I will build a custom image enhancement or super resolution model in pytorch


About this gig
I'll build a production-ready image enhancement solution using state-of-the-art deep learning models (Real-ESRGAN, GFPGAN).
I have a PhD in Computer Vision and 5 years of experience at Samsung Research America where I built super resolution and diffusion-based image restoration systems.
What I deliver:
Working PyTorch model (pre-trained or integrated into your pipeline)
Clean inference script you can run immediately
FastAPI endpoint (Standard and Premium)
Before/after image comparison with PSNR/SSIM metrics (Premium)
Live demo of my work: huggingface.co/spaces/sumitjh/image-super-resolution
Message me before ordering to discuss your specific use case.
Get to know Sumit J
Computer Vision Engineer
- FromUnited States
- Member sinceMay 2026
Languages
English, Nepali, Hindi
FAQ
Can you work with my custom dataset?
For the Standard and Premium packages I can adapt the model pipeline to work with your specific image domain (medical, satellite, e-commerce, etc.). Message me with details before ordering.
What image formats do you support?
JPG, PNG, and TIFF. If you have a different format, message me and we can discuss.

