UltraZoom: Generating Gigapixel Images from Regular Photos

TL;DR

UltraZoom is a system that creates high-resolution gigapixel images from casual photos by combining global low-detail images with local high-detail close-ups, using a learned object-specific upscaling model.

Contribution

It introduces a novel method for generating gigapixel images from casual photos by constructing paired datasets and adapting generative models for object-specific upscaling.

Findings

Produces seamless, photorealistic gigapixel images from minimal input.

Effectively registers close-ups within full images for accurate upscaling.

Enables pan and zoom across objects with consistent detail.

Abstract

We present UltraZoom, a system for generating gigapixel-resolution images of objects from casually captured inputs, such as handheld phone photos. Given a full-shot image (global, low-detail) and one or more close-ups (local, high-detail), UltraZoom upscales the full image to match the fine detail and scale of the close-up examples. To achieve this, we construct a per-instance paired dataset from the close-ups and adapt a pretrained generative model to learn object-specific low-to-high resolution mappings. At inference, we apply the model in a sliding window fashion over the full image. Constructing these pairs is non-trivial: it requires registering the close-ups within the full image for scale estimation and degradation alignment. We introduce a simple, robust method for getting registration on arbitrary materials in casual, in-the-wild captures. Together, these components form a system that enables seamless pan and zoom across the entire object, producing consistent, photorealistic gigapixel imagery from minimal input.