Countering Inconsistent Labelling by Google's Vision API for Rotated Images

by   Aman Apte, et al.

Google's Vision API analyses images and provides a variety of output predictions, one such type is context-based labelling. In this paper, it is shown that adversarial examples that cause incorrect label prediction and spoofing can be generated by rotating the images. Due to the black-boxed nature of the API, a modular context-based pre-processing pipeline is proposed consisting of a Res-Net50 model, that predicts the angle by which the image must be rotated to correct its orientation. The pipeline successfully performs the correction whilst maintaining the image's resolution and feeds it to the API which generates labels similar to the original correctly oriented image and using a Percentage Error metric, the performance of the corrected images as compared to its rotated counter-parts is found to be significantly higher. These observations imply that the API can benefit from such a pre-processing pipeline to increase robustness to rotational perturbances.



There are no comments yet.


page 2

page 4

page 7

page 8


Simple Transparent Adversarial Examples

There has been a rise in the use of Machine Learning as a Service (MLaaS...

Image Pre-processing Using OpenCV Library on MORPH-II Face Database

This paper outlines the steps taken toward pre-processing the 55,134 ima...

Deceiving Google's Cloud Video Intelligence API Built for Summarizing Videos

Despite the rapid progress of the techniques for image classification, v...

Attacking Automatic Video Analysis Algorithms: A Case Study of Google Cloud Video Intelligence API

Due to the growth of video data on Internet, automatic video analysis ha...

A Comprehensive Pipeline for Hotel Recommendation System

This paper addresses a comprehensive pipeline to build a hotel recommend...

FrugalMCT: Efficient Online ML API Selection for Multi-Label Classification Tasks

Multi-label classification tasks such as OCR and multi-object recognitio...

Text Mining to Identify and Extract Novel Disease Treatments From Unstructured Datasets

Objective: We aim to learn potential novel cures for diseases from unstr...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.