Multi-label classification of promotions in digital leaflets using textual and visual information
This work addresses the challenge of processing embedded promotion information in e-commerce for retailers and consumers, but it is incremental as it applies existing methods like region detection and text recognition to a specific domain.
The paper tackles the problem of extracting and classifying promotions from digital leaflet images in e-commerce by developing an end-to-end approach that uses visual and textual information for region detection, text recognition, and multi-label classification, achieving results that consistently outperform the baseline by a large margin.
Product descriptions in e-commerce platforms contain detailed and valuable information about retailers assortment. In particular, coding promotions within digital leaflets are of great interest in e-commerce as they capture the attention of consumers by showing regular promotions for different products. However, this information is embedded into images, making it difficult to extract and process for downstream tasks. In this paper, we present an end-to-end approach that classifies promotions within digital leaflets into their corresponding product categories using both visual and textual information. Our approach can be divided into three key components: 1) region detection, 2) text recognition and 3) text classification. In many cases, a single promotion refers to multiple product categories, so we introduce a multi-label objective in the classification head. We demonstrate the effectiveness of our approach for two separated tasks: 1) image-based detection of the descriptions for each individual promotion and 2) multi-label classification of the product categories using the text from the product descriptions. We train and evaluate our models using a private dataset composed of images from digital leaflets obtained by Nielsen. Results show that we consistently outperform the proposed baseline by a large margin in all the experiments.