[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-121080-en":3,"doc-seo-121080-105":30,"detail-sidebar-cat-0-en-105":90},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},121080,8796095462418,"Noah","https://ap-avatar.wpscdn.com/avatar/80000253c1241d02b47?x-image-process=image/resize,m_fixed,w_180,h_180&k=1778826106357471780",6,"Technology","Demonstration of MaskSearch - Efficiently Querying Image Masks for Machine Learning Workflows","MaskSearch is a system that accelerates queries over databases of image masks produced by machine learning models. It defines and speeds up a mask-property-based query category that retrieves both images and their corresponding masks, enabling tasks such as uncovering spurious correlations learned by models and comparing model saliency with human attention. The demonstration introduces a GUI for interactive mask-property exploration, provides hands-on access to capabilities and limitations within ML workflows, and explains how MaskSearch improves query performance for image masks.","Demonstration of MaskSearch: Efficiently Querying Image Masks for Machine Learning Workflows  \nLindsey Linxi Wei*  \nUniversity of Washington [linxiwei@cs](linxiwei@cs.washington.edu)[.](linxiwei@cs.washington.edu)[washington](linxiwei@cs.washington.edu)[.](linxiwei@cs.washington.edu)[edu](linxiwei@cs.washington.edu)  \nJingchuan Zhou*  \nUniversity of Washington [jzhou27@cs](jzhou27@cs.washington.edu)[.](jzhou27@cs.washington.edu)[washington](jzhou27@cs.washington.edu)[.](jzhou27@cs.washington.edu)[edu](jzhou27@cs.washington.edu)  \nChung Yik Edward Yeung*  \nUniversity of Washington [chungy04@cs](chungy04@cs.washington.edu)[.](chungy04@cs.washington.edu)[washington](chungy04@cs.washington.edu)[.](chungy04@cs.washington.edu)[edu](chungy04@cs.washington.edu)  \nDong He University of Washington [donghe@cs](donghe@cs.washington.edu)[.](donghe@cs.washington.edu)[washington](donghe@cs.washington.edu)[.](donghe@cs.washington.edu)[edu](donghe@cs.washington.edu)  \nHongjian Yu* University of Washington [hjyu@cs](hjyu@cs.washington.edu)[.](hjyu@cs.washington.edu)[washington](hjyu@cs.washington.edu)[.](hjyu@cs.washington.edu)[edu](hjyu@cs.washington.edu)  \nMagdalena Balazinska University of Washington [magda@cs](magda@cs.washington.edu)[.](magda@cs.washington.edu)[washington](magda@cs.washington.edu)[.](magda@cs.washington.edu)[edu](magda@cs.washington.edu)  \narXiv :2404 .06563v 1 [ cs .DB] 9 Apr 2024  \nABSTRACT  \nWe demonstrate MaskSearch, a system designed to accelerate queries over databases of image masks generated by machine learning models. MaskSearch formalizes and accelerates a new category of queries for retrieving images and their corresponding masks based on mask properties, which support various applications, from identifying spurious correlations learned by models to exploring discrepancies between model saliency and human attention. This demonstration makes the following contributions: (1) the introduction of MaskSearch’s graphical user interface (GUI), which enables interactive exploration of image databases through mask properties,(2) hands-on opportunities for users to explore MaskSearch’s capabilities and constraints within machine learning workflows, and (3) an opportunity for conference attendees to understand how MaskSearch accelerates queries over image masks.  \n1 INTRODUCTION  \nMasking is a way to highlight or isolate certain parts of an image based on desired properties for further processing or analysis. Machine learning tasks over image databases often involve generating and using masks, such as image segmentation masks [13] and modelsaliency maps [16] . These masks are crucial for a wide range of applications, from model explanation [6, 16] to real-world analysis [1] . For example, practitioners developing image classification models can generate model saliency maps to understand which pixels contribute the most to the model’s predictions.  \nConsider a scenario further discussed in §4, Alice, a data engineer, uses the iWildCam dataset [4] for developing a wild animal image classification model. Facing validation accuracy issues, she computessaliency maps [16] and YOLO-generated bounding boxes [13] for themisclassified images, an example of which is shown in Figure 1 . In thesaliency map, the red pixels indicate higher importance for the model’s prediction, and the blue pixels indicate lower importance. She finds that the model focuses on the background pixels, notably outside the ground-truth object bounding boxes, rather than the animals, leading  \n*Equal contribution.  \nThis work is licensed under the Creative Commons BY-NC-ND 4.0 International License. Visit [https://creativecommons](https://creativecommons.org/licenses/by-nc-nd/4.0/ to view a copy of)[.](https://creativecommons.org/licenses/by-nc-nd/4.0/ to view a copy of)[org/licenses/by-nc-nd/4](https://creativecommons.org/licenses/by-nc-nd/4.0/ to view a copy of)[.](https://creativecommons.org/licenses/by-nc-nd/4.0/ to view a copy of)[0/ to view a copy of](ht","cbCainBBQKwNgQ99","https://ap.wps.com/l/cbCainBBQKwNgQ99","pdf",4689720,1,4,"English","en",105,"# Abstract\n# Introduction\n## Image masking and its role in ML\n## Example workflow: correcting spurious background reliance\n## Need for efficient systems support\n# MaskSearch overview","[{\"question\":\"What is MaskSearch designed to do?\",\"answer\":\"MaskSearch accelerates queries over databases of image masks generated by machine learning models. It retrieves images together with their masks using properties of the masks.\"},{\"question\":\"Why are mask-property queries useful in machine learning workflows?\",\"answer\":\"They help practitioners identify spurious correlations, and they support analysis by comparing model saliency with human attention. This is valuable for both debugging models and improving reliability.\"},{\"question\":\"What does the MaskSearch demonstration include?\",\"answer\":\"The demonstration provides a GUI for interactive exploration of image databases through mask properties. It also offers hands-on experience with MaskSearch’s capabilities and constraints within ML workflows.\"}]","Demonstration of MaskSearch - Efficiently Querying Image Masks for Machine Learning Workflows | PDF",1785733619,10,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":85,"head_meta":87,"extra_data":89,"updated_unix":28},"demonstration-of-masksearch-efficiently-querying-image-masks-for-machine-learning-workflows","",{"@graph":36,"@context":84},[37,53,67],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/technology/",3,{"item":52,"name":13,"@type":43,"position":21},"https://docshare.wps.com/document/demonstration-of-masksearch-efficiently-querying-image-masks-for-machine-learning-workflows/121080/",{"url":52,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":61,"encodingFormat":60,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-08-03",true,{"@type":64,"interactionType":65,"userInteractionCount":4},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80],{"name":71,"@type":72,"acceptedAnswer":73},"What is MaskSearch designed to do?","Question",{"text":74,"@type":75},"MaskSearch accelerates queries over databases of image masks generated by machine learning models. It retrieves images together with their masks using properties of the masks.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"Why are mask-property queries useful in machine learning workflows?",{"text":79,"@type":75},"They help practitioners identify spurious correlations, and they support analysis by comparing model saliency with human attention. This is valuable for both debugging models and improving reliability.",{"name":81,"@type":72,"acceptedAnswer":82},"What does the MaskSearch demonstration include?",{"text":83,"@type":75},"The demonstration provides a GUI for interactive exploration of image databases through mask properties. It also offers hands-on experience with MaskSearch’s capabilities and constraints within ML workflows.","https://schema.org",{"og:url":52,"og:type":86,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":88,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":91},[92,96,100,104,109,112,117,122,127,130,133],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":93,"show_sort_weight":94,"slug":95},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":97,"show_sort_weight":98,"slug":99},"Literature",80,"literature",{"id":21,"doc_module":4,"doc_module_name":46,"category_name":101,"show_sort_weight":102,"slug":103},"Exam",70,"exam",{"id":105,"doc_module":4,"doc_module_name":46,"category_name":106,"show_sort_weight":107,"slug":108},5,"Comic",60,"comic",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":110,"slug":111},50,"technology",{"id":113,"doc_module":4,"doc_module_name":46,"category_name":114,"show_sort_weight":115,"slug":116},7,"Healthcare",40,"healthcare",{"id":118,"doc_module":4,"doc_module_name":46,"category_name":119,"show_sort_weight":120,"slug":121},8,"Research & Report",30,"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":29,"doc_module":4,"doc_module_name":46,"category_name":131,"show_sort_weight":29,"slug":132},"Lifestyle","lifestyle",{"id":134,"doc_module":4,"doc_module_name":46,"category_name":135,"show_sort_weight":105,"slug":136},19,"General","general"]