Neural Radiance Fields (NeRF) synthesize photorealistic novel views from dense image collections, but degrade sharply when few views are available. We attribute a specific, measurable defect to the ob...
DriftNeRF introduces a regularizer that addresses cross‑ray density disagreement in Neural Radiance Fields, a defect that causes poor performance when only a few views are available. By aggregating densities from nearby points on other rays into a stop‑gradient consensus target, DriftNeRF penalizes unsupported disagreement while preserving agreed boundaries. The method improves 3‑view PSNR by 0.42–0.65 dB on NeRF‑Synthetic, DTU, and LLFF datasets and remains competitive with sparse‑view Gaussian pipelines without requiring additional geometry or preprocessing.
NaCR: Visual Localization via NeRF-aided Camera Ray Regression proposes a unified framework that integrates Neural Radiance Fields (NeRF) with Camera Ray Regression (CRR) to improve visual localization accuracy. The method enhances the CRR baseline with three simple improvements, augments training data by synthesizing novel views from a pre‑trained NeRF, and employs a closed‑loop supervision pipeline that back‑propagates photometric rendering errors to refine predicted camera rays. A two‑stage training curriculum ensures stable convergence, and experiments on indoor and outdoor benchmarks show competitive accuracy with validated component efficacy.
Recent advances in neural scene representations enable photorealistic novel-view synthesis, yet most methods remain tightly coupled to a single rendering paradigm, limiting their versatility and integration with conventional graphics workflows. We introduce Floating Radiance Networks (FlaRe), a neural scene representation combining explicit ray-traceable geometry with continuous neural radiance functions.
arXiv:2605.10360v3 Announce Type: replace Abstract: While novel view synthesis (NVS) for dynamic scenes has seen significant progress, reconstructing temporally consistent geometric surfaces remains...
arXiv:2607. 16286v1 Announce Type: cross Abstract: The 3D geometry of real-world scene data is often incomplete.
arXiv:2609.09606v1 Announce Type: new Abstract: Neural radiance fields (NeRFs) and 3D Gaussian Splatting (3DGS) encode a scene with complementary inductive biases, but existing cross-representation d...
OX-NeRF is a new framework for 3D X‑ray tomography reconstruction from ultra‑sparse views, combining cross‑scene feature learning with scene‑specific optimisation. It uses a CNN to extract shared features, maintains multi‑resolution hash grids for each scene, and fuses these with an MLP that is jointly optimised end‑to‑end. Benchmarks on parallel‑beam and cone‑beam X‑ray datasets show that OX‑NeRF achieves significantly higher reconstruction accuracy than existing radiance field methods for scenes with ten or fewer views.
Reconstructing 3D shapes from a single image remains a fundamental yet challenging problem in computer vision. Traditional monocular 3D generation pipelines typically synthesize multiple views from a single input image before applying Neural Radiance Field (NeRF)-based reconstruction.
arXiv:2607. 00885v1 Announce Type: cross Abstract: Recent advances in neural rendering have established 3D Gaussian Splatting (3DGS) as a highly efficient representation for novel view synthesis, enabling fast training and real-time rendering with strong fidelity.
CrossDepth introduces geometry-constrained attention for multi-view surround depth estimation, addressing cross-image inconsistencies caused by varying camera intrinsics and limited receptive fields. The method conditions features on per-pixel camera-aware ray embeddings and extends pixel context via cross-image attention limited to geometrically plausible regions. Trained self-supervised with photometric consistency, it achieves better depth accuracy and consistency on DDAD and nuScenes compared to existing self-supervised approaches.
Reinforcement learning from human feedback (RLHF) for 3D generation is now established across a number of works, but most existing pipelines optimise explicit surface representations, often by converting radiance fields into meshes and training heavily on surface-supervised data. We instead fine-tune a pretrained 3D-aware generative model directly from a learned reward over radiance-field density ($σ$) values, with no externally supplied mesh or shape prior.