UniSHARP: Universal Sharp Monocular View Synthesis
Paper Overview
- Research Area: CV (Computer Vision)
- Authors: Meixi Song, Dizhe Zhang, Hao Ren
- Published: 2025-06-11
- arXiv: 2506.08646
- UniSHARP performs implicit alignment in both feature space and Gaussian space.
- Gaussian primitives are arranged along rays and radial distances in a ray-based universal representation.
- 2D semantic and 3D spatial features extracted from UniK3D-inspired encoders are jointly decoded to generate the complete Gaussian cloud.
- A new benchmark covering diverse scenes and imaging systems is introduced, stratified by field of view (FoV) for fine-grained evaluation of universal monocular rendering.
Summary
This work extends SHARP, a popular photorealistic view synthesis method, to universal monocular rendering across a continuum of camera systems — from conventional perspective cameras to wide-field-of-view, fisheye, and omnidirectional panoramic settings. To overcome SHARP's pinhole-specific assumptions, the key idea is to align various images in a unified omnidirectional latent space.
Key Contributions
Results
Extensive experiments demonstrate that UniSHARP significantly outperforms existing methods across camera settings.
Original Abstract (excerpt)
> In this work, we focus on extending SHARP, the popular photorealistic view synthesis method, for universal monocular rendering across a continuum of camera systems, from conventional perspective cameras to wide-field-of-view, fisheye and omnidirectional panoramic settings. To overcome the pinhole-specific assumptions of SHARP, our key idea is to align various images in a unified omnidirectional latent space. Thus, we propose UniSHARP, which performs implicit alignment in both feature and Gaussian spaces...
---
*Auto-collected on 2026-06-09*