arXiv Open Access 2023

Robust Average Networks for Monte Carlo Denoising

Javor Kalojanov Kimball Thurston

Lihat Sumber

Abstrak

We present a method for converting denoising neural networks from spatial into spatio-temporal ones by modifying the network architecture and loss function. We insert Robust Average blocks at arbitrary depths in the network graph. Each block performs latent space interpolation with trainable weights and works on the sequence of image representations from the preceding spatial components of the network. The temporal connections are kept live during training by forcing the network to predict a denoised frame from subsets of the input sequence. Using temporal coherence for denoising improves image quality and reduces temporal flickering independent of scene or image complexity.

Topik & Kata Kunci

cs.GR

Penulis (2)

Javor Kalojanov

Kimball Thurston

Format Sitasi

APA MLA BibTeX

Kalojanov, J., Thurston, K. (2023). Robust Average Networks for Monte Carlo Denoising. https://arxiv.org/abs/2310.04080

Akses Cepat

Lihat di Sumber

Informasi Jurnal

Tahun Terbit: 2023
Bahasa: en
Sumber Database: arXiv
Akses: Open Access ✓