HL-OutPaint: Coarse-to-Fine Video Outpainting for High-Resolution Long-Range Videos
Summary
HL-OutPaint is a coarse-to-fine video outpainting framework for high-resolution long-range videos, using global coarse guidance to enable large spatial extrapolation while maintaining spatio-temporal consistency.
View Cached Full Text
Cached at: 06/01/26, 03:20 PM
Paper page - HL-OutPaint: Coarse-to-Fine Video Outpainting for High-Resolution Long-Range Videos
Source: https://huggingface.co/papers/2605.17543
Abstract
HL-OutPaint is a high-resolution video outpainting framework that uses a coarse-to-fine strategy with global coarse guidance to enable large spatial extrapolation and long sequence generation while maintaining spatio-temporal consistency.
Video outpaintinggenerates plausible visual content beyond the original spatial extent of a video, playing a key role in adapting videos to diverse display formats. To support such use cases, it must enable largespatial extrapolationover long sequences. However, most existing methods address only one of these challenges or lack explicit mechanisms for ensuring globalspatio-temporal consistency, leading to notable limitations. In this paper, we propose HL-OutPaint, a high-resolutionvideo outpaintingframework for long sequences. Our approach follows acoarse-to-fine strategywith a two-stage pipeline. We first constructGlobal Coarse Guidance(GCG), a low-resolution representation that captures global structure and dominant motion across the video. Unlike naive downsampling, GCG is built via a novelglobal-local frame swapping mechanismthat couples sparse global keyframes with local temporal windows and exchanges information during sampling. This enables GCG to encode both long-term structural consistency and short-term temporal dynamics in a unified representation. Guided by this representation, HL-OutPaint then performs high-resolution outpainting to generate spatially detailed and temporally consistent content. By separating global structure modeling from fine-grained synthesis, our framework achieves stable, coherent generation for large spatial expansion andlong video sequences. Extensive experiments show that HL-OutPaint outperforms existing methods in challenging scenarios involving widespatial extrapolationandlong video sequences.
View arXiv pageView PDFProject pageAdd to collection
Get this paper in your agent:
hf papers read 2605\.17543
Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash
Models citing this paper0
No model linking this paper
Cite arxiv.org/abs/2605.17543 in a model README.md to link it from this page.
Datasets citing this paper0
No dataset linking this paper
Cite arxiv.org/abs/2605.17543 in a dataset README.md to link it from this page.
Spaces citing this paper0
No Space linking this paper
Cite arxiv.org/abs/2605.17543 in a Space README.md to link it from this page.
Collections including this paper0
No Collection including this paper
Add this paper to acollectionto link it from this page.
Similar Articles
Video outpainting is getting really good
Video outpainting techniques have significantly advanced, enabling better extrapolation of video content beyond original boundaries.
DALL·E: Introducing outpainting
OpenAI introduces Outpainting for DALL·E, enabling users to extend generated or uploaded images to create large-scale images in any aspect ratio while maintaining visual consistency with shadows, reflections, and textures.
Sketch Then Paint: Hierarchical Reinforcement Learning for Diffusion Multi-Modal Large Language Models
This paper proposes HT-GRPO, a hierarchical reinforcement learning method for diffusion multi-modal large language models that uses a sketch-then-paint training scheme and hierarchical credit assignment to improve image generation quality and reward alignment.
HDR Video Generation via Latent Alignment with Logarithmic Encoding
This paper presents a method for HDR video generation by leveraging pretrained generative models through logarithmic encoding alignment and camera-mimicking degradation training, enabling effective HDR synthesis without architectural redesign. The approach demonstrates that HDR generation can be achieved simply by adapting existing models to a representation naturally aligned with their learned priors.
KVPO: ODE-Native GRPO for Autoregressive Video Alignment via KV Semantic Exploration
KVPO introduces an ODE-native online GRPO framework that aligns streaming autoregressive video generators with human preferences using causal-semantic KV cache exploration and a velocity-field surrogate policy, achieving consistent improvements in visual quality and alignment.