Gaussian Splatting has revolutionized the field of Novel View Synthesis (NVS) with faster training and real-time rendering. However, its reconstruction fidelity still trails behind the powerful radiance models such as Zip-NeRF. Motivated by our theoretical result that both queries (such as coordinates) and neighborhood are important to learn high-fidelity signals, this paper proposes Queried-Convolutions (Qonvolutions), a simple yet powerful modification using the neighborhood properties of convolution. Qonvolutions convolve a low-fidelity signal with queries to output residual and achieve high-fidelity reconstruction. We empirically demonstrate that combining Gaussian splatting with Qonvolution neural networks (QNNs) results in state-of-the-art NVS on real-world scenes, even outperforming Zip-NeRF on image fidelity. QNNs also enhance performance of 1D regression, 2D regression and 2D super-resolution tasks.
@article{arxiv.2512.12898,
title = {Towards High-Fidelity Gaussian Splatting with Queried-Convolution Neural Networks},
author = {Abhinav Kumar and Tristan Aumentado-Armstrong and Lazar Valkov and Gopal Sharma and Alex Levinshtein and Radek Grzeszczuk and Suren Kumar},
journal= {arXiv preprint arXiv:2512.12898},
year = {2026}
}