |
| </div> | |
| <div class="carousel-item"> | |
| <div> | |
| <img class="teaser-img" src="images/plane.jpg" height="390" /> | |
| </div> | |
| </div> | |
| <div class="carousel-item"> | |
| <div> | |
| <img class="teaser-img" src="images/horse.jpg" height="390" /> | |
| </div> | |
| </div> | |
| </div> | |
| <a class="carousel-control-prev" href="#imageCarousel" role="button" data-slide="prev"> | |
| <div class="slider-navigation-previous"> | |
| <svg viewBox="0 0 50 80" xml:space="preserve"> | |
| <polyline fill="white" stroke-width=".5em" stroke-linecap="round" | |
| stroke-linejoin="round" points="45.63,75.8 0.375,38.087 45.63,0.375 "> | |
| </polyline> | |
| </svg> | |
| </div> | |
| <span class="sr-only">Previous</span> | |
| </a> | |
| <a class="carousel-control-next" href="#imageCarousel" role="button" data-slide="next"> | |
| <div class="slider-navigation-next"> | |
| <svg viewBox="0 0 50 80" xml:space="preserve"> | |
| <polyline fill="white" stroke-width=".5em" stroke-linecap="round" | |
| stroke-linejoin="round" points="0.375,0.375 45.63,38.087 0.375,75.8 "> | |
| </polyline> | |
| </svg> | |
| </div> | |
| <span class="sr-only">Next</span> | |
| </a> | |
| </div> | |
| </tr> | |
| <!-- <tr> <br /> </tr> --> | |
| <tr align="center"></tr> | |
| </tbody> | |
| </table> | |
| <div style="text-align: justify;text-justify: inter-word;"> | |
| <p><span class="section"><b>Abstract</b></span> </p> | |
| <p>This project focuses on image retrieval, aiming to retrieve relevant images given an image query. Leveraging both Histogram of Oriented Gradients (HoG) and Convolutional Neural Network (CNN) features extracted through provided implementations, various methodologies are explored, including classification and clustering and mean based techniques. The CIFAR-10 dataset serves as the foundation for this study, offering a diverse collection of images for training and evaluation purposes. Through systematic experimentation and analysis, this project seeks to enhance image retrieval systems, contributing to advancements in image recognition and retrieval technologies..</p> | |
| </div> | |
| <!-- <div style="text-align: justify;text-justify: inter-word;"> | |
| <p><b>Keywords:</b> sketch+text-based image retrieval, cross-modal retrieval, image retrieval, SBIR, CSTBIR</p> | |
| </div> --> | |
| <p class="section"> </p> | |
| <p class="section"><b>The Image Retrieval Problem</b></p> | |
| <div style="text-align: center"> | |
| <img src="./images/image2image.png" alt="" width="700px" style="margin: auto" /> | |
| <br><br><p>In today's digital age, the vast amount of image data available on the internet poses a significant challenge in efficiently retrieving relevant images based on user queries. To address this challenge, the task at hand is to develop an Image Retrieval System capable of retrieving relevant images given an image query.</p> | |
| </div> | |
| <p class="section"> </p> | |
| <!-- <p class="section"><b>Sketches in CSTBIR</b></p> | |
| <div style="text-align: center"> | |
| <img src="./resources/supp_sketch_examples-min.png" alt="" width="700px" style="margin: auto" /><br> | |
| <br><br><p>Examples of sketches.</p> | |
| </div> | |
| <p class="section"> </p> --> | |
| <p class="section"><b>Youtube Video</b></p> | |
| <centre> | |
| <iframe width="667" height="500" src="https://www.youtube.com/watch?v=bciby07CSLA"> | |
| </iframe> | |
| </centre> | |
| <p class="section"> </p> | |
| <div class="papers-list"> | |
| <p class="section" id="related-paper"><b>Our Approaches</b></p> | |
| <ul> | |
| <li> | |
| <h5>KNN:</h5> Utilizes similarity of feature vectors for classification. | |
| Without additional feature extraction techniques, KNN operates directly on raw pixel values or basic features. | |
| </li><br> | |
| <li> | |
| <h5>PCA+KNN:</h5> rincipal Component Analysis (PCA) reduces dimensionality to capture significant variations.By reducing the dimensionality of the feature space, PCA aims to capture the most significant variations in the data. | |
| </li><br> | |
| <li> | |
| <h5>Resnet50 + KNN:</h5> Used the pretrained Resnet50, along with 3 additional 3 hidden linear layers to reduce the output | |
| layers size from 1000 to 10. Trained the above model on CIFAR-10 dataset to fine the model to give required output | |
| </li><br> | |
| <li> | |
| <h5>Resnet-18 + KNN:</h5> Resnet-18 for feature extraction of training images and testing images. Next used KNN on extracted features, which gives K nearest images from training data as output. This KNN modal has 72 percent accuracy on testing accuracy. Also, wherever the images are retrieved of the wrong class, these wrong class images are internally similar to the original class. For example, If the input image is of a cat, the output images will contain mostly cats with some other animals like dogs or horses, but any vehicle won't be output. | |
| </li><br> | |
| <li> | |
| <h5>ANN+KNN:</h5> First trained ANN with ReLU activation function and SGD optimizer to classify input images. Extracted images from training data which has the same class as the predicted class. Applied KNN to find Similar images. Ann has 52 percent accuracy, so retrieved images will have 52 percent accuracy as well. | |
| </li><br> | |
| </ul> | |
| </div> | |
| <p class="section"> </p> | |
| <div class="team"> | |
| <p class="section" id="team"><b>Team</b></p> | |
| <div class="row justify-content-center"> | |
| <div class="col-xs-12 col-sm-6 col-md-6 col-lg-2"> | |
| <div class="mainflip"> | |
| <div class="card"> | |
| <div class="card-body text-center card-no-padding"> | |
| <p><img src="images/jay.jpeg" alt="card image" height="130" width="100"></p> | |
| <h5 class="card-title">Jay <br> Mehta</h5> | |
| <ul class="list-inline"> | |
| <li class="list-inline-item"> | |
| <a class="social-icon text-xs-center" target="_blank" href="https://www.linkedin.com/in/jay-mehta-86a745258/"> | |
| <i class="fa fa-linkedin"></i> | |
| </a> | |
| </li> | |
| <li class="list-inline-item"> | |
| <a class="social-icon text-xs-center" target="_blank" href="https://jay18mehta.github.io/Portfolio/"> | |
| <i class="fa fa-user"></i> | |
| </a> | |
| </li> | |
| </ul> | |
| </div> | |
| </div> | |
| </div> | |
| </div> | |
| <div class="col-xs-12 col-sm-6 col-md-6 col-lg-2"> | |
| <div class="mainflip"> | |
| <div class="card"> | |
| <div class="card-body text-center card-no-padding"> | |
| <p><img src="images/akshat.jpeg" alt="card image" height="130" width="100"></p> | |
| <h5 class="card-title">Akshat <br> Jain</h5> | |
| <ul class="list-inline"> | |
| <li class="list-inline-item"> | |
| <li class="list-inline-item"> | |
| <a class="social-icon text-xs-center" target="_blank" href="https://www.linkedin.com/in/akshat-jain-2bba94280/"> | |
| <i class="fa fa-linkedin"></i> | |
| </a> | |
| </li> | |
| </a> | |
| </li> | |
| </ul> | |
| </div> | |
| </div> | |
| </div> | |
| </div> | |
| <div class="col-xs-12 col-sm-6 col-md-6 col-lg-2"> | |
| <div class="mainflip"> | |
| <div class="card"> | |
| <div class="card-body text-center card-no-padding"> | |
| <p><img src="images/dog.jpg" alt="card image" height="130" width="100"></p> | |
| <h5 class="card-title">Harshiv Shah</h5> | |
| <ul class="list-inline"> | |
| <li class="list-inline-item"> | |
| <a class="social-icon text-xs-center" target="_blank" href="https://www.linkedin.com/in/harshivshah27?utm_source=share&utm_campaign=share_via&utm_content=profile&utm_medium=android_app"> | |
| <i class="fa fa-linkedin"></i> | |
| </a> | |
| </li> | |
| </ul> | |
| </div> | |
| </div> | |
| </div> | |
| </div> | |
| <div class="col-xs-12 col-sm-6 col-md-6 col-lg-2"> | |
| <div class="mainflip"> | |
| <div class="card"> | |
| <div class="card-body text-center card-no-padding"> | |
| <p><img src="images/rhythm.jpeg" alt="card image" height="130" width="100"></p> | |
| <h5 class="card-title">Rhythm <br> Baghel</h5> | |
| <ul class="list-inline"> | |
| <li class="list-inline-item"> | |
| <a class="social-icon text-xs-center" target="_blank" href="https://www.linkedin.com/in/rhythm-baghel-80675a25a/"> | |
| <i class="fa fa-linkedin"></i> | |
| </a> | |
| </li> | |
| </ul> | |
| </div> | |
| </div> | |
| </div> | |
| </div> | |
| <div class="col-xs-12 col-sm-6 col-md-6 col-lg-2"> | |
| <div class="mainflip"> | |
| <div class="card"> | |
| <div class="card-body text-center card-no-padding"> | |
| <p><img src="images/jyotin.jpeg" alt="card image" height="130" width="100"></p> | |
| <h5 class="card-title">Jyotin <br> Goel</h5> | |
| <ul class="list-inline"> | |
| <li class="list-inline-item"> | |
| <a class="social-icon text-xs-center" target="_blank" href="https://www.linkedin.com/in/jyotin-goel-16924b263/"> | |
| <i class="fa fa-linkedin"></i> | |
| </a> | |
| </li> | |
| </ul> | |
| </div> | |
| </div> | |
| </div> | |
| </div> | |
| </div> | |
| <p class="section"> </p> | |
| <p class="section"> </p> | |
| <p class="section"> </p> | |
| <div class="ack"> | |
| <p class="section" id="ack"><b>Acknowledgment</b></p> | |
| We are very grateful to Dr Anand Mishra for providing us with this opportunity to work on this project. This project helped us to explore many different techniques for image retrieval which helped us strengthen our basics and learn some of the advanced concepts. We were able to get a hands on experience while working on this project which was very beneficial for all of us. | |
| </div> | |
| <p class="section"></p> | |
| <p class="section"> </p> | |
| <div class="contact"> | |
| <p class="section" id="related-paper"><b>Contact</b></p> | |
| For questions, please contact <a href="https://github.com/gjyotin305" target="_blank">Jyotin Goel</a> or raise an issue on <a class="publink" href="https://github.com/gjyotin305/CSL2050_CourseProject" target="_blank" style="text-decoration: none">GitHub</a>. | |
| </div> | |
| <p class="section"></p> | |
| <!-- Footer --> | |
| <footer> | |
| <p class="m-0 text-center">Copyright © This is an open source project.</p> | |
| </footer> | |
| </body> | |
| <script> | |
| function prompt_on(prompt_element) { | |
| prompt_element.classList.add("caption-active"); | |
| } | |
| function prompt_off(prompt_element) { | |
| prompt_element.classList.remove("caption-active"); | |
| } | |
| function toggle_prompt(active_prompt_id, inactive_prompt_ids, result_id) { | |
| let active_prompt = document.getElementById(active_prompt_id); | |
| prompt_on(active_prompt); | |
| for (let i = 0; i < inactive_prompt_ids.length; i++) { | |
| let inactive_prompt = document.getElementById(inactive_prompt_ids[i]); | |
| prompt_off(inactive_prompt); | |
| } | |
| let result = document.getElementById(result_id); | |
| result.src = file_paths[active_prompt_id]; | |
| } | |
| </script> | |
| <script src="https://code.jquery.com/jquery-3.2.1.slim.min.js" integrity="sha384-KJ3o2DKtIkvYIK3UENzmM7KCkRr/rE9/Qpg6aAZGJwFDMVNA/GpGFF93hXpG5KkN" crossorigin="anonymous"></script> | |
| <script src="https://cdn.jsdelivr.net/npm/popper.js@1.12.9/dist/umd/popper.min.js" integrity="sha384-ApNbgh9B+Y1QKtv3Rn7W3mgPxhU9K/ScQsAP7hUibX39j7fakFPskvXusvfa0b4Q" crossorigin="anonymous"></script> | |
| <script src="https://cdn.jsdelivr.net/npm/bootstrap@4.0.0/dist/js/bootstrap.min.js" integrity="sha384-JZR6Spejh4U02d8jOt6vLEHfe/JQGiRRSQQxSfFWpi1MquVdAyjUar5+76PVCmYl" crossorigin="anonymous"></script> | |
| </html> |