Multi-method Integration with Confidence-based Weighting for Zero-shot Image Classification

TL;DR

This paper presents a multi-method zero-shot image classification framework that leverages ChatGPT, DALL-E, CLIP, and DINO with confidence-based weighting to improve accuracy on standard datasets.

Contribution

It introduces a novel integration framework combining multiple models and alignment strategies with adaptive confidence weighting for enhanced zero-shot learning performance.

Findings

Achieves over 96% AUROC on CIFAR-10, CIFAR-100, and TinyImageNet.

Surpasses 99% AUROC on CIFAR-10.

Significantly outperforms single-model approaches.

Abstract

This paper introduces a novel framework for zero-shot learning (ZSL), i.e., to recognize new categories that are unseen during training, by using a multi-model and multi-alignment integration method. Specifically, we propose three strategies to enhance the model's performance to handle ZSL: 1) Utilizing the extensive knowledge of ChatGPT and the powerful image generation capabilities of DALL-E to create reference images that can precisely describe unseen categories and classification boundaries, thereby alleviating the information bottleneck issue; 2) Integrating the results of text-image alignment and image-image alignment from CLIP, along with the image-image alignment results from DINO, to achieve more accurate predictions; 3) Introducing an adaptive weighting mechanism based on confidence levels to aggregate the outcomes from different prediction methods. Experimental results on multiple datasets, including CIFAR-10, CIFAR-100, and TinyImageNet, demonstrate that our model can significantly improve classification accuracy compared to single-model approaches, achieving AUROC scores above 96% across all test datasets, and notably surpassing 99% on the CIFAR-10 dataset.