View-wised discriminative ranking for 3D object retrieval

In this paper, we propose a new framework which can capture the latent relative information within the multiple views of 3D model, named View-wised Discriminative Ranking(VDR). Different to existing view-based methods which treat the multiple views as the independent information, we want to model the relative information within multiple views. By placing the views of model in certain order, we learn the parameters of ranking function as a new robust model representation. We evaluate our proposal on several challenging datasets for 3D retrieval and the comparison experiments demonstrate the superiority of the proposed method in both retrieval accuracy and efficiency.     

Off-the-shelf CNN features for 3D object retrieval

Multimedia Tools and Applications - Effective feature representation is crucial to view-based 3D object retrieval (V3OR). Most previous works employed hand-crafted features to represent the views...     

Review of multi-view 3D object recognition methods based on deep learning

Three-dimensional (3D) object recognition is widely used in automated driving, medical image analysis, virtual/augmented reality, artificial intelligence robots, and other areas. Deep learning is increasingly being used to solve 3D vision problems. Multi-view 3D object recognition based on the deep learning technique has become one of the rigorously researched topics because it can directly use the pretrained and successful advanced classification network as the backbone network, and views from multiple viewpoints can complement each other’s detailed features of the object. However, some challenges still exist in this area. Recently, many methods have been proposed to solve the problems pertaining to this research topic. This paper presents a comprehensive review and classification of the latest developments in the deep learning methods for multi-view 3D object recognition. It also summarizes the results of these methods on a few mainstream datasets, provides an insightful summary, and puts forward enlightening future research directions.     

Local commute-time guided MDS for 3D non-rigid object retrieval

In this paper, we propose a 3D non-rigid shape retrieval method based on canonical shape analysis. Our main idea is to transform the problem of non-rigid shape retrieval into a rigid shape retrieval problem via the well-known multidimensional scaling (MDS) approach and random walk on graphs. We first segment the non-rigid shape into local partitions based on its salient features. Then, we calculate a local MDS problem for each partition, where the local commute time distance is used as weighting function in order to preserve local shape details. Finally, we aggregate the set of local MDS problems as a global constrained problem. The constraint is formulated using the biharmonic function between local salient features. In contrast to MDS method, the proposed local MDS is computationally efficient, parameters free and gives isometry-invariant forms with minimum features distortion. Due to these advantageous properties, the proposed method achieved good retrieval accuracy on non-rigid shape benchmark datasets.     

Single image 3D object reconstruction based on deep learning: A review

Multimedia Tools and Applications - The reconstruction of 3D object from a single image is an important task in the field of computer vision. In recent years, 3D reconstruction of single image...     

Reeb graph path dissimilarity for 3D object matching and retrieval

We introduce a skeletal graph for topological 3D shape representation using Morse theory. The proposed skeletonization algorithm encodes a 3D shape into a topological Reeb graph using a normalized mixture distance function. We also propose a novel graph matching algorithm by comparing the relative shortest paths between the skeleton endpoints. Experimental results demonstrate the feasibility of the proposed topological Reeb graph as a shape signature for 3D object matching and retrieval.     

基于混合光场描述符的三维目标检索研究

针对三维目标（3D object）检索问题,提出了一种基于新型描述符的3D目标检索方法。首先,在分析现行基于视图的3D模型描述符在描述方法上不充分的基础上,提出了混合描述符H_D的总体思路。进而讨论了H_D总体框架,即在光场图像阵列自适应的基础上,实现了直方图颜色描述符H_CD,shock图形状描述符H_SD及贝叶斯网络（Bayesian Network,BN）纹理描述符H_TD的优化组合。其次,讨论了H_D各部分的具体实现及度量机制,最后,对H_D检索性能进行了实验分析,结果表明提出的方法是优于其他基于视图的检索方法。     

Two-stage deep learning for supervised cross-modal retrieval

This paper deals with the problem of modeling internet images and associated texts for cross-modal retrieval such as text-to-image retrieval and image-to-text retrieval. Recently, supervised cross-modal retrieval has attracted increasing attention. Inspired by a typical two-stage method, i.e., semantic correlation matching(SCM), we propose a novel two-stage deep learning method for supervised cross-modal retrieval. Limited by the fact that traditional canonical correlation analysis (CCA) is a 2-view method, the supervised semantic information is only considered in the second stage of SCM. To maximize the value of semantics, we expand CCA from 2-view to 3-view and conduct supervised learning in both stages. In the first learning stage, we embed 3-view CCA into a deep architecture to learn non-linear correlation between image, text and semantics. To overcome over-fitting, we add the reconstruct loss of each view into the loss function, which includes the correlation loss of every two views and regularization of parameters. In the second stage, we build a novel fully-convolutional network (FCN), which is trained by joint supervision of contrastive loss and center loss to learn better features. The proposed method is evaluated on two publicly available data sets, and the experimental results show that our method is competitive with state-of-the-art methods.

     

Instance-level object retrieval via deep region CNN

Multimedia Tools and Applications - Instance retrieval is a fundamental problem in the multimedia field for its various applications. Since the relevancy is defined at the instance level, it is...     

3D object retrieval based on Spatial+LDA model

Multimedia Tools and Applications - Latent Dirichlet Allocation (LDA) is one popular topic extraction method, which has been applied in many applications such as textual retrieval, user...