Learning Training Samples for Occlusion Edge Detection and Its Application in Depth Ordering Inference
Learning Training Samples for Occlusion Edge Detection and Its Application in Depth Ordering Inference
复制标题
DOI：
10.1109/icpr.2018.8545622
复制
发表时间：
2018-08
期刊：
影响因子：
--
通讯作者：
Yu Zhou;Jianxiang Ma;Anlong Ming;X. Bai
中科院分区：
文献类型：
--
作者：
Yu Zhou;Jianxiang Ma;Anlong Ming;X. Bai
本文研究了遮挡边缘检测问题，并将其应用于单目图像中物体深度顺序的推断。关键的观察是，给定固定的回归目标，遮挡边缘检测的准确性，有效地提高通过选择适当的训练样本中的判别特征子空间。具体地说，$\ell_{1}$ -正则化逻辑回归用于学习更稀疏但具有鉴别力的特征子空间，而训练样本的选择被公式化为具有鲁棒Huber损失的二次优化。该方法有效地避免了噪声干扰，从而能够检测出理想的遮挡边缘。我们验证了我们的方法在深度顺序推理问题上的有效性。实验在两个著名的数据集上进行，即，Cornell深度顺序数据集和NYU 2数据集。令人鼓舞的结果表明，我们的方法优于国家的最先进的方法。
This paper studies the problem of occlusion edge detection, which is applied to infer the depth order of objects in a monocular image. The key observation is that, given the fixed regression objective, the accuracy of occlusion edge detection is effectively boosted by selecting appropriate training samples in a discriminative feature subspace. Specifically, the $\ell_{1}$ -regularized logistic regression is employed to learn a more sparse yet discriminative feature subspace, while the training sample selection is formulated as a quadratic optimization with the robust Huber loss. The presented formulation avoids the noises efficiently, and hence the desirable occlusion edges can be detected. We validate the effectiveness of our approach on depth order inference problem. Experiments are conducted on two famous datasets, i.e., the Cornell depth-order dataset and the NYU2 dataset. Promising results demonstrate the superiority of our approach over the state-of-the-art approaches.