# Personal AI Agent for Camera Roll VQA

Source: [Personal AI Agent for Camera Roll VQA](<https://www.sota2.com/research/paper/6a22584f86173ad9c1468888>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-06-08T06:00:00.979Z

Authors: Thao Nguyen, Krishna Kumar Singh, Donghyun Kim, Yong Jae Lee, Yuheng Li

Published: 2026-06-03T17:59:30.000Z

## Abstract

We study the personal camera roll visual question answering setting. In this setting, a conversational AI assistant can access a user's personal camera roll and retrieve relevant photos to answer queries, ranging from simple factual questions (e.g., \`\`Name of the food I tried yesterday?'') to more open-ended ones (e.g., \`\`Recommend some dishes I have never eaten before''). Given the vast nature of the personal camera roll (i.e., multiple years, hundreds to thousands of photos), a successful AI assistant needs to understand a long-horizon, highly personalized visual content stream in order to navigate and locate the correct and/or relevant information. To support this, we collect and manually annotate questions that mimic real-world usage. The final dataset, camroll, contains 50 users, 31,476 images, and 2,500 QA pairs. We further design camroll-agent, a conversational AI agent equipped with hierarchical memory and a minimal set of tools for efficient navigation over large, personalized visual memory. Experimental results show that camroll-agent outperforms numerous baselines and methods for long-context understanding AI agents system. Together, the camroll dataset and camroll-agent highlight the gap in AI agents' long-context reasoning: personalized visual memory requires different approaches from standard long-context textual memory, especially when consistency, visual details, and user-specific context are present.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Free-form Visual Question Answering on camroll](<https://www.sota2.com/research/sota/free-form-visual-question-answering-on-camroll.md>): Free-form Visual Question Answering; camroll; Judge Score: 4.11.

- [Multi-choice Visual Question Answering on camroll](<https://www.sota2.com/research/sota/multi-choice-visual-question-answering-on-camroll.md>): Multi-choice Visual Question Answering; camroll; Accuracy: 70.5.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2606.05275>)

[PDF](<https://arxiv.org/pdf/2606.05275.pdf>)

[Implementation](<https://github.com/thaoshibe/camroll>)