From same photo: cheating on visual kinship challenges
With the propensity for deep learning models to learn unintended signals from data sets there is always the possibility that the network can `cheat' in order to solve a task. In the instance of data sets for visual kinship verification, one such unintended signal could be that the faces are cro...
| Main Authors: | , , |
|---|---|
| Format: | Internet publication |
| Language: | English |
| Published: |
2018
|
| _version_ | 1817931600443211776 |
|---|---|
| author | Dawson, M Zisserman, A Nellåker, C |
| author_facet | Dawson, M Zisserman, A Nellåker, C |
| author_sort | Dawson, M |
| collection | OXFORD |
| description | With the propensity for deep learning models to learn unintended signals from data sets there is always the possibility that the network can `cheat' in order to solve a task. In the instance of data sets for visual kinship verification, one such unintended signal could be that the faces are cropped from the same photograph, since faces from the same photograph are more likely to be from the same family. In this paper we investigate the influence of this artefactual data inference in published data sets for kinship verification.<br>
To this end, we obtain a large dataset, and train a CNN classifier to determine if two faces are from the same photograph or not. Using this classifier alone as a naive classifier of kinship, we demonstrate near state of the art results on five public benchmark data sets for kinship verification - achieving over 90% accuracy on one of them. Thus, we conclude that faces derived from the same photograph are a strong inadvertent signal in all the data sets we examined, and it is likely that the fraction of kinship explained by existing kinship models is small. |
| first_indexed | 2024-12-09T03:24:36Z |
| format | Internet publication |
| id | oxford-uuid:f14dd12b-ec04-4e44-bfeb-9be00f862db2 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-12-09T03:24:36Z |
| publishDate | 2018 |
| record_format | dspace |
| spelling | oxford-uuid:f14dd12b-ec04-4e44-bfeb-9be00f862db22024-11-26T13:53:07ZFrom same photo: cheating on visual kinship challengesInternet publicationhttp://purl.org/coar/resource_type/c_7ad9uuid:f14dd12b-ec04-4e44-bfeb-9be00f862db2EnglishSymplectic Elements2018Dawson, MZisserman, ANellåker, CWith the propensity for deep learning models to learn unintended signals from data sets there is always the possibility that the network can `cheat' in order to solve a task. In the instance of data sets for visual kinship verification, one such unintended signal could be that the faces are cropped from the same photograph, since faces from the same photograph are more likely to be from the same family. In this paper we investigate the influence of this artefactual data inference in published data sets for kinship verification.<br> To this end, we obtain a large dataset, and train a CNN classifier to determine if two faces are from the same photograph or not. Using this classifier alone as a naive classifier of kinship, we demonstrate near state of the art results on five public benchmark data sets for kinship verification - achieving over 90% accuracy on one of them. Thus, we conclude that faces derived from the same photograph are a strong inadvertent signal in all the data sets we examined, and it is likely that the fraction of kinship explained by existing kinship models is small. |
| spellingShingle | Dawson, M Zisserman, A Nellåker, C From same photo: cheating on visual kinship challenges |
| title | From same photo: cheating on visual kinship challenges |
| title_full | From same photo: cheating on visual kinship challenges |
| title_fullStr | From same photo: cheating on visual kinship challenges |
| title_full_unstemmed | From same photo: cheating on visual kinship challenges |
| title_short | From same photo: cheating on visual kinship challenges |
| title_sort | from same photo cheating on visual kinship challenges |
| work_keys_str_mv | AT dawsonm fromsamephotocheatingonvisualkinshipchallenges AT zissermana fromsamephotocheatingonvisualkinshipchallenges AT nellakerc fromsamephotocheatingonvisualkinshipchallenges |