摘要
光场相机通过单次曝光同时从多个视角采样单个场景,在深度估计领域具有独特优势。噪声场景下的深度获取是光场图像深度估计的难点之一。传统针对噪声场景的深度获取方法大多仅适用于非遮挡情况,无法较好处理包含遮挡区域的噪声场景。针对含遮挡的噪声场景深度估计问题,提出了基于内联遮挡处理的深度估计方法。该方法采用内联遮挡处理框架,通过将遮挡处理集成进抗噪成本量中,在保证抗噪性能的同时提升算法的抗遮挡能力。在成本量建立完成后,为进一步滤除剩余噪声,采用提出的适应遮挡的多模板滤波策略对成本量进行遮挡感知优化,该策略通过为不同方向的遮挡分别设计滤波模板,在滤波的同时能较好保留图像的边缘结构,有效改善了传统滤波算法无法保留遮挡边界的问题。实验结果表明,相比其它先进深度估计算法,该方法在高噪场景下具有显著优势,并能更好处理噪声场景深度估计的遮挡问题。
A light field camera can simultaneously sample a scene from multiple viewpoints with a single exposure,which has unique advantages in portability and depth accuracy over other depth sensors.Noise is a challenging is-sue for light field depth estimation.Most of the traditional depth estimation methods for noisy scenes are only suita-ble for non-occluded scenes,and cannot handle the noisy scenes with occluded regions.To solve this problem,we present a light field depth estimation method based on inline occlusion handling.The proposed method integrates the occlusion handling into the anti-noise cost volume,which can improve the anti-occlusion capability while main-taining the anti-noise performance.After the cost volume is constructed,we propose a multi-template filtering algo-rithm to smooth the data cost while preserving the edge structure.Experimental results show that the proposed method has better performance over other state-of-the-art depth estimation methods in high noise scenes,and can better handle the occlusion problem of depth estimation in noisy scenes.
作者
吴迪
张旭东
范之国
孙锐
Wu Di;Zhang Xudong;Fan Zhiguo;Sun Rui(School of Computer Science and Information Engineering,Hefei University of Technology,Hefei,Anhui 230601,China)
出处
《光电工程》
CAS
CSCD
北大核心
2021年第7期9-22,共14页
Opto-Electronic Engineering
基金
国家自然科学基金资助项目(61876057,61971177)。
关键词
光场
深度估计
散焦线索
噪声抑制
遮挡处理
light field
depth estimation
defocus cue
noise suppression
occlusion handling
