.png){kind=icon}
.png){kind=icon}
Return
Adaptive sparse polynomial regression for camera lens simulation
.png){kind=icon}
PRE
AI
DOI：10.1007/s00371-017-1402-9.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Lens effects are crucial visual elements in the synthetic imagery, but rendering lens effects with complex full lens models is time-consuming. This paper proposes a polynomial regression-based approach for constructing a sparse and accurate polynomial lens model. Terms of a polynomial are built adaptively in a bottom-up approach. Depending on the distribution of aberrations, this approach partitions the light field and builds separate polynomial models for local light fields. A line pupil-based sampling method is presented to accelerate the generation of camera rays. In addition, a new Monte Carlo estimator is derived to support general Monte Carlo rendering. Experiments show that this approach significantly reduces the time cost of constructing a polynomial lens model in comparison to state-of-the-art methods, while achieving high imaging accuracy.
Keywords:
Sparse polynomial regression
Camera lens simulation
Lens effects
Photo-realistic rendering
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.