StreamEmbed speeds up screenshot search by reusing visual data
StreamEmbed: Real-Time Indexing from Pixels for Screenshot Retrieval
Computer Vision and Pattern RecognitionInformation RetrievalMachine Learning
Summary
Screenshots change mostly by small amounts from one to the next, but most current methods treat each screenshot as completely new, making search slow or less accurate. The authors show a way to speed this up by only updating parts that change and reusing data from previous images, keeping the search quality high. Their approach balances speed and accuracy better than older methods, making it practical to quickly find screenshots based on their visual content.
What this means in practice
- •For user interface developers: Build tools that retrieve recent screen content instantly by updating visual embeddings as the screen changes.
- •For personal assistant software teams: Create faster and more memory-efficient visual memory systems to support local, privacy-preserving screen content assistants.
Authors
Dayoon Ko, Yichuan Wang, Soyeong Jeong, Joseph E. Gonzalez, Gunhee Kim, Sewon Min
Abstract
Screen history can serve as searchable visual memory for people and agents, but unlike an offline corpus, it must be encoded and indexed as each screenshot arrives. OCR-based indexing is fast but loses visual information; visual encoding takes over two seconds. Yet screens change incrementally: most pixels repeat within and across screenshots. This raises a question: do visual encoders need to recompute their representations from scratch for every new screenshot? We present StreamEmbed, a system for incremental visual embedding that exploits this redundancy. Guided solely by pixel differences, it combines spatial token pruning, which skips blocks that repeat a neighbor, with temporal KV reuse, which reuses cached keys and values for blocks unchanged across screenshots. A small learned adapter compensates for pruning and reuse to preserve fidelity to full-screenshot embeddings. On ScreenMemEval, our benchmark of 10,426 questions over 184 hours of screen histories from two humans and a computer-use agent, StreamEmbed sets a new accuracy-latency Pareto frontier on an M2 Pro GPU: it encodes up to 3.7x faster than full-screenshot encoding without sacrificing accuracy and improves accuracy over OCR-based indexing by up to 6.3 Recall@10 points at matched latency. A memory-efficient vector index stores a simulated year of screenshots in 0.13 GB of RAM. Together, these results make real-time indexing of screen history in its native visual form practical, laying a foundation for fully local visual assistants.