Images altered to trick machine vision can influence humans too

Re:Images altered to trick machine vision can influence huma... by pain2022 on Tuesday January 02, 2024 @05:52PM (#38848168)

> If brain activations are insensitive to subtle adversarial attacks, we would expect people to choose each picture 50% of the time on average. However, we found that the choice rate—which we refer to as the perceptual bias—was reliably above chance for a wide variety of perturbed picture pairs

Ok, but the article doesn’t say what was the actual rate?

[ Parent | on HN ]

Re:Images altered to trick machine vision can influence huma... by jaschasd on Tuesday January 02, 2024 @07:34PM (#38849173)

It does! See the effect strength plot near the bottom of the blog post, or Figures 2b and 3d in the paper.

The effect strength on humans ranges from a few percent deviation of human judgements from chance for subtle adversarial perturbations (epsilon=2), to ~15% deviations of human judgement from chance for large magnitude perturbations in the largest magnitude experimental condition.

[ Parent | on HN ]