analiza aiqbench
This research introduces QuADA-GS, a method that resolves the efficiency-quality trade-off in arbitrary-scale super-resolution by dynamically allocating Gaussian primitives based on image content rather than uniform grids. By predicting sparse feature distributions directly from low-resolution inputs, the approach significantly reduces memory and inference costs while maintaining high-fidelity reconstruction where structural complexity demands it. This advancement represents a meaningful step forward in optimizing 3D representation learning for scalable AI applications.
Adaptive Densification for High-Fidelity and Efficient Sparse Gaussian Splatting in Arbitrary-Scale Super-Resolution
arXiv:2606.29400v2 Announce Type: replace-cross Abstract: Arbitrary-Scale Super-Resolution (ASR) aims to reconstruct high-resolution images at any continuous magnification. While 2D Gaussian Splatting (GS) has recently shown great promise for ASR, current methods struggle to balance visual quality and computational cost. Approaches targeting high fidelity rely on powerful backbones and uniform, highly dense Gaussian grids, leading to prohibitive memory and inference costs. Conversely, methods prioritizing efficiency aggressively simplify their architectures, severely compromising visual quality. To bridge this gap, we observe that a core capability of GS remains largely underexplored in ASR: the potential for dynamic densification, i.e., the spatially adaptive allocation of Gaussians based on image content. Unlike standard scene fitting, where densification is guided by a known ground truth, applying this to ASR is highly non-trivial because the high-resolution target is exactly what the model must predict. To address this challenge, we propose QuADA-GS, an approach that retains a powerful representational backbone but autonomously predicts where to allocate Gaussians relying strictly on the low-resolution input. By adopting a sparse approach, QuADA-GS refines features and increases Gaussian density strictly where structural complexity demands it. Because this adaptive allocation produces a non-uniform hierarchical topology, we introduce a novel, highly efficient communication mechanism to process these sparse features, bypassing standard dense bottlenecks. Extensive experiments indicate that our approach successfully balances visual quality and computational requirements, providing an improved and competitive trade-off for ASR.
Abstract:Arbitrary-Scale Super-Resolution (ASR) aims to reconstruct high-resolution images at any continuous magnification. While 2D Gaussian Splatting (GS) has recently shown great promise for ASR, current methods struggle to balance visual quality and computational cost. Approaches targeting high fidelity rely on powerful backbones and uniform, highly dense Gaussian grids, leading to prohibitive memory and inference costs. Conversely, methods prioritizing efficiency aggressively simplify their architectures, severely compromising visual quality. To bridge this gap, we observe that a core capability of GS remains largely underexplored in ASR: the potential for dynamic densification, i.e., the spatially adaptive allocation of Gaussians based on image content. Unlike standard scene fitting, where densification is guided by a known ground truth, applying this to ASR is highly non-trivial because the high-resolution target is exactly what the model must predict. To address this challenge, we propose QuADA-GS, an approach that retains a powerful representational backbone but autonomously predicts where to allocate Gaussians relying strictly on the low-resolution input. By adopting a sparse approach, QuADA-GS refines features and increases Gaussian density strictly where structural complexity demands it. Because this adaptive allocation produces a non-uniform hierarchical topology, we introduce a novel, highly efficient communication mechanism to process these sparse features, bypassing standard dense bottlenecks. Extensive experiments indicate that our approach successfully balances visual quality and computational requirements, providing an improved and competitive trade-off for ASR.
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Graphics (cs.GR)
arXiv:2606.29400 [cs.CV]
(or arXiv:2606.29400v2 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2606.29400 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 324 cuvinte
Subiecte conexe
LightSplat: Real-Time High-Fidelity 3D Gaussian SLAM with Loop Closure
arXiv · 9 sept. 2026
Adaptive multi-resolution Gaussian processes: Scalable exact inference with naturally data-sparse covariance matrices
arXiv · 28 sept. 2026
OpenSplatGraph: From Dense Semantic Maps to Structured Scene Graphs for Open-Vocabulary Robot Perception
arXiv · 7 oct. 2026
LatentAM: Real-Time, Large-Scale Latent Gaussian Attention Mapping via Online Dictionary Learning
arXiv · 17 aug. 2026