A Lightweight Generalizable Evaluation and Enhancement Framework for Generative Models and Generated Samples
2024 IEEE WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION WORKSHOPS, WACVW 2024(2024)
Abstract
While extensive research has been conducted on evaluating generative models, little research has been conducted on the quality assessment and enhancement of individual-generated samples. We propose a lightweight generalizable evaluation framework, designed to evaluate and enhance the generative models and generated samples. Our framework trains a classifier-based dataset-specific model, enabling its application to unseen generative models and extending its compatibility with both deep learning and efficient machine learning-based methods. We propose three novel evaluation metrics aiming at capturing distribution correlation, quality, and diversity of generated samples. These metrics collectively offer a more thorough performance evaluation of generative models compared to the Frechet Inception Distance (FID). Our approach assigns individual quality scores to each generated sample for sample-level evaluation. This enables better sample mining and thereby improves the performance of generative models by filtering out lower-quality generations. Extensive experiments across various datasets and generative models demonstrate the effectiveness and efficiency of the proposed method.
MoreTranslated text
Key words
Evaluation Framework,Machine Learning,Model Performance,Deep Learning,Quality Assessment,Model Evaluation,Machine Learning-based Methods,Inception Distance,Fréchet Inception Distance,Training Set,Support Vector Machine,Deep Neural Network,Feature Space,Sample Quality,Feature Learning,Model Size,Real Samples,Score Calculation,Discriminative Features,Deep Features,Soft Labels,Efficient Machine Learning,Spectral Representation,Tail Of Distribution,Image Inpainting,Deep Feature Extraction,Real Class,Object Dataset,Dataset Construction
AI Read Science
Must-Reading Tree
Example
![](https://originalfileserver.aminer.cn/sys/aminer/pubs/mrt_preview.jpeg)
Generate MRT to find the research sequence of this paper
Chat Paper
Summary is being generated by the instructions you defined