Denoising is an integral part of production rendering pipelines that use Monte-Carlo (MC) path tracing. Machine learning based denoisers have been proven to effectively remove the residual noise and produce a clean image. However, denoising volumetric rendering remains a problem due to the lack of useful features and large-scale volume datasets. We have seen issues such as over-blurring and temporal flickering in the denoised sequence. In this work, we modify the production renderer to generate potential features that might improve the denoising quality, and then run a state-of-the-art feature selection algorithm to detect the best combination. We collect thousands of unique volumetric scenes from our recent films to create a large dataset for training. Our evaluation shows a good amount of quality gain compared to the version currently in use.
14 August 2023 in IEEE Transactions on Visualization and Computer Graphics
Line work is a core element for the stylization of computer animations used by recent shows. However, existing stylization techniques are limited to edge treatments based on brush strokes or textures applied solely on top of curves. In this work, we propose new stylization effects by offering artists direct control over the inside and outside of surface contours. To this end, we introduce a method that creates ribbons , geometry strips of possibly varying width, that extrude from each side of the surface contour with temporally coherent orientations. Our contributions include the generation of spatially and temporally consistent normal orientations along visible contours and a trimming routine that converts arrangements of offset curves into ribbons free of intersections. We demonstrate the expressiveness and versatility of stylized ribbons by applying various effects on both character and shadow edges from animation sequences.