Wavelet Transparency

TL;DR

This paper introduces a wavelet-based method for order-independent transparency rendering that improves image quality and performance, especially in complex scenes like smoke and refraction, using a GPU-efficient algorithm.

Contribution

It presents a novel wavelet representation of transmittance, an efficient GPU algorithm, and extends phenomenological transparency to include chromatic aberration effects.

Findings

Comparable image quality to A-buffering in complex scenes

Better performance and bandwidth than Moment transparency

Simpler implementation of transparency rendering techniques

Abstract

Order-independent transparency schemes rely on low-order approximations of transmittance as a function of depth. We introduce a new wavelet representation of this function and an algorithm for building and evaluating it efficiently on a GPU. We then extend the order-independent Phenomenological Transparency algorithm to our representation and introduce a new phenomenological approximation of chromatic aberration under refraction. This generates comparable image quality to reference A-buffering for challenging cases such as smoke coverage, more realistic refraction, and comparable or better performance and bandwidth to the state-of-the-art Moment transparency with a simpler implementation.