Just Noticeable Difference for Deep Machine Vision

TL;DR

This paper investigates the existence of a Just Noticeable Difference (JND) in deep machine vision (DMV), proposing a model to identify the threshold of perceptible image distortion for DMV systems, with implications for various applications.

Contribution

It is the first to demonstrate DMV has a JND (DMV-JND) and introduces a novel JND model for image classification in deep machine vision.

Findings

DMV can tolerate images with PSNR of only 9.56dB

Proposed DMV-JND-NET effectively generates JND for DMV

Semantic-guided strategy controls JND spatial distribution

Abstract

As an important perceptual characteristic of the Human Visual System (HVS), the Just Noticeable Difference (JND) has been studied for decades with image and video processing (e.g., perceptual visual signal compression). However, there is little exploration on the existence of JND for the Deep Machine Vision (DMV), although the DMV has made great strides in many machine vision tasks. In this paper, we take an initial attempt, and demonstrate that the DMV has the JND, termed as the DMV-JND. We then propose a JND model for the image classification task in the DMV. It has been discovered that the DMV can tolerate distorted images with average PSNR of only 9.56dB (the lower the better), by generating JND via unsupervised learning with the proposed DMV-JND-NET. In particular, a semantic-guided redundancy assessment strategy is designed to restrain the magnitude and spatial distribution of the DMV-JND. Experimental results on image classification demonstrate that we successfully find the JND for deep machine vision. Our DMV-JND facilitates a possible direction for DMV-oriented image and video compression, watermarking, quality assessment, deep neural network security, and so on.