Seed Selection for Human-Oriented Image Reconstruction via Guided Diffusion

TL;DR

This paper introduces a seed selection technique for diffusion-based image reconstruction that enhances image quality without increasing bitrate by choosing the best seed from multiple candidates based on early diffusion outputs.

Contribution

It proposes a novel seed selection method that improves image quality in diffusion models without additional bitrate or computational cost, using early diffusion outputs for efficiency.

Findings

Outperforms baseline with single random seed in quality metrics

Uses early diffusion outputs to efficiently select optimal seed

Enhances human-oriented image reconstruction without extra bitrate

Abstract

Conventional methods for scalable image coding for humans and machines require the transmission of additional information to achieve scalability. A recent diffusion-based approach avoids this by generating human-oriented images from machine-oriented images without extra bitrate. However, it utilizes a single random seed, which may lead to suboptimal image quality. In this paper, we propose a seed selection method that identifies the optimal seed from multiple candidates to improve image quality without increasing the bitrate. To reduce the computational cost, selection is performed based on intermediate outputs obtained from early steps of the reverse diffusion process. Experimental results demonstrate that our proposed method outperforms the baseline, which uses a single random seed without selection, across multiple evaluation metrics.