Saved in:
Bibliographic Details
Main Authors: Cetin, Doruk, Schesch, Benedikt, Stamenkovic, Petar, Huber, Niko Benjamin, Zünd, Fabio, Helou, Majed El
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2406.18430
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917706022780928
author Cetin, Doruk
Schesch, Benedikt
Stamenkovic, Petar
Huber, Niko Benjamin
Zünd, Fabio
Helou, Majed El
author_facet Cetin, Doruk
Schesch, Benedikt
Stamenkovic, Petar
Huber, Niko Benjamin
Zünd, Fabio
Helou, Majed El
contents Assessing distances between images and image datasets is a fundamental task in vision-based research. It is a challenging open problem in the literature and despite the criticism it receives, the most ubiquitous method remains the Fréchet Inception Distance. The Inception network is trained on a specific labeled dataset, ImageNet, which has caused the core of its criticism in the most recent research. Improvements were shown by moving to self-supervision learning over ImageNet, leaving the training data domain as an open question. We make that last leap and provide the first analysis on domain-specific feature training and its effects on feature distance, on the widely-researched facial image domain. We provide our findings and insights on this domain specialization for Fréchet distance and image neighborhoods, supported by extensive experiments and in-depth user studies.
format Preprint
id arxiv_https___arxiv_org_abs_2406_18430
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Facial Image Feature Analysis and its Specialization for Fréchet Distance and Neighborhoods
Cetin, Doruk
Schesch, Benedikt
Stamenkovic, Petar
Huber, Niko Benjamin
Zünd, Fabio
Helou, Majed El
Computer Vision and Pattern Recognition
Assessing distances between images and image datasets is a fundamental task in vision-based research. It is a challenging open problem in the literature and despite the criticism it receives, the most ubiquitous method remains the Fréchet Inception Distance. The Inception network is trained on a specific labeled dataset, ImageNet, which has caused the core of its criticism in the most recent research. Improvements were shown by moving to self-supervision learning over ImageNet, leaving the training data domain as an open question. We make that last leap and provide the first analysis on domain-specific feature training and its effects on feature distance, on the widely-researched facial image domain. We provide our findings and insights on this domain specialization for Fréchet distance and image neighborhoods, supported by extensive experiments and in-depth user studies.
title Facial Image Feature Analysis and its Specialization for Fréchet Distance and Neighborhoods
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2406.18430