Abstract We present a kernel‐predicting neural denoising method for path‐traced deep‐Z images that facilitates their usage in animation and visual effects production. Deep‐Z images provide enhanced flexibility during compositing as they contain color, opacity, and other rendered data at multiple depth‐resolved bins within each pixel. However, they are subject to noise, and rendering until convergence is prohibitively expensive. The current state of the art in deep‐Z denoising yields objectionable artifacts, and current neural denoising methods are incapable of handling the variable number of depth bins in deep‐Z images. Our method extends kernel‐predicting convolutional neural networks to address the challenges stemming from denoising deep‐Z images. We propose a hybrid reconstruction architecture that combines the depth‐resolved reconstruction at each bin with the flattened reconstruction at the pixel level. Moreover, we propose depth‐aware neighbor indexing of the depth‐resolved inputs to the convolution and denoising kernel application operators, which reduces artifacts caused by depth misalignment present in deep‐Z images. We evaluate our method on a production‐quality deep‐Z dataset, demonstrating significant improvements in denoising quality and performance compared to the current state‐of‐the‐art deep‐Z denoiser. By addressing the significant challenge of the cost associated with rendering path‐traced deep‐Z images, we believe that our approach will pave the way for broader adoption of deep‐Z workflows in future productions.

We propose an adaptive sampling and reconstruction method for offline Monte Carlo rendering. Our method produces sampling maps constrained by a user-defined budget that minimize the expected future denoising error. Compared to other state-of-the-art methods, which produce the necessary training data on the fly by composing pre-rendered images, our method samples from analytic noise distributions instead. These distributions are compact and closely approximate the pixel value distributions stemming from Monte Carlo rendering. Our method can efficiently sample training data by leveraging only a few per-pixel statistics of the target distribution, which provides several benefits over the current state of the art. Most notably, our analytic distributions' modeling accuracy and sampling efficiency increase with sample count, essential for high-quality offline rendering. Although our distributions are approximate, our method supports joint end-to-end training of the sampling and denoising networks. Finally, we propose the addition of a global summary module to our architecture that accumulates valuable information from image regions outside of the network's receptive field. This information discourages sub-optimal decisions based on local information. Our evaluation against other state-of-the-art neural sampling methods demonstrates denoising quality and data efficiency improvements.

We present a modular convolutional architecture for denoising rendered images. We expand on the capabilities of kernel-predicting networks by combining them with a number of task-specific modules, and optimizing the assembly using an asymmetric loss. The source-aware encoder---the first module in the assembly---extracts low-level features and embeds them into a common feature space, enabling quick adaptation of a trained network to novel data. The spatial and temporal modules extract abstract, high-level features for kernel-based reconstruction, which is performed at three different spatial scales to reduce low-frequency artifacts. The complete network is trained using a class of asymmetric loss functions that are designed to preserve details and provide the user with a direct control over the variance-bias trade-off during inference. We also propose an error-predicting module for inferring reconstruction error maps that can be used to drive adaptive sampling. Finally, we present a theoretical analysis of convergence rates of kernel-predicting architectures, shedding light on why kernel prediction performs better than synthesizing the colors directly, complementing the empirical evidence presented in this and previous works. We demonstrate that our networks attain results that compare favorably to state-of-the-art methods in terms of detail preservation, low-frequency noise removal, and temporal stability on a variety of production and academic datasets.

We propose a simple method to enhance the image quality of modern Bayer pattern based cameras that are offering an optional sub-pixel accurate sensor shift to capture full rgb images of static scenes. By capturing a series of image pairs of the same, unaltered scene, once captured with the Bayer pattern and once with the full rgb image data, a database of corresponding images can be generated in which the former contains artifacts resulting from the spatial interpolation which is absent in the latter. Using this data, we train a convolutional neural network (CNN) to generate a camera dependent image processing operation which reduces the image artifacts and enhances the image quality to approximate the quality of the full rgb image within a single exposure, even if moving scenes are captured. We present a simple, do-it-yourself method to capture and pre-process the data, train the network, and to enhance the images. An evaluation using several image quality assessment methods shows the effectiveness of the proposed method.

Coloring books capture the imagination of children and provide them with one of their earliest opportunities for creative expression. However, given the proliferation and popularity of digital devices, real-world activities like coloring can seem unexciting, and children become less engaged in them. Augmented reality holds unique potential to impact this situation by providing a bridge between real-world activities and digital enhancements. In this paper, we present an augmented reality coloring book App in which children color characters in a printed coloring book and inspect their work using a mobile device. The drawing is detected and tracked, and the video stream is augmented with an animated 3-D version of the character that is textured according to the child's coloring. This is possible thanks to several novel technical contributions. We present a texturing process that applies the captured texture from a 2-D colored drawing to both the visible and occluded regions of a 3-D character in real time. We develop a deformable surface tracking method designed for colored drawings that uses a new outlier rejection algorithm for real-time tracking and surface deformation recovery. We present a content creation pipeline to efficiently create the 2-D and 3-D content. And, finally, we validate our work with two user studies that examine the quality of our texturing algorithm and the overall App experience.

The decoupling of eye vergence and accommodation (V/A) has been found to negatively impact depth interpretation, visual comfort and fatigue. In this paper, we explore a hypothesis that placement of visual cues within a scene can assist a viewer in the process of maintaining the V/A decoupling. This effect is demonstrated through the use of a continuous depth plane that connects spatially distinct scene elements. Our experimental design enables us to make the following three contributions: (1) We show that a continuous depth element can improve the time it takes to transition visual attention in depth. (2) We observe that the subjective assessment of fatigue emerges before we detect a quantitative decline in performance. (3) We aim to motivate that stereoscopic 3D content creators may learn scene composition, framing and montage from visual psychophysics.