# Privacy Preserving Visual Question Answering

Source: [Privacy Preserving Visual Question Answering](<https://www.sota2.com/research/paper/68fa6bda1e209131cea30992>)

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

Data updated: 2025-10-25T23:58:04.948Z

Authors: Cristian-Paul Bara, Qing Ping, Abhinav Mathur, Govind Thattai, Rohith MV, Gaurav S. Sukhatme

Published: 2022-02-15T20:22:50.000Z

## Abstract

We introduce a novel privacy-preserving methodology for performing Visual Question Answering on the edge. Our method constructs a symbolic representation of the visual scene, using a low-complexity computer vision model that jointly predicts classes, attributes and predicates. This symbolic representation is non-differentiable, which means it cannot be used to recover the original image, thereby keeping the original image private. Our proposed hybrid solution uses a vision model which is more than 25 times smaller than the current state-of-the-art (SOTA) vision models, and 100 times smaller than end-to-end SOTA VQA models. We report detailed error analysis and discuss the trade-offs of using a distilled vision model and a symbolic representation of the visual scene.

## 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.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2202.07712>)

[PDF](<https://arxiv.org/pdf/2202.07712.pdf>)