Point-cloud denoising should suppress noise without harming geometry, classical approaches often oversmooth complex object surfaces, erasing fine details. Although recent learning-based techniques alleviate these shortcomings, they still face two critical limitations: (i) single-scale feature extractors without explicit positional encodings struggle to capture long-range dependencies, and (ii) stage-wise pipelines that update normals and coordinates separately accumulate errors. We propose an end-to-end framework that fuses multi-scale convolutions with a position-encoded Transformer and a geometry-guided feedback loop. Convolutions capture local detail, the Transformer models global relations, and residual fusion unifies their features. A key-point selector, driven by normal orthogonality and cross-scale agreement, retains only high-confidence points, while a bidirectional module jointly refines coordinates and normals under a geometric-consistency loss. On synthetic CAD and non-CAD datasets corrupted with 0.6– \(2.0\,\%\) Gaussian noise, the method achieves lower Chamfer and point-to-surface distances than five state-of-the-art baselines. Ablations confirm each component’s value, and qualitative results preserve sharp edges and high curvature, providing an efficient, practical solution for point-cloud denoising.