RaPD: Resolution-Agnostic Pixel Diffusion via Semantics-Enriched Implicit Representations

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Ge, Yanhao, Guan, Shanyan, Wang, Weihao, Tai, Ying, You, Mingyu
Format:	Preprint
Published:	2026
Subjects:	Computer Vision and Pattern Recognition Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

_version_	1866910223721037824
author	Ge, Yanhao Guan, Shanyan Wang, Weihao Tai, Ying You, Mingyu
author_facet	Ge, Yanhao Guan, Shanyan Wang, Weihao Tai, Ying You, Mingyu
contents	Natural images are continuous, yet most generative models synthesize them on discrete grids, limiting resolution-flexible generation. Continuous neural fields enable resolution-free rendering, but prior methods introduce continuity only at the decoding stage as an interpolation module, leaving the generative latent space discretized and reconstruction-oriented. We propose RaPD (Resolution-agnostic Pixel Diffusion), which performs diffusion in a continuous Neural Image Field (NIF) latent space. RaPD bridges this reconstruction-generation gap with Semantic Representation Guidance for generation-aware latent learning and a Coordinate-Queried Attention Renderer for coordinate-conditioned, scale-aware rendering. A single denoised latent can be rendered at arbitrary resolutions by changing only the query coordinates, keeping diffusion cost fixed. Experiments demonstrate superior generation quality and resolution scalability.
format	Preprint
id	arxiv_https___arxiv_org_abs_2605_15908
institution	arXiv
publishDate	2026
record_format	arxiv
spellingShingle	RaPD: Resolution-Agnostic Pixel Diffusion via Semantics-Enriched Implicit Representations Ge, Yanhao Guan, Shanyan Wang, Weihao Tai, Ying You, Mingyu Computer Vision and Pattern Recognition Artificial Intelligence Natural images are continuous, yet most generative models synthesize them on discrete grids, limiting resolution-flexible generation. Continuous neural fields enable resolution-free rendering, but prior methods introduce continuity only at the decoding stage as an interpolation module, leaving the generative latent space discretized and reconstruction-oriented. We propose RaPD (Resolution-agnostic Pixel Diffusion), which performs diffusion in a continuous Neural Image Field (NIF) latent space. RaPD bridges this reconstruction-generation gap with Semantic Representation Guidance for generation-aware latent learning and a Coordinate-Queried Attention Renderer for coordinate-conditioned, scale-aware rendering. A single denoised latent can be rendered at arbitrary resolutions by changing only the query coordinates, keeping diffusion cost fixed. Experiments demonstrate superior generation quality and resolution scalability.
title	RaPD: Resolution-Agnostic Pixel Diffusion via Semantics-Enriched Implicit Representations
topic	Computer Vision and Pattern Recognition Artificial Intelligence
url	https://arxiv.org/abs/2605.15908