arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2410.02004cs.CVcs.AIcs.LG
收藏

基于归一化流的图像生成评价指标

Normalizing Flow-Based Metric for Image Generation

Indian Institute of Technology Bombay(印度理工学院孟买分校)

Pranav Jeevan, Neeraj Nixon, Amit Sethi

AI总结:

本研究提出两种基于归一化流的生成图像真实感评价指标FLD和D-FLD,相比FID等指标需样本量更少、计算效率更高,在新域场景优势显著,且实验验证其与图像退化程度呈预期单调关系。

摘要:

我们提出了两种基于归一化流的新型评价指标,用于评估生成图像的真实感:一种更简洁高效的基于流的似然距离(flow-based likelihood distance,FLD),以及一种更精确的基于双流的似然距离(dual-flow based likelihood distance,D-FLD)。由于归一化流可用于计算精确似然,所提指标能够衡量生成图像与给定域真实图像分布的契合程度。这一特性使所提指标相较于广泛使用的Fréchet inception distance(FID)及其他近期指标具备若干优势。首先,所提指标仅需数百张图像即可稳定(均值收敛),而FID需要数万张,其他指标至少需要数千张。这使得即便针对小规模生成图像集(如训练循环内的验证批次)也能开展可靠评估。其次,用于计算所提指标的网络参数量比计算FID所用的Inception-V3少一个数量级以上,计算效率更高。在评估新域(例如X光图像)生成图像的真实感时,理想情况下需在真实图像上重新训练这些网络以建模其独特分布,因此我们的小型网络在新域场景下优势更显著。大量实验表明,所提指标与各类图像退化程度之间呈现预期的单调关系。

英文摘要:

We propose two new evaluation metrics to assess realness of generated images based on normalizing flows: a simpler and efficient flow-based likelihood distance (FLD) and a more exact dual-flow based likelihood distance (D-FLD). Because normalizing flows can be used to compute the exact likelihood, the proposed metrics assess how closely generated images align with the distribution of real images from a given domain. This property gives the proposed metrics a few advantages over the widely used Fréchet inception distance (FID) and other recent metrics. Firstly, the proposed metrics need only a few hundred images to stabilize (converge in mean), as opposed to tens of thousands needed for FID, and at least a few thousand for the other metrics. This allows confident evaluation of even small sets of generated images, such as validation batches inside training loops. Secondly, the network used to compute the proposed metric has over an order of magnitude fewer parameters compared to Inception-V3 used to compute FID, making it computationally more efficient. For assessing the realness of generated images in new domains (e.g., x-ray images), ideally these networks should be retrained on real images to model their distinct distributions. Thus, our smaller network will be even more advantageous for new domains. Extensive experiments show that the proposed metrics have the desired monotonic relationships with the extent of image degradation of various kinds.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑