Abstract. A vast amount of soft shadow map algorithms have been presented in recent years. Most use a single sample hard shadow map together with some clever filtering technique to calculate perceptually or even physically plausible soft shadows. On the other hand there is the class of much slower algorithms that calculate physically correct soft shadows by taking and combining many samples of the light. In this paper we present a new soft shadow method that combines the benefits of these approaches. It samples the light source over multiple frames instead of a single frame, creating only a single shadow map each frame. Where temporal coherence is low we use spatial filtering to estimate additional samples to create correct and very fast soft shadows. 1

This paper introduces a framebuffer level of detail algorithm for controlling the pixel workload in an interactive rendering application. Our basic strategy is to evaluate the shading in a low resolution buffer and, in a second rendering pass, resample this buffer at the desired screen resolution. The size of the lower resolution buffer provides a trade-off between rendering time and the level of detail in the final shading. In order to reduce approximation error we use a feature-preserving reconstruction technique that more faithfully approximates the shading near depth and normal discontinuities. We also demonstrate how intermediate components of the shading can be selectively resized to provide finer-grained control over resource allocation. Finally, we introduce a simple control mechanism that continuously adjusts the amount of resizing necessary to maintain a target framerate. These techniques do not require any preprocessing, are straightforward to implement on modern GPUs, and are shown to provide significant performance gains for several pixel-bound scenes. Categories and Subject Descriptors (according to ACM CCS): I.3.3 [Computer Graphics]: Picture/Image Generation 1. Introduction and Related

Real-time rendering imposes the challenging task of creating a new rendering of an input scene at least 60 times a second. Although computer graphics hardware has made staggering advances in terms of speed and freedom of programmability, there still exist a number of algorithms that are too expensive to be calculated in this time budget, like exact shadows or an exact global illumination solution. One way to circumvent this hard time limit is to capitalize on temporal coherence to formulate algorithms incremental in time. The main thesis of this work is that temporal coherence is a characteristic of real-time graphics that can be used to redesign well-known rendering methods to become faster, while exhibiting better visual fidelity. To this end we present our adaptations of algorithms from the fields of exact hard shadows, physically correct soft shadows and fast discrete LOD blending, in which we have successfully incorporated temporal coherence. Additionally, we provide a detailed context of previous work not only in the

Balancing the trade off between the spatial and temporal quality of interactive computer graphics imagery is one of the fundamental design challenges in the construction of rendering systems. Inexpensive interactive rendering hardware may deliver a high level of temporal performance if the level of spatial image quality is sufficiently constrained. In these cases, the spatial fidelity level is an independent parameter of the system and temporal performance is a dependent variable. The spatial quality parameter is selected for the system by the designer based on the anticipated graphics workload. Interactive ray tracing is one example; the algorithm is often selected due to its ability to deliver a high level of spatial fidelity, and the relatively lower level of temporal performance is readily accepted. This dissertation proposes an algorithm to perform fine-grained adjustments to the trade off between the spatial quality of images produced by an interactive renderer, and the temporal performance or quality of the rendered image sequence. The approach first determines the minimum amount of sampling work necessary to achieve a certain fidelity level, and then allows the surplus capacity to be directed towards spatial or temporal fidelity improvement. The algorithm consists of an efficient parallel spatial and temporal adaptive rendering mechanism and a control optimization

Abstract. Multi-frame rate systems decouple viewing from rendering in an asynchronous pipeline. Multiple GPUs can be used as frame sources, while a primary GPU is responsible for viewing and display update. Conventionally, the last rendering result is used for display. However, modern GPUs are equipped with a fairly large amount of memory which allows frames to be cached in video memory. As long as the data is static, caching allows for a more sophisticated reference frame selection that increases the output quality. With a growing frame database, images for most viewpoints can be queried from the cache and the system converges into a conventional image-based rendering system. However, multi-frame rate systems use purely virtual image sources. As a consequence, the rendering process can be actively steered by the viewing process, which allows for advanced strategies. Moreover, by picking multiple reference frames from the cache, we can avoid display artifacts arising from occlusions. 1