Text-to-image synthesis method evaluation based on visual patterns
This addresses the need for better evaluation metrics in text-to-image synthesis, though it is incremental as it builds on existing methods like Inception networks and t-SNE.
The paper tackled the problem of evaluating text-to-image synthesis methods by proposing a new metric that assesses realism, variety, and semantic accuracy, showing that classification accuracy based on visual concepts effectively gauges semantic accuracy.
A commonly used evaluation metric for text-to-image synthesis is the Inception score (IS) \cite{inceptionscore}, 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 \cite{inceptionnet} 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) \cite{tsne}. 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.