English

TrackNeRF: Bundle Adjusting NeRF from Sparse and Noisy Views via Feature Tracks

Computer Vision and Pattern Recognition 2024-08-21 v1

Abstract

Neural radiance fields (NeRFs) generally require many images with accurate poses for accurate novel view synthesis, which does not reflect realistic setups where views can be sparse and poses can be noisy. Previous solutions for learning NeRFs with sparse views and noisy poses only consider local geometry consistency with pairs of views. Closely following \textit{bundle adjustment} in Structure-from-Motion (SfM), we introduce TrackNeRF for more globally consistent geometry reconstruction and more accurate pose optimization. TrackNeRF introduces \textit{feature tracks}, \ie connected pixel trajectories across \textit{all} visible views that correspond to the \textit{same} 3D points. By enforcing reprojection consistency among feature tracks, TrackNeRF encourages holistic 3D consistency explicitly. Through extensive experiments, TrackNeRF sets a new benchmark in noisy and sparse view reconstruction. In particular, TrackNeRF shows significant improvements over the state-of-the-art BARF and SPARF by 8\sim8 and 1\sim1 in terms of PSNR on DTU under various sparse and noisy view setups. The code is available at \href{https://tracknerf.github.io/}.

Keywords

Cite

@article{arxiv.2408.10739,
  title  = {TrackNeRF: Bundle Adjusting NeRF from Sparse and Noisy Views via Feature Tracks},
  author = {Jinjie Mai and Wenxuan Zhu and Sara Rojas and Jesus Zarzar and Abdullah Hamdi and Guocheng Qian and Bing Li and Silvio Giancola and Bernard Ghanem},
  journal= {arXiv preprint arXiv:2408.10739},
  year   = {2024}
}

Comments

ECCV 2024 (supplemental pages included)

R2 v1 2026-06-28T18:17:59.411Z