Background removal

An ML model that accurately separates foreground subjects from backgrounds, enabling easy and efficient generation of transparent assets

Overview

The Background remover is an advanced machine-learning model, designed to make background removal an effortless and time-saving process. By leveraging the power of deep learning algorithms, it can accurately separate foreground subjects from their backgrounds, resulting in high-quality transparent assets.

The model is built on a foundation of state-of-the-art deep learning techniques, particularly focused on semantic segmentation. It is trained on extensive datasets containing a diverse range of images, ensuring robustness and adaptability to handle various edge cases and complexities. By utilizing a multi-layered neural network architecture, it processes each pixel in an image to classify it as part of the foreground or background.

The model undergoes a rigorous training process, learning to identify different object shapes, fine details and semi-transparent elements in the images. As a result, it can accurately separate foreground subjects from their backgrounds, even in challenging scenarios.

Automatic background removal can be a game-changer for many workflows. By transforming product photos, portraits and creative artworks into transparent assets in a matter of seconds, it eliminates any fiddling with tedious image editing software or outsourcing the task to graphic designers. Professional-grade background removal can be achieved effortlessly, saving valuable time and resources.

Use cases

Automatic background removal can prove invaluable for multiple use cases, such as:

• E-commerce product catalogs - The model ensures consistent, visually appealing product images that seamlessly blend into any website or marketing material and can streamline any e-commerce business.

• Portrait photography - The background remover offers a quick and efficient way to remove distracting backgrounds from portrait shots, enabling better focus on the subject's features and expressions.

• Design projects - Designers can explore boundless creative possibilities by easily overlaying graphics, text or new backgrounds, allowing for eye-catching collages, posters, and social media posts.

• Presentations and marketing materials - Creation of professional presentations by placing images on any background, ensuring a clean and polished look that captivates the audience.

• Image localization - The model facilitates localization for global audiences by enabling easy background replacement to suit different cultural contexts and brand aesthetics.

API endpoints

An up-to-date reference with all API endpoints is available here:

Scaleflex API for Digital Asset Management (DAM), Visual AI and Media OptimizationScaleflex API

Example API responses

Input image	Input image

Output preview	Output preview

API response	API response
{ "status": "success", "version": "2.9.3", "request_uuid": "2fafe426-3c7d-4c71-9f55-c2378fd5a4fd", "result": "https://fdocs.filerobot.com/https://ask.filerobot.com/deliver/fdocs/2fafe426-3c7d-4c71-9f55-c2378fd5a4fd.png", "sha1": "b5d07a2d7648df181c94b2e2c8616ff7a91ca8c0", "method": "sfx_cpd", "file_downloaded": "face.png" }	{ "status": "success", "version": "2.9.3", "request_uuid": "fd04f383-e8a2-49af-98fd-1abf412a67b0", "result": "https://fdocs.filerobot.com/https://ask.filerobot.com/deliver/fdocs/fd04f383-e8a2-49af-98fd-1abf412a67b0.png", "sha1": "9fac9bbd65b80423f45691a7f4a3f4868c84ec55", "method": "sfx_cpd", "file_downloaded": "17043984.jpg" }