# The Elephant in the Room

Source: [The Elephant in the Room](<https://www.sota2.com/research/paper/68fa64b91e209131cea06139>)

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

Data updated: 2025-10-25T21:09:58.940Z

Authors: Amir Rosenfeld, Richard Zemel, John K. Tsotsos

Published: 2018-08-09T18:58:59.000Z

## 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.

## 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/1808.03305>)

[PDF](<https://arxiv.org/pdf/1808.03305.pdf>)
