Video Method for Estimating the Apparel Worn Size

Online commerce has experienced a great opportunity in recent years due to the pandemic. The forecasts show that trends for online purchases are to be maintained, so the competitive challenge consists further in simplifying the interaction with the customer and reducing the time of selecting products. In fashion segment, one of the main factors influencing purchase decision is the certainty of the product fitting on the customer’s size. As solution to this issue, paper presents a video method for determining the size worn in clothes, based on the use of an anatomical reference, the interpupillary distance. The average interpupillary distance in female is 61.63mm and has a standard deviation of 2.66mm and the average interpupillary distance in male is 65.32mm and has a standard deviation of 1.50mm. The method of video estimation of the worn apparel size includes six phases: the identification of the pupils, the determination of the centers of pupils, the calculation of the interpupillary distance in pixels, the video identification of shoulders, the determination of the width of the shoulders in pixels, the conversion of the width of the shoulders in centimeters and determining the size worn.