Technical Report of NICE Challenge at CVPR 2024: Caption Re-ranking Evaluation Using Ensembled CLIP and Consensus Scores

TL;DR

This paper introduces the ECO framework, combining ensembled CLIP and consensus scores, to effectively evaluate and rank image captions, achieving top results in the CVPR 2024 NICE challenge.

Contribution

The paper presents a novel ECO pipeline that integrates ensembled CLIP and consensus scores for improved caption ranking in image captioning evaluation.

Findings

Secured top positions in multiple metrics at CVPR 2024 NICE challenge.

Demonstrated the effectiveness of combining semantic alignment and caption essentialness.

Achieved state-of-the-art results in caption re-ranking evaluation.

Abstract

This report presents the ECO (Ensembled Clip score and cOnsensus score) pipeline from team DSBA LAB, which is a new framework used to evaluate and rank captions for a given image. ECO selects the most accurate caption describing image. It is made possible by combining an Ensembled CLIP score, which considers the semantic alignment between the image and captions, with a Consensus score that accounts for the essentialness of the captions. Using this framework, we achieved notable success in the CVPR 2024 Workshop Challenge on Caption Re-ranking Evaluation at the New Frontiers for Zero-Shot Image Captioning Evaluation (NICE). Specifically, we secured third place based on the CIDEr metric, second in both the SPICE and METEOR metrics, and first in the ROUGE-L and all BLEU Score metrics. The code and configuration for the ECO framework are available at https://github.com/DSBA-Lab/ECO .