# Spatial Memory for AR Glasses: Hierarchical Visual Acquisition and Source-Linked Recall

DevFeed: [Spatial Memory for AR Glasses: Hierarchical Visual Acquisition and Source-Linked Recall](<https://devfeed.tech/articles/spatial-memory-for-ar-glasses-hierarchical-visual-acquisition-and-source-linked-recall-117496.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.04494>)

Author: Tianxing Zhu, Dan Huang

Published: 2026-11-04T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Glasses & AR Tech](<https://devfeed.tech/topics/glasses-ar-tech.md>), [recording](<https://devfeed.tech/topics/recording.md>)

Tags: [abstract](<https://devfeed.tech/tags/abstract.md>), [acceptance](<https://devfeed.tech/tags/acceptance.md>), [acquisition](<https://devfeed.tech/tags/acquisition.md>), [appearance](<https://devfeed.tech/tags/appearance.md>), [ar](<https://devfeed.tech/tags/ar.md>), [ar-glasses](<https://devfeed.tech/tags/ar-glasses.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [bounded](<https://devfeed.tech/tags/bounded.md>), [camera](<https://devfeed.tech/tags/camera.md>), [capture](<https://devfeed.tech/tags/capture.md>), [combines](<https://devfeed.tech/tags/combines.md>), [control](<https://devfeed.tech/tags/control.md>), [core](<https://devfeed.tech/tags/core.md>), [cost](<https://devfeed.tech/tags/cost.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [glasses](<https://devfeed.tech/tags/glasses.md>), [memory](<https://devfeed.tech/tags/memory.md>), [recall](<https://devfeed.tech/tags/recall.md>), [source](<https://devfeed.tech/tags/source.md>), [spatial](<https://devfeed.tech/tags/spatial.md>)

## AI overview

The prototype links past encounters captured by AR glasses to later recall while distinguishing recorded images, semantic interpretations, and evidence for a specific answer. It screens monochrome observations for quality, visual novelty, and temporal support before selecting detailed RGB captures; semantic records remain linked to their source images so retrieval can answer indexed questions or inspect visual details missing from descriptions. It also supports bounded reassessment of familiar views when evidence is lacking and uses source dependencies to guide retention. In a 131.35-second replay, it selected six RGB views from 648 monitoring observations, while feature extraction ran on 97.7% of observations. The authors identify frequent local inference as a remaining cost, and note that exposure duration and timing across views constrain what evidence is preserved.

## Source excerpt

arXiv:2611.04494v1 Announce Type: new Abstract: We present a spatial-memory system that enables optical see-through AR glasses to recall earlier physical encounters. The dual-camera prototype combines hierarchical visual acquisition with source-linked storage and staged retrieval. Quality screen-ing, global and regional appearance novelty, and temporal support select monochrome observations for detailed RGB cap-ture. Semantic observations retain links to their source images, allowing retrieval to answer indexed historical questions direct-ly or inspect visual details omitted from scene descriptions. The architecture separates capture acceptance, interpretation out-come, and support for a particular question. This distinction motivates two core control mechanisms: bounded reassessment of familiar views when earlier processing leaves an evidence gap, and retention guided by source dependencies. In a 131.35 s egocentric replay, the implemented acquisition policy selected six RGB views from 648 monitoring observations, while feature extraction ran on 97.7% of observations. These results demon-strate selective RGB acquisition and identify frequent local inference as a remaining cost. Exposure duration and cross-view timing further constrain the evidence preserved from se-lected encounters. The system provides a practical acquisition-to-recall architecture for persistent AR assistance, with explicit boundaries between captured evidence, indexed interpretations, and supported historical claims.