arXiv AI
Jun 30
Learning to Adaptively Allocate Gaussians for Arbitrary-Scale Image Super-Resolution
arXiv:2606. 29400v1 Announce Type: cross Abstract: In computer graphics, visual content is continuously warped, zoomed and resampled.
SP-MoMamba: Superpixel-driven Mixture of State Space Experts for Efficient Image Super-Resolution
SP-MoMamba introduces a superpixel-driven mixture of state space experts for efficient image super‑resolution. By grouping spatially coherent features into region‑level tokens, the Superpixel‑SSM performs global sequence modeling over compact representations, reducing redundant computation while enabling long‑range structural interaction. The Multi‑Scale Superpixel Mixture of State Space Experts further adapts to varying representation granularities, and a Local Spatial Modulation Expert refines local high‑frequency details, resulting in strong reconstruction performance with a favorable trade‑off among model size, computational cost, and inference efficiency.
Related stories
arXiv AI
Sep 10
Adaptive Densification for High-Fidelity and Efficient Sparse Gaussian Splatting in Arbitrary-Scale Super-Resolution
The paper introduces QuADA-GS, a method for Arbitrary-Scale Super-Resolution that dynamically densifies 2D Gaussian splatting based on low‑resolution input. By allocating Gaussians adaptively to structurally complex regions and employing a sparse communication mechanism, it balances high visual fidelity with lower computational cost. Experiments show that this approach achieves a competitive trade‑off between quality and efficiency for super‑resolution tasks.
arXiv Computer Vision
Aug 27
Compact Feed-Forward 3D Gaussians via Saliency-Guided Primitive Merging
The paper introduces a structure‑aware merging pipeline that consolidates per‑pixel 3D Gaussian primitives from any feed‑forward reconstruction method into a compact, content‑adaptive Gaussian set. By grouping spatially coherent Gaussians with adaptive superpixel segmentation guided by a saliency map, compressing clusters via a learned encoder, and merging representations across views using geometric overlap and feature similarity, the method reduces the number of Gaussians to about one‑twentieth of the original while preserving visual quality. A level‑of‑detail decoder allows controllable resolution, and the pipeline operates as a backbone‑agnostic post‑processing module, improving robustness and rendering efficiency.
arXiv Computer Vision
Sep 22
MambaX: Image Super-Resolution with State Predictive Control
arXiv:2511.18028v2 Announce Type: replace Abstract: Image super-resolution (SR) is a critical technology for overcoming the inherent hardware limitations of sensors. However, existing approaches main...
arXiv Computer Vision
Oct 27
UltraFlash: Accelerating Megapixel Visual Synthesis
arXiv:2610.33581v1 Announce Type: new Abstract: Megapixel visual synthesis with latent diffusion models requires operating far beyond the resolutions at which these models are typically trained. Casc...
arXiv Computer Vision
5d ago
PixelMoE: Scaling Pixel-Space Diffusion with LLM-Style Mixture-of-Experts
PixelMoE introduces a pixel‑space generative framework that applies LLM‑style Mixture‑of‑Experts (MoE) to diffusion models, enabling sparse scaling of model capacity without diffusion‑specific routing. It combines Token‑Choice routing, fine‑grained routed experts, a shared expert, and bias‑based load balancing, and adds Pixel‑Space Hierarchical Flow Matching to vary spatial resolution during denoising. Across model scales, PixelMoE outperforms dense counterparts and achieves competitive FID scores on class‑conditional ImageNet‑1K, with code and pretrained models publicly released.
arXiv Computer Vision
Oct 1
PixelUMM: Encoder-Free Unified Image and Video Understanding and Generation
PixelUMM is an encoder‑free model that unifies image and video understanding and generation directly in pixel space. It represents images as spatial patches and videos as spatiotemporal tubelets, feeding both through single‑layer linear projections into a shared multimodal backbone. The Mixture‑of‑Transformers architecture blends shared attention with task‑specific parameters, enabling autoregressive text prediction, pixel‑space flow matching, and clean‑pixel video generation, and experiments show competitive performance across tasks while providing design insights for future pixel‑space multimodal models.
arXiv AI
Jun 2
Beyond Visual Fidelity: Benchmarking Super-Resolution Models for Large-Scale Remote Sensing Imagery via Downstream Task Integration
arXiv:2605. 00310v2 Announce Type: replace-cross Abstract: Super-resolution (SR) techniques have made major advances in reconstructing high-resolution images from low-resolution inputs.
arXiv Computer Vision
Oct 27
HiFiPix: Bringing High-Fidelity Real-World Super-Resolution Back to Pixel Space
arXiv:2610.35515v1 Announce Type: new Abstract: Diffusion-based real-world image super-resolution suffers from limited reconstruction fidelity due to information loss in VAE-based latent representati...
arXiv Computer Vision
Sep 1
Keyframe-Centric State-Space Modeling for Burst Image Super-Resolution
arXiv:2503.19634v2 Announce Type: replace Abstract: Burst image super-resolution (BISR) reconstructs a high-resolution keyframe by aggregating complementary sub-pixel evidence from a short burst of l...
arXiv Machine Learning
Sep 23
SPARC: SuperPixel-Aware Region Contrastive Learning for Self-Supervised Dense Prediction
arXiv:2609.25067v1 Announce Type: cross Abstract: Self-supervised learning (SSL) has become an effective approach for learning visual representations without manual annotations. Among SSL approaches,...
Hugging Face Trending Papers
Oct 13
PixelMoE: Scaling Pixel-Space Diffusion with LLM-Style Mixture-of-Experts
PixelMoE introduces a pixel-space generative framework that applies LLM-style Mixture-of-Experts (MoE) to diffusion models, enabling sparse scaling of model capacity without diffusion-specific routing. The approach combines Token-Choice routing, fine-grained routed experts, a shared expert, and bias-based load balancing, while also employing Pixel-Space Hierarchical Flow Matching to adjust spatial resolution during denoising. PixelMoE outperforms dense counterparts at comparable active parameter counts, achieving strong FID scores on class-conditional ImageNet-1K.