Describing and Localizing Multiple Changes with Transformers

TL;DR

This paper introduces MCCFormers, a transformer-based model for detecting and describing multiple changes in image pairs, supported by a new dataset and benchmark results showing significant improvements over existing methods.

Contribution

The paper presents a novel multi-change captioning transformer model, a simulation dataset, and benchmarks that demonstrate superior performance in multi-change detection and description tasks.

Findings

MCCFormers achieved the highest scores on four evaluation metrics.

The method effectively separates attention maps for each change.

Outperformed previous state-of-the-art on CLEVR-Change benchmark.

Abstract

Change captioning tasks aim to detect changes in image pairs observed before and after a scene change and generate a natural language description of the changes. Existing change captioning studies have mainly focused on a single change.However, detecting and describing multiple changed parts in image pairs is essential for enhancing adaptability to complex scenarios. We solve the above issues from three aspects: (i) We propose a simulation-based multi-change captioning dataset; (ii) We benchmark existing state-of-the-art methods of single change captioning on multi-change captioning; (iii) We further propose Multi-Change Captioning transformers (MCCFormers) that identify change regions by densely correlating different regions in image pairs and dynamically determines the related change regions with words in sentences. The proposed method obtained the highest scores on four conventional change captioning evaluation metrics for multi-change captioning. Additionally, our proposed method can separate attention maps for each change and performs well with respect to change localization. Moreover, the proposed framework outperformed the previous state-of-the-art methods on an existing change captioning benchmark, CLEVR-Change, by a large margin (+6.1 on BLEU-4 and +9.7 on CIDEr scores), indicating its general ability in change captioning tasks.