# Deduplication / Image Similarity ## Overview Over time it is common to accumulate duplicate or near-duplicate images in a photo library, whether from importing the same files twice, shooting bursts, or keeping multiple edited versions of the same photo. Compendia can scan your library for visually similar images and present them in groups so you can review them and decide which to keep. ## Finding Duplicates To scan for duplicates and similar images, use **Autos > Find Similar Images**. You can choose to scan the selected files, all currently visible files, or your entire project. Compendia analyses the visual content of each image and compares it against the others to identify pairs and groups that look alike. ## Understanding Similarity Scores Compendia calculates a similarity score for each pair of images to determine how visually alike they are. Exact or near-exact duplicates receive the highest scores, while images that share strong visual similarities but are not identical, such as photos taken in quick succession or the same scene at slightly different exposures, receive lower scores. The similarity threshold is set automatically and is not user-configurable. ## Reviewing Similar Image Groups Results are presented as groups of similar images. You can view each group side by side in the review panel and compare them directly. For each group, examine the images and decide which version or versions you want to keep. Compendia can also tag each group with a Similarity Set tag if you choose. This gives each group a shared tag that you can use later with the standard tag filtering system to recall the set for further review or cleanup. This is useful if you want to work through your duplicates gradually rather than all at once. ## Removing Duplicates Compendia does not delete files directly. To remove a duplicate, right-click the file and use the menu option to open its containing folder in your file system browser (Windows Explorer on Windows, Finder on Mac). From there you can move or delete the file as you normally would. Take care when removing files, as changes made in the file system are not undone by Compendia. If you are unsure, it is safer to tag duplicates for later review rather than removing them immediately. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://compendia.gitbook.io/compendia/deduplication.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.