DWTNeRF: Boosting Few-shot Neural Radiance Fields via Discrete Wavelet Transform

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Nguyen, Hung, Li, Blark Runfa, Nguyen, Truong
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915435786534912
author Nguyen, Hung
Li, Blark Runfa
Nguyen, Truong
author_facet Nguyen, Hung
Li, Blark Runfa
Nguyen, Truong
contents Neural Radiance Fields (NeRF) has achieved superior performance in novel view synthesis and 3D scene representation, but its practical applications are hindered by slow convergence and reliance on dense training views. To this end, we present DWTNeRF, a unified framework based on Instant-NGP's fast-training hash encoding. It is coupled with regularization terms designed for few-shot NeRF, which operates on sparse training views. Our DWTNeRF additionally includes a novel Discrete Wavelet loss that allows explicit prioritization of low frequencies directly in the training objective, reducing few-shot NeRF's overfitting on high frequencies in earlier training stages. We also introduce a model-based approach, based on multi-head attention, that is compatible with INGP, which are sensitive to architectural changes. On the 3-shot LLFF benchmark, DWTNeRF outperforms Vanilla INGP by 15.07% in PSNR, 24.45% in SSIM and 36.30% in LPIPS. Our approach encourages a re-thinking of current few-shot approaches for fast-converging implicit representations like INGP or 3DGS.
format Preprint
id arxiv_https___arxiv_org_abs_2501_12637
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DWTNeRF: Boosting Few-shot Neural Radiance Fields via Discrete Wavelet Transform
Nguyen, Hung
Li, Blark Runfa
Nguyen, Truong
Computer Vision and Pattern Recognition
Neural Radiance Fields (NeRF) has achieved superior performance in novel view synthesis and 3D scene representation, but its practical applications are hindered by slow convergence and reliance on dense training views. To this end, we present DWTNeRF, a unified framework based on Instant-NGP's fast-training hash encoding. It is coupled with regularization terms designed for few-shot NeRF, which operates on sparse training views. Our DWTNeRF additionally includes a novel Discrete Wavelet loss that allows explicit prioritization of low frequencies directly in the training objective, reducing few-shot NeRF's overfitting on high frequencies in earlier training stages. We also introduce a model-based approach, based on multi-head attention, that is compatible with INGP, which are sensitive to architectural changes. On the 3-shot LLFF benchmark, DWTNeRF outperforms Vanilla INGP by 15.07% in PSNR, 24.45% in SSIM and 36.30% in LPIPS. Our approach encourages a re-thinking of current few-shot approaches for fast-converging implicit representations like INGP or 3DGS.
title DWTNeRF: Boosting Few-shot Neural Radiance Fields via Discrete Wavelet Transform
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2501.12637