Positive-Augmented Contrastive Learning for Image and Video Captioning Evaluation
This work addresses the need for more accurate caption evaluation metrics in vision-and-language tasks, offering a novel approach that could improve benchmarking for researchers and developers, though it appears incremental as it builds on existing CLIP-based methods.
The paper tackles the problem of evaluating image and video captions by proposing PAC-S, a contrastive learning-based metric that unifies visual-semantic space learning with generated data augmentation, achieving the highest correlation with human judgments across multiple datasets, outperforming existing metrics like CIDEr, SPICE, and CLIP-Score.
The CLIP model has been recently proven to be very effective for a variety of cross-modal tasks, including the evaluation of captions generated from vision-and-language architectures. In this paper, we propose a new recipe for a contrastive-based evaluation metric for image captioning, namely Positive-Augmented Contrastive learning Score (PAC-S), that in a novel way unifies the learning of a contrastive visual-semantic space with the addition of generated images and text on curated data. Experiments spanning several datasets demonstrate that our new metric achieves the highest correlation with human judgments on both images and videos, outperforming existing reference-based metrics like CIDEr and SPICE and reference-free metrics like CLIP-Score. Finally, we test the system-level correlation of the proposed metric when considering popular image captioning approaches, and assess the impact of employing different cross-modal features. Our source code and trained models are publicly available at: https://github.com/aimagelab/pacscore.