Object-based Illumination Estimation with Rendering-aware Neural Networks
This work addresses the need for realistic virtual object rendering in AR scenarios by providing fast and accurate illumination estimation, though it is incremental as it builds on existing inverse rendering and learning-based methods.
The paper tackles the problem of real-time environment light estimation from RGBD object appearances for augmented reality, achieving high accuracy and real-time performance by combining physical principles from inverse rendering with neural networks.
We present a scheme for fast environment light estimation from the RGBD appearance of individual objects and their local image areas. Conventional inverse rendering is too computationally demanding for real-time applications, and the performance of purely learning-based techniques may be limited by the meager input data available from individual objects. To address these issues, we propose an approach that takes advantage of physical principles from inverse rendering to constrain the solution, while also utilizing neural networks to expedite the more computationally expensive portions of its processing, to increase robustness to noisy input data as well as to improve temporal and spatial stability. This results in a rendering-aware system that estimates the local illumination distribution at an object with high accuracy and in real time. With the estimated lighting, virtual objects can be rendered in AR scenarios with shading that is consistent to the real scene, leading to improved realism.