2022/11/14 by Junyang Wang, Yi Zhang, Wang, Junyang +7 · 1 citation
Computer Science · #Advanced Image and Video Retrieval Techniques #Computer Vision and Pattern Recognition (cs.CV) #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences #Multimodal Machine Learning Applications
paper · pdf · doi:10.48550/arxiv.2211.07275
openalex publication_date 2022/11/14 · openalex created_date 2022/11/23 · openalex updated_date 2026/07/28
CLIP (Contrastive Language-Image Pre-Training) has shown remarkable zero-shot transfer capabilities in cross-modal correlation tasks such as visual classification and image retrieval. However, its performance in cross-modal generation tasks like zero-shot image captioning remains unsatisfied. In this work, we discuss that directly employing CLIP for zero-shot image captioning relies more on the textual modality in context and largely ignores the visual information, which we call contextual language prior. To address this, we propose Cross-modal Language Models (CLMs) to facilitate unsupervised cross-modal learning. We further propose Anchor Augment to guide the generative model's attention to the fine-grained information in the representation of CLIP. Experiments on MS COCO and Flickr 30K validate the promising performance of proposed approach in both captioning quality and computational efficiency.