CVDec 8, 2021

Assessing a Single Image in Reference-Guided Image Synthesis

arXiv:2112.04163v118 citations
Originality Incremental advance
AI Analysis

This addresses a specific bottleneck in generative model evaluation for tasks like style transfer, but it is incremental as it builds on existing evaluation methods.

The paper tackles the problem of evaluating single generated images in reference-guided image synthesis, where existing metrics are inadequate, by proposing a learning-based framework called RISA that achieves high consistency with human preference without requiring human annotations.

Assessing the performance of Generative Adversarial Networks (GANs) has been an important topic due to its practical significance. Although several evaluation metrics have been proposed, they generally assess the quality of the whole generated image distribution. For Reference-guided Image Synthesis (RIS) tasks, i.e., rendering a source image in the style of another reference image, where assessing the quality of a single generated image is crucial, these metrics are not applicable. In this paper, we propose a general learning-based framework, Reference-guided Image Synthesis Assessment (RISA) to quantitatively evaluate the quality of a single generated image. Notably, the training of RISA does not require human annotations. In specific, the training data for RISA are acquired by the intermediate models from the training procedure in RIS, and weakly annotated by the number of models' iterations, based on the positive correlation between image quality and iterations. As this annotation is too coarse as a supervision signal, we introduce two techniques: 1) a pixel-wise interpolation scheme to refine the coarse labels, and 2) multiple binary classifiers to replace a naïve regressor. In addition, an unsupervised contrastive loss is introduced to effectively capture the style similarity between a generated image and its reference image. Empirical results on various datasets demonstrate that RISA is highly consistent with human preference and transfers well across models.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes