POLO Shirt Occlusion Image Dataset

The retail e-commerce industry faces significant challenges with product visibility due to various occlusions, such as clothing folds and body postures that obscure branding. Existing solutions often fail to accurately identify and classify products under these conditions, leading to reduced sales and customer dissatisfaction. This dataset aims to address the specific need for robust algorithms that can effectively handle occlusions in product images, enhancing visibility and recognition for better customer engagement. The dataset comprises images collected from diverse retail environments, utilizing high-resolution cameras in controlled lighting to ensure clarity. Quality control measures include multi-round annotations, consistency checks among annotators, and expert reviews to maintain high accuracy. The data is organized in a JPG format, with structured folders for easy access to images and associated metadata.