Context-Aware Group Captioning via Self-Attention and Contrastive Features
This work addresses a novel task in image captioning for AI researchers, focusing on multi-image contexts, but it is incremental as it builds on existing captioning techniques.
The paper tackles the problem of generating captions for groups of target images by contrasting them with reference image groups, introducing a new task called context-aware group captioning. It proposes a method using self-attention and contrastive features, achieving effectiveness on datasets built from Conceptual Captions and a new Stock Captions dataset.
While image captioning has progressed rapidly, existing works focus mainly on describing single images. In this paper, we introduce a new task, context-aware group captioning, which aims to describe a group of target images in the context of another group of related reference images. Context-aware group captioning requires not only summarizing information from both the target and reference image group but also contrasting between them. To solve this problem, we propose a framework combining self-attention mechanism with contrastive feature construction to effectively summarize common information from each image group while capturing discriminative information between them. To build the dataset for this task, we propose to group the images and generate the group captions based on single image captions using scene graphs matching. Our datasets are constructed on top of the public Conceptual Captions dataset and our new Stock Captions dataset. Experiments on the two datasets show the effectiveness of our method on this new task. Related Datasets and code are released at https://lizw14.github.io/project/groupcap .