Rujie Liu

Attributed Graph Matching Based Engineering Drawings Retrieval

This paper presents a method for engineering drawings retrieval by their shape appearances. In this method, an engineering drawing is represented by an attributed graph, where each node corresponds to a meaningful primitive extracted from the original drawing image. This representation, which characterizes the primitives as well as their spatial relationships by graph nodes attributes and edges attributes respectively, provides a global vision of the drawings. Thus, the retrieval problem can be formulated as one of attributed graph matching, which is realized by mean field theory in this paper. The effectiveness of this method is verified by experiments.

Shape detection from line drawings with local neighborhood structure

An object detection method from line drawings is presented. The method adopts the local neighborhood structure as the elementary descriptor, which is formed by grouping several nearest neighbor lines/curves around one reference. With this representation, both the appearance and the geometric structure of the line drawing are well described. The detection algorithm is a hypothesis-test scheme. The top k most similar local structures in the drawing are firstly obtained for each local structure of the model, and the transformation parameters are estimated for each of the k candidates, such as object center, scale and rotation factors. By treating each estimation result as a point in the parameter space, a dense region around the ground truth is then formed provided that there exist a model in the drawing. The mean shift method is used to detect the dense regions, and the significant modes are accepted as the occurrence of object instances.

An images-based 3d model retrieval approach

This paper presents an images based 3D model retrieval method in which each model is described by six 2D images. The images are generated by three steps: 1) the model is normalized based on the distribution of the surface normal directions; 2) then, the normalized model is uniformly sampled to generate a number of random points; 3) finally, the random points are projected along six directions to create six images, each of which is described by Zernike moment feature. In the comparison of two models, six images of each model are naturally divided into three pairs, and the similarity between two models is calculated by summing up the distances of all corresponding pairs. The effectiveness of our method is verified by comparative experiments. Meanwhile, high matching speed is achieved, e.g., it takes about 3e-5 seconds to compare two models using a computer with Pentium IV 3.00GHz CPU.

Similarity-based partial image retrieval system for engineering drawings

Designers of mechanical products frequently refer to engineering drawings which are stored as image data in databases to design a new mechanical product efficiently. Multiple mechanical parts are usually drawn on each engineering drawing. Therefore designers want to find engineering drawings containing parts similar to a query image in the shape of a part drawn on an engineering drawing. In this paper, we propose a novel similarity based partial image retrieval system for engineering drawings. A unique aspect of this system is that a graph representation is utilized to robustly find engineering drawings containing similar parts which are invariant to the size, position, and rotation. We verified the performance for the similarity based partial image retrieval system through experiments using industrial engineering drawings. The results show that the top five similar engineering drawings for every query image are always accurately retrieved by our proposed system. This finding suggests that this system could be useful for the reuse of stored engineering drawings.

SVM-based active feedback in image retrieval using clustering and unlabeled data

In content based image retrieval, relevance feedback has been extensively studied to bridge the gap between low level image features and high level semantic concepts. However, it is still challenged by small sample size problem, since users are usually not so patient to label a large number of training instances. In this paper, two strategies are proposed to tackle this problem: (1) a novel active selection criterion. It takes into consideration both the informative and the representative measures. With this criterion, the diversities of the selected images are increased while their informative powers are kept, thus more information gain can be obtained from the feedback images; and (2) incorporation of unlabeled images within the co-training framework. Unlabeled data partially alleviates the training data scarcity problem, thus can improve the efficiency of SVM active learning. Systematic experimental results verify the superiority of our method over some existing active learning methods.

Semi-supervised learning by locally linear embedding in kernel space

Graph based semi-supervised learning methods (SSL) implicitly assume that the intrinsic geometry of the data points can be fully specified by an Euclidean distance based local neighborhood graph, however, this assumption may not always be necessarily true. To overcome this problem, we propose to apply locally linear embedding (LLE) method to characterize the geometric structure of the data points; besides this, the embedding process is performed in the kernel induced feature space rather than the original input space. After embedding, the proposed transductive learning method predicts the labels of the unlabeled data within the regularization framework. Experimental results on image retrieval and pattern recognition verify the performance of the proposed approach.

Geo-referenced Tourist Attraction Photo Tagging by Mining Community Photo Collections

The advent of photo sharing sites like Flickr has drastically increased the volume of community photo collections on the web. Also the rising popularity of the mobile devices with GPS cameras like iPhone has made most of the photos geo-tagged. These provide new opportunities for automatically tagging the geo-referenced photos such as the tourist attraction photos. In this paper, we propose a framework for automatically tagging geo-referenced tourist attraction photos through mining the community photo collections. The photos collected from social sites are first clustered by fusing several modalities such as GPS and visual features, and then the tags of each cluster are extracted via a simple TF-IDF weighted voting scheme. Finally, for a tourist attraction photo taken with a GPS camera, it is annotated by the tags of the best matched cluster. We download a lot of photos located around some places of interest in Beijing from Flickr and manually construct a geo-referenced photo dataset. Experiment on the dataset shows an overall good performance.

Device parts retrieval from assembly drawings with SVM based active relevance feedback

Content based assembly drawing retrieval is valued highly in many application areas, and it is a common thing to seek assembly drawings from a large collection where a specified device part is contained. Different from object detection techniques, a novel solution is presented in this paper to find the occurrences of target objects. Firstly, all device parts are extracted from assembly drawing images according to their specific characteristics. In later retrieval, these parts are compared with the query image to realize the search task. Furthermore, SVM based relevance feedback is adopted to incrementally improve the retrieval performance, and two strategies are proposed: (1) a novel active selection criterion, which takes into consideration both the informative and the representative measures to obtain more information from the feedback images; (2) incorporation of unlabeled images to alleviate the small sample size problem. The performance of this method is verified by extensive experiments.

Shape similarity based on contour decomposition and correspondence

A shape similarity measure based on contour decomposition and part correspondence is introduced. Shapes are approximated by polygons with split-and-merge technique, and a set of perceptual and easily reconfigurable attributes are designed to model the polygonal curves. To compute our similarity measure, the correspondence of polygonal curves is established by DP technique. Then, the similarity between them is computed and aggregated. To verify the effectiveness of our algorithm, we apply it to shape retrieval in two different datasets, one of marine species shapes and another of mechanical shapes, and compare it with some traditional approaches

Color-Introduced Frame-to-Model Registration for 3D Reconstruction

3D reconstruction has become an active research topic with the popularity of consumer-grade RGB-D cameras, and registration for model alignment is one of the most important steps. Most typical systems adopt depth-based geometry matching, while the captured color images are totally discarded. Some recent methods further introduce photometric cue for better results, but only frame-to-frame matching is used. In this paper, a novel registration approach is proposed. According to both geometric and photometric consistency, depth and color information are involved in a unified optimization framework. With the available depth maps and color images, a global model with colored surface vertices is maintained. The incoming RGB-D frames are aligned based on frame-to-model matching for more effective camera pose estimation. Both quantitative and qualitative experimental results demonstrate that better reconstruction performance can be obtained by our proposal.