Similar image search: Find visually similar images using a reference file
Users want to find images that look similar to a selected reference file so they can discover relevant assets based on visual content, not just keywords or metadata. Problem / Pain: Finding visually similar images currently relies entirely on keywords and metadata. Users who are looking for assets that share a visual style, subject, or composition, for example, during campaign planning, duplicate checking, or content expansion have no way to search by visual similarity. This means relevant assets can easily be missed if they are not tagged in a way that matches the user's search terms. Impact / Benefit: Visual similarity search would allow users to find relevant images based on how they look, not just how they are described. This makes asset discovery faster and more intuitive, especially in cases where metadata is incomplete, tagging conventions are unfamiliar, or the user simply cannot describe what they are looking for in words. Example use case: A marketing user has found one strong campaign image and wants to find others with a similar visual style for the same campaign. Instead of guessing which keywords might match, they select the image as a reference and immediately see a list of visually similar assets sorted by relevance and ready to review. Idea source: Customer feedback Voting note: Vote for this idea if this need is relevant to you as well.
marked this as Planned· 4mo
marked this as In development· 2mo
marked this as Released· 3w
More in Search
- AI-powered search available in Public PortalsVotes 1
- Faceted search in Public Portals for faster, more targeted content discoveryVotes 1
- Include metadata fields in AI-powered search results across all modulesVotes 1
- Add in-document text search with highlighting to the document viewerVotes 1
- AI-powered search: Support for videos and GIFsVotes 1
