← Search

Tomasz Szczepanik

2 accepted papers

2026

Affine-Equivariant Kernel Space Encoding for NeRF Editing

ICML 2026poster

Neural scene representations achieve high-fidelity rendering by encoding 3D scenes as continuous functions, but their latent spaces are typically implicit and globally entangled, making localized editing and physically grounded manipulation difficult. While several works introduce explicit control s…

Cited by 0SourceScholar
2025

MiraGe: Editable 2D Images using Gaussian Splatting

ICML 2025poster

Implicit Neural Representations (INRs) approximate discrete data through continuous functions and are commonly used for encoding 2D images. Traditional image-based INRs employ neural networks to map pixel coordinates to RGB values, capturing shapes, colors, and textures within the network’s weights.…