Color-Correcting Product Photo Enhancer
Budget: ₹12,500 – ₹37,500 INR
I’m building a small, focused tool that automatically fixes colour issues in product photos so the final images look true-to-life on any screen or marketplace. The entire aim is straightforward: take an existing shot, analyse its colours, and return a new file whose tones and white balance match the real product without introducing artefacts or noise.
Here’s what I need from you:
• An algorithm or lightweight script (Python, OpenCV, or your preferred stack) that detects colour cast and applies accurate correction in a single pass.
• Batch processing support so hundreds of images can be processed in one go.
• Simple inputs—JPEG or PNG—and outputs that retain original resolution.
• A brief usage guide plus two or three before-and-after samples using generic product images to prove the improvement.
Acceptance criteria
• Average colour error reduced enough that side-by-side comparisons look visually identical to the physical item.
• Processing time under five seconds per 4-megapixel image on a typical desktop.
• Clean, well-commented code so I can extend or embed the function later.
If you already have experience with colour science, ICC profiles, or camera RAW workflows, that’s a strong bonus. Let me know how you’d approach the correction logic and any libraries you plan to use, and include a quick turnaround estimate.
Here’s what I need from you:
• An algorithm or lightweight script (Python, OpenCV, or your preferred stack) that detects colour cast and applies accurate correction in a single pass.
• Batch processing support so hundreds of images can be processed in one go.
• Simple inputs—JPEG or PNG—and outputs that retain original resolution.
• A brief usage guide plus two or three before-and-after samples using generic product images to prove the improvement.
Acceptance criteria
• Average colour error reduced enough that side-by-side comparisons look visually identical to the physical item.
• Processing time under five seconds per 4-megapixel image on a typical desktop.
• Clean, well-commented code so I can extend or embed the function later.
If you already have experience with colour science, ICC profiles, or camera RAW workflows, that’s a strong bonus. Let me know how you’d approach the correction logic and any libraries you plan to use, and include a quick turnaround estimate.