K-means Clustering Based Feature Consistency Alignment for Label-free Model Evaluation
This addresses the problem of evaluating models without ground truths for researchers in computer vision, though it is incremental as it builds on existing autoeval methods.
The paper tackled label-free model evaluation by proposing K-means Clustering Based Feature Consistency Alignment (KCFCA) to handle distribution shifts, achieving 2nd place in a CVPR 2023 challenge with an RMSE of 6.8526 and a 36% improvement over the best baseline.
The label-free model evaluation aims to predict the model performance on various test sets without relying on ground truths. The main challenge of this task is the absence of labels in the test data, unlike in classical supervised model evaluation. This paper presents our solutions for the 1st DataCV Challenge of the Visual Dataset Understanding workshop at CVPR 2023. Firstly, we propose a novel method called K-means Clustering Based Feature Consistency Alignment (KCFCA), which is tailored to handle the distribution shifts of various datasets. KCFCA utilizes the K-means algorithm to cluster labeled training sets and unlabeled test sets, and then aligns the cluster centers with feature consistency. Secondly, we develop a dynamic regression model to capture the relationship between the shifts in distribution and model accuracy. Thirdly, we design an algorithm to discover the outlier model factors, eliminate the outlier models, and combine the strengths of multiple autoeval models. On the DataCV Challenge leaderboard, our approach secured 2nd place with an RMSE of 6.8526. Our method significantly improved over the best baseline method by 36\% (6.8526 vs. 10.7378). Furthermore, our method achieves a relatively more robust and optimal single model performance on the validation dataset.