HackerPapers

arXiv:1808.03305cs.CV

The Elephant in the Room

Amir Rosenfeld, Richard Zemel, John K. Tsotsos

Early discovery. Linked by one identifiable commenter. This paper is outside the main ranking; a mention is not an endorsement.

Abstract

We showcase a family of common failures of state-of-the art object detectors. These are obtained by replacing image sub-regions by another sub-image that contains a trained object. We call this "object transplanting". Modifying an image in this manner is shown to have a non-local impact on object detection. Slight changes in object position can affect its identity according to an object detector as well as that of other objects in the image. We provide some analysis and suggest possible reasons for the reported phenomena.

arXiv comments
Journal reference
DOI
Subjects
cs.CVComputer Vision and Pattern Recognition cs.LGMachine Learning

On Hacker News

Linked by 1 identifiable commenter in 3 comments across 3 threads, from Jul 2020 to Mar 2022. Found in 2 HN submissions from Aug 2018 to Sep 2018. Open submission discussions.

Comments linking this paper, per year.

Discussed in comments3 threads

  1. A visual proof that neural nets can compute any function 1 comment

    • [NOT THE OP] >> All you see is numbers and functions, and all I see is problems being solved by these AI/ML methods and often in quantifiably better ways than intelligent humans can. Maybe in playing chess and go, but in any other task that neural networks have been compared to…

      YeGoblynQueenneView on HNParent context

  2. Deep learning can’t be trusted, brain modelling pioneer says 1 comment

    • With the exception of counterfactual regret minimisation, which has nothing to do with deep learning, all those are applications of existing approaches that have become possible because of the increased expenditure of resources- processing power and data. There have been no…

      YeGoblynQueenneView on HNParent context

  3. The Bitter Lesson (2019) 1 comment

    • >> AI as a field relied mostly on 'understanding' based approaches for 50 years without much success. These approaches were too brittle and ungrounded. Why return to something that doesn't work? To begin with, because they do work and much better than the new approaches in a…

      YeGoblynQueenneView on HNParent context

Submitted to Hacker News

Connected through shared HN threads, not inferred similarity or a recommendation.