Text-To-Image Synthesis Method Evaluation Based on Visual Patterns

Publikation: Bidrag til bog/antologi/rapport/proceedingKonferencebidrag i proceedingsForskningpeer review

A commonly used evaluation metric for text-to-image synthesis is the Inception score (IS) [1], which has been shown to be a quality metric that correlates well with human judgment. However, IS does not reveal properties of the generated images indicating the ability of a text-to-image synthesis method to correctly convey semantics of the input text descriptions. In this paper, we introduce an evaluation metric and a visual evaluation method allowing for the simultaneous estimation of the realism, variety and semantic accuracy of generated images. The proposed method uses a pre-trained Inception network [2] to produce high dimensional representations for both real and generated images. These image representations are then visualized in a 2-dimensional feature space defined by the t-distributed Stochastic Neighbor Embedding (t-SNE) [3]. Visual concepts are determined by clustering the real image representations, and are subsequently used to evaluate the similarity of the generated images to the real ones by classifying them to the closest visual concept. The resulting classification accuracy is shown to be a effective gauge for the semantic accuracy of text-to-image synthesis methods.

OriginalsprogEngelsk
Titel2020 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2020 - Proceedings
Antal sider5
ForlagIEEE
Udgivelsesår2020
Sider4097-4101
Artikelnummer9053034
ISBN (Elektronisk)9781509066315
DOI
StatusUdgivet - 2020
Begivenhed2020 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2020 - Barcelona, Spanien
Varighed: 4 maj 20208 maj 2020

Konference

Konference2020 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2020
LandSpanien
ByBarcelona
Periode04/05/202008/05/2020
SponsorThe Institute of Electrical and Electronics Engineers, Signal Processing Society

Se relationer på Aarhus Universitet Citationsformater

ID: 194897708