2000 character limit reached
Robust Average Networks for Monte Carlo Denoising (2310.04080v3)
Published 6 Oct 2023 in cs.GR
Abstract: 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.
- TensorFlow: Large-Scale Machine Learning on Heterogeneous Systems. https://www.tensorflow.org/
- Kernel-predicting convolutional networks for denoising Monte Carlo renderings. ACM Trans. Graph. 36, 4 (2017), 97–1.
- A non-local algorithm for image denoising. In CVPR’05, Vol. 2. 60–65 vol. 2. https://doi.org/10.1109/CVPR.2005.38
- Interactive Reconstruction of Monte Carlo Image Sequences Using a Recurrent Denoising Autoencoder. ACM Trans. Graph. 36, 4 (2017).
- Weakly-Supervised Contrastive Learning in Path Manifold for Monte Carlo Image Reconstruction. ACM Trans. Graph. 40, 4, Article 38 (jul 2021), 14 pages. https://doi.org/10.1145/3450626.3459876
- Image Denoising by Sparse 3-D Transform-Domain Collaborative Filtering. IEEE Transactions on Image Processing 16, 8 (2007), 2080–2095. https://doi.org/10.1109/TIP.2007.901238
- ImageNet: A large-scale hierarchical image database. In 2009 IEEE conference on computer vision and pattern recognition. IEEE, 248–255.
- Image Style Transfer Using Convolutional Neural Networks. In 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2414–2423. https://doi.org/10.1109/CVPR.2016.265
- Sample-based Monte Carlo Denoising Using a Kernel-splatting Network. ACM Trans. Graph. 38, 4 (2019), 125:1–125:12.
- Neural Temporal Adaptive Sampling and Denoising. Computer Graphics Forum 39, 2 (2020), 147–155.
- Guided Image Filtering. IEEE transactions on pattern analysis and machine intelligence 35 (06 2013), 1397–1409.
- Deep Residual Learning for Image Recognition. In CVPR. 770–778. https://doi.org/10.1109/CVPR.2016.90
- Perceptual losses for real-time style transfer and super-resolution. In European conference on computer vision. 694–711. http://arxiv.org/abs/1603.08155
- A.C. Kokaram. 2014. Motion Picture Restoration: Digital Algorithms for Artefact Suppression in Degraded Motion Picture Film and Video. Springer London.
- Noise2Noise: Learning Image Restoration without Clean Data. CoRR abs/1803.04189 (2018). arXiv:1803.04189 http://arxiv.org/abs/1803.04189
- Burst Denoising with Kernel Prediction Networks. In CVPR. 2502–2510. https://arxiv.org/abs/1712.02327
- Karen Simonyan and Andrew Zisserman. 2014. Very Deep Convolutional Networks for Large-Scale Image Recognition. https://arxiv.org/abs/1409.1556
- Denoising with kernel prediction and asymmetric loss functions. ACM Trans. Graph. 37, 4 (2018), 1–15.
- An Improved Algorithm for TV-L1 Optical Flow. Springer-Verlag, Berlin, Heidelberg, 23–45.
- Monte Carlo Denoising via Auxiliary Feature Guided Self-Attention. ACM Trans. Graph. 40, 6, Article 273 (dec 2021), 13 pages. https://doi.org/10.1145/3478513.3480565
- Deep Compositional Denoising for High-quality Monte Carlo Rendering. Computer Graphics Forum (2021). https://doi.org/10.1111/cgf.14337
- Deep Compositional Denoising on Frame Sequences. In Eurographics Symposium on Rendering (EGSR).
- Recent Advances in Adaptive Sampling and Reconstruction for Monte Carlo Rendering. Comput. Graph. Forum 34, 2 (may 2015), 667–681.