CineScale: Free Lunch in High-Resolution Cinematic Visual Generation

TL;DR

CineScale introduces a new inference method that allows existing pre-trained visual diffusion models to generate high-resolution images and videos, including 8K images and 4K videos, without extensive fine-tuning.

Contribution

The paper presents CineScale, a novel inference paradigm that extends high-resolution visual generation capabilities of pre-trained models across various architectures without requiring retraining.

Findings

Enables 8K image generation without fine-tuning.

Achieves 4K video generation with minimal LoRA fine-tuning.

Outperforms existing methods in high-resolution visual synthesis.

Abstract

Visual diffusion models achieve remarkable progress, yet they are typically trained at limited resolutions due to the lack of high-resolution data and constrained computation resources, hampering their ability to generate high-fidelity images or videos at higher resolutions. Recent efforts have explored tuning-free strategies to exhibit the untapped potential higher-resolution visual generation of pre-trained models. However, these methods are still prone to producing low-quality visual content with repetitive patterns. The key obstacle lies in the inevitable increase in high-frequency information when the model generates visual content exceeding its training resolution, leading to undesirable repetitive patterns deriving from the accumulated errors. In this work, we propose CineScale, a novel inference paradigm to enable higher-resolution visual generation. To tackle the various issues introduced by the two types of video generation architectures, we propose dedicated variants tailored to each. Unlike existing baseline methods that are confined to high-resolution T2I and T2V generation, CineScale broadens the scope by enabling high-resolution I2V and V2V synthesis, built atop state-of-the-art open-source video generation frameworks. Extensive experiments validate the superiority of our paradigm in extending the capabilities of higher-resolution visual generation for both image and video models. Remarkably, our approach enables 8k image generation without any fine-tuning, and achieves 4k video generation with only minimal LoRA fine-tuning. Generated video samples are available at our website: https://eyeline-labs.github.io/CineScale/.