Partners  |  About Us  |  Blog  |  Resources  |  Training  |  Customer Reviews  |  Case Studies  |  Contact Us

Practical Algorithms for Product Photography Background Removal

Practical Algorithms for Product Photography Background Removal

by Darian Muresan, Ph.D.

At Iconasys  we develop software and hardware solutions for increasing product photography efficiencies. With the advent of eCommerce taking over virtually all aspects of our purchasing experiences, good and efficient product photography solutions enable everyone, from a moms and pops store on eBay to very large stores, to create a professional presence on the internet.  A particularly important feature of product photography is consistent, pure white backgrounds. This article presents a bird’s eye overview of the background removal algorithms integrated in Iconasys’ Shutter Stream Product Photography Software.

First, Shutter Stream is a product photography software suite that runs on Mac and Windows and allows users to connect directly to selected cameras: Canon, Nikon, Sony and mobile phones (iOS and Android). Once the camera connects to Shutter Stream, users see a live view of what the camera sees. Camera’s settings (shutter speed, aperture, exposure compensation, ISO, white balance and focus) can be controlled from Shutter Stream and with a click of a button a picture is taken and automatically transferred to the computer for further processing.

Second, the product photography background removal tools are part of Shutter Stream’s editing tools. The background removal tools are separated into several categories:

  1. The first background removal algorithm is based on using a foreground and a background image. The algorithm requires two images. The first image is the image with the object of interest in the image and the second image is with the object removed. The algorithm uses an intelligent background subtraction method and region growing technique, as described in the figure below.
Product Photography: Background Removal 01
(click to view larger image)
  1. The second background removal tool is the “magic wand tool,” which works similarly to Photoshop’s magic wand tool. Pixels of similar background colors can be grouped together to be classified as background. This is a powerful tool when the background pixels are uniform in color, as is the case of products that are shot in a studio and under very controlled lighting conditions. This is our recommended solution for achieving the best possible background removal results.
  2. The third background removal algorithm is based on the green screen (sometimes called chroma keying) techniques. This tool is especially efficient for doing background removal of objects that are transparent, such as bottles. For opaque objects, one drawback of this technique, is that the green background may reflect on the object, especially around edges and when the object is close to the green screen. This can cause false “transparent” regions around edges.
  3. The final set of background removal tools is based on manually removing background or adding back in foreground. We have extended the lasso tool to allow users to cut the background or add foreground, by controlling the alpha transparency channel in the manually selected region.

In conclusion, creating high quality eCommerce product photography is a critical component of any business with a web presence. At Iconasys we have developed Shutter Stream specifically for product photography and all product photography applications. In this article, we have described the background removal features and showed the four different ways in which background removal can be achieved inside Shutter Stream Product Photography Software.

About the authorDarian Muresan manages the software and hardware development at Iconasys and is a key contributor to Iconasys’ image processing algorithms.  Darian has undergraduate degrees in Electrical Engineering and Mathematics from University of Washington (Seattle, WA.); and a Masters and Ph.D. degree in Electrical and Computer Engineering from Cornell University (Ithaca, NY.).

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top