RobustNeRF: Ignoring Distractors with Robust Losses
Sabour, Sara, Vora, Suhani, Duckworth, Daniel, Krasin, Ivan, Fleet, David J., Tagliasacchi, Andrea
–arXiv.org Artificial Intelligence
Neural radiance fields (NeRF) excel at synthesizing new views given multi-view, calibrated images of a static scene. When scenes include distractors, which are not persistent during image capture (moving objects, lighting variations, shadows), artifacts appear as view-dependent effects or 'floaters'. To cope with distractors, we advocate a form of robust estimation for NeRF training, modeling distractors in training data as outliers of an optimization problem. Our method successfully removes outliers from a scene and improves upon our baselines, on synthetic and real-world scenes. Our technique is simple to incorporate in modern NeRF frameworks, with few hyper-parameters. It does not assume a priori knowledge of the types of distractors, and is instead focused on the optimization problem rather than pre-processing or modeling transient objects. More results on our page https://robustnerf.github.io/public.
arXiv.org Artificial Intelligence
Feb-1-2023
- Country:
- Asia (0.28)
- North America > Canada (0.28)
- Genre:
- Research Report (0.50)
- Industry:
- Media > Photography (0.48)
- Technology: