FaD-VLP: Fashion Vision-and-Language Pre-training towards Unified Retrieval and Captioning
This work addresses multimodal learning problems in the fashion domain for e-commerce applications, representing an incremental improvement by adapting existing methods to fashion-specific data.
The paper tackles the challenge of multimodal tasks in fashion, such as retrieving items based on image and text feedback, by proposing a fashion-specific pre-training framework and a flexible decoder-based model. The result is a competitive performance on diverse fashion tasks including cross-modal retrieval and captioning.
Multimodal tasks in the fashion domain have significant potential for e-commerce, but involve challenging vision-and-language learning problems - e.g., retrieving a fashion item given a reference image plus text feedback from a user. Prior works on multimodal fashion tasks have either been limited by the data in individual benchmarks, or have leveraged generic vision-and-language pre-training but have not taken advantage of the characteristics of fashion data. Additionally, these works have mainly been restricted to multimodal understanding tasks. To address these gaps, we make two key contributions. First, we propose a novel fashion-specific pre-training framework based on weakly-supervised triplets constructed from fashion image-text pairs. We show the triplet-based tasks are an effective addition to standard multimodal pre-training tasks. Second, we propose a flexible decoder-based model architecture capable of both fashion retrieval and captioning tasks. Together, our model design and pre-training approach are competitive on a diverse set of fashion tasks, including cross-modal retrieval, image retrieval with text feedback, image captioning, relative image captioning, and multimodal categorization.