English

Efficient High-Resolution Image Editing with Hallucination-Aware Loss and Adaptive Tiling

Computer Vision and Pattern Recognition 2025-10-09 v1 Artificial Intelligence Machine Learning

Abstract

High-resolution (4K) image-to-image synthesis has become increasingly important for mobile applications. Existing diffusion models for image editing face significant challenges, in terms of memory and image quality, when deployed on resource-constrained devices. In this paper, we present MobilePicasso, a novel system that enables efficient image editing at high resolutions, while minimising computational cost and memory usage. MobilePicasso comprises three stages: (i) performing image editing at a standard resolution with hallucination-aware loss, (ii) applying latent projection to overcome going to the pixel space, and (iii) upscaling the edited image latent to a higher resolution with adaptive context-preserving tiling. Our user study with 46 participants reveals that MobilePicasso not only improves image quality by 18-48% but reduces hallucinations by 14-51% over existing methods. MobilePicasso demonstrates significantly lower latency, e.g., up to 55.8×\times speed-up, yet with a small increase in runtime memory, e.g., a mere 9% increase over prior work. Surprisingly, the on-device runtime of MobilePicasso is observed to be faster than a server-based high-resolution image editing model running on an A100 GPU.

Cite

@article{arxiv.2510.06295,
  title  = {Efficient High-Resolution Image Editing with Hallucination-Aware Loss and Adaptive Tiling},
  author = {Young D. Kwon and Abhinav Mehrotra and Malcolm Chadwick and Alberto Gil Ramos and Sourav Bhattacharya},
  journal= {arXiv preprint arXiv:2510.06295},
  year   = {2025}
}

Comments

Preprint. Under review

R2 v1 2026-07-01T06:22:17.482Z