
Fashion Image Generation Pipeline Developer (Cloud-based)
- or -
Post a project like this$50/hr
- Posted:
- Proposals: 17
- Remote
- #4382688
- Expired
Working on .NET core, react, Azure, AWS, .NET, C#, WPF, Oracle , SSIS, react native, Design

118184561074983012275455121193804890886801493519927831210943333657361159864614770546906687





Description
Experience Level: Expert
Project Overview:
We’re looking for an experienced developer to build a fully automated, cloud-hosted system that transforms plain “ghost mannequin” photos of our products into five unique, styled images on a consistent model. The solution must be easy to scale up or down in the cloud as our image-generation needs fluctuate.
What You’ll Deliver:
- Automated Garment Extraction:
Take front/back mannequin photos and isolate the clothing items.
Accessory & Styling Selection:
-For each garment, automatically pick matching shoes, bags, jewelry - and for tops, also bottoms- so that each outfit looks cohesive. Accessories are optional if none make sense.
Pose, Hair & Makeup Variation:
- Generate five distinct, natural full-body shots per product with different poses, hairstyles, and make-up looks drawn from a preset library.
Consistent Model & Background:
- Ensure the same slender model appears in every image, head-to-toe, set against a seamless white backdrop.
High-Resolution Output:
- Produce JPGs at 1801×2600 pixels, ready for our online store.
End-to-End Orchestration & Scaling:
- Wire everything together so that uploading a new product triggers the entire flow automatically.
Deploy in the cloud with auto-scaling of compute resources (GPUs) to match demand.
Key Responsibilities:
- Design and implement each stage - from garment isolation through final image delivery - so there’s zero manual intervention.
- Create and manage libraries of poses, hairstyles, and make-up styles.
- Build logic for selecting and matching accessories based on product type and style.
- Integrate an AI-based image generator that applies the clothing and styling to the model.
- Set up automatic workflows that start when a new product image arrives, handle retries/failures, and notify on errors.
- Deploy and operate the solution in the cloud, ensuring it can easily spin up GPUs or other resources as needed.
- Provide monitoring and simple dashboards so we can track progress and spot issues.
What We’re Looking For:
- Proven experience building automated image-processing or AI pipelines
- Strong background in generative AI for images.
- Comfort with setting up cloud-based workflows and auto-scaling compute resources.
- Familiarity with end-to-end deployment: triggering on new inputs, handling retries, and updating databases or storage with results.
- Excellent Python development skills and ability to integrate multiple services into a single automated flow.
- Clear communication in English and ability to document design and operation steps.
We’re looking for an experienced developer to build a fully automated, cloud-hosted system that transforms plain “ghost mannequin” photos of our products into five unique, styled images on a consistent model. The solution must be easy to scale up or down in the cloud as our image-generation needs fluctuate.
What You’ll Deliver:
- Automated Garment Extraction:
Take front/back mannequin photos and isolate the clothing items.
Accessory & Styling Selection:
-For each garment, automatically pick matching shoes, bags, jewelry - and for tops, also bottoms- so that each outfit looks cohesive. Accessories are optional if none make sense.
Pose, Hair & Makeup Variation:
- Generate five distinct, natural full-body shots per product with different poses, hairstyles, and make-up looks drawn from a preset library.
Consistent Model & Background:
- Ensure the same slender model appears in every image, head-to-toe, set against a seamless white backdrop.
High-Resolution Output:
- Produce JPGs at 1801×2600 pixels, ready for our online store.
End-to-End Orchestration & Scaling:
- Wire everything together so that uploading a new product triggers the entire flow automatically.
Deploy in the cloud with auto-scaling of compute resources (GPUs) to match demand.
Key Responsibilities:
- Design and implement each stage - from garment isolation through final image delivery - so there’s zero manual intervention.
- Create and manage libraries of poses, hairstyles, and make-up styles.
- Build logic for selecting and matching accessories based on product type and style.
- Integrate an AI-based image generator that applies the clothing and styling to the model.
- Set up automatic workflows that start when a new product image arrives, handle retries/failures, and notify on errors.
- Deploy and operate the solution in the cloud, ensuring it can easily spin up GPUs or other resources as needed.
- Provide monitoring and simple dashboards so we can track progress and spot issues.
What We’re Looking For:
- Proven experience building automated image-processing or AI pipelines
- Strong background in generative AI for images.
- Comfort with setting up cloud-based workflows and auto-scaling compute resources.
- Familiarity with end-to-end deployment: triggering on new inputs, handling retries, and updating databases or storage with results.
- Excellent Python development skills and ability to integrate multiple services into a single automated flow.
- Clear communication in English and ability to document design and operation steps.

Yan J.
0% (0)Projects Completed
-
Freelancers worked with
-
Projects awarded
0%
Last project
16 Oct 2025
Poland
New Proposal
Login to your account and send a proposal now to get this project.
Log inClarification Board Ask a Question
-
My Clarification Questions Yan:
Do you already have a pose, hairstyle, and makeup reference library, or should I create one from scratch?
What type of cloud infrastructure are you currently using, if any?
Should the system support different product categories (e.g., menswear, kids) in the future?
Are you looking for continuous support after deployment or a handover-only setup?
Is the consistent model image preselected, or should it be AI-generated from a template?
1131779
We collect cookies to enable the proper functioning and security of our website, and to enhance your experience. By clicking on 'Accept All Cookies', you consent to the use of these cookies. You can change your 'Cookies Settings' at any time. For more information, please read ourCookie Policy
Cookie Settings
Accept All Cookies