Qian Yu

Bio: Qian Yu is an academic researcher from Fudan University. The author has contributed to research in topics: Computer science & Artificial intelligence. The author has an hindex of 5, co-authored 6 publications receiving 80 citations. Previous affiliations of Qian Yu include Jiangsu University.

Part-Wise AtlasNet for 3D point cloud reconstruction from a single image

Abstract: Learning to generate three dimensional (3D) point clouds from a single image remains a challenging task. Numerous approaches with encoder–decoder architectures have been proposed. However, these methods are hard to realize structured reconstructions and usually lack constraints on the local structures of 3D objects. AtlasNet as a representative model of 3D reconstruction consists of many branches, and each branch is a neural network used to reconstruct one local patch of a 3D object. However, the neural networks in AtlasNet and the patches of 3D objects are not in one-to-one correspondence before training. This case is not conducive to adding some reconstruction constraints to the local structures of 3D objects. Based on the architecture of AtlasNet, we propose Part-Wise AtlasNet in which each neural network is only responsible for reconstructing one specific part of a 3D object. This kind of restriction facilitates imposition of several local constraints on the final reconstruction loss, hence better recovering 3D objects with finer local structures. Both the qualitative results and quantitative analysis show that the variants of the proposed method with the local reconstruction losses generate structured point clouds with a higher visual quality and achieve better performance than other methods in 3D point cloud generation from a single image.

Fruit detection in natural environment using partial shape matching and probabilistic Hough transform

Abstract: This paper proposes a novel technique for fruit detection in natural environments which is applicable in automatic harvesting robots, yield estimation systems and quality monitoring systems. As most color-based techniques are highly sensitive to illumination changes and low contrasts between fruits and leaves, the proposed technique, conversely, is based on contour information. Firstly, a discriminative shape descriptor is derived to represent geometrical properties of arbitrary fragment, and applied to a bidirectional partial shape matching to detect sub-fragments of interest that match parts of a reference contour. Then, a novel probabilistic Hough transform is developed to aggregate these sub-fragments for obtaining fruit candidates. Finally, all fruit candidates are verified by a support vector machine classifier trained on color and texture features. Citrus, tomato, pumpkin, bitter gourd, towel gourd and mango datasets were provided. Experiments on these datasets demonstrated that the proposed approach was competitive for detecting most type of fruits, such as green, orange, circular and non-circular, in natural environments.

Scene Semantic Recognition Based on Modified Fuzzy C-Mean and Maximum Entropy Using Object-to-Object Relations

Abstract: With advances in machine vision systems (e.g., artificial eye, unmanned aerial vehicles, surveillance monitoring) scene semantic recognition (SSR) technology has attracted much attention due to its related applications such as autonomous driving, tourist navigation, intelligent traffic and remote aerial sensing. Although tremendous progress has been made in visual interpretation, several challenges remain (i.e., dynamic backgrounds, occlusion, lack of labeled data, changes in illumination, direction, and size). Therefore, we have proposed a novel SSR framework that intelligently segments the locations of objects, generates a novel Bag of Features, and recognizes scenes via Maximum Entropy. First, denoising and smoothing are applied on scene data. Second, modified Fuzzy C-Means integrates with super-pixels and Random Forest for the segmentation of objects. Third, these segmented objects are used to extract a novel Bag of Features that concatenate different blobs, multiple orientations, Fourier transform and geometrical points over the objects. An Artificial Neural Network recognizes the multiple objects using the different patterns of objects. Finally, labels are estimated via Maximum Entropy model. During experimental evaluation, our proposed system illustrated a remarkable mean accuracy rate of 90.07% over the MSRC dataset and 89.26% over the Caltech 101 for object recognition, and 93.53% over the Pascal-VOC12 dataset for scene recognition, respectively. The proposed system should be applicable to various emerging technologies, such as augmented reality, to represent the real-world environment for military training and engineering design, as well as for entertainment, artificial eyes for visually impaired people and traffic monitoring to avoid congestion or road accidents.

2D Object Recognition Techniques: State-of-the-Art Work

Abstract: Object recognition is one of the research area in the field of computer vision and image processing because of its varied applications in surveillance and security systems, biometrics, intelligent vehicle system, content based image retrieval, etc. Many researchers have already done a lot of work in this area, but still there are many issues like scale, rotation, illumination, viewpoint, occlusion, background clutter among many more that draw the attention of the researchers. Object recognition is the task of recognizing the object and labeling the object in an image. The main goal of this survey is to present a comprehensive study in the field of 2D object recognition. An object is recognized by extracting the features of object like color of the object, texture of the object or shape or some other features. Then based on these features, objects are classified into various classes and each class is assigned a name. In this paper, various feature extraction techniques and classification algorithms are discussed which are required for object recognition. As the deep learning has made a tremendous improvement in object recognition process, so the paper also presents the recognition results achieved with various deep learning methods. The survey also includes the applications of object recognition system and various challenges faced while recognizing the object. Pros and cons of feature extraction and classification algorithms are also discussed which may help other researchers during their initial period of study. In this survey, the authors have also reported an analysis of various researches that describes the techniques used for object recognition with the accuracy achieved on particular image dataset. Finally, this paper ends with concluding notes and future directions. The aim of this study is to introduce the researchers about various techniques used for object recognition system.

Multiscale Fourier descriptor based on triangular features for shape retrieval

Abstract: Shape information plays an important role in the human visual system, and how to represent the shape accurately is a very challenging issue in shape recognition. In recent years, many shape description approaches have been proposed. However, most of the existing methods are still facing challenges on accuracy and computational efficiency. Fourier descriptor is regarded as a promising shape description method because it has a solid theoretical foundation and at the same time possesses the strength of appealing invariance properties and high computational efficiency. We introduce a novel multiscale Fourier descriptor based on triangular features which is used to identify shapes. The local and global characteristics of a shape are effectively captured by the proposed shape descriptor. Meanwhile, the proposed shape descriptor has good properties, such as the invariant of the geometric transformation and the starting point of an object. We tested this descriptor on four popular benchmarking datasets, including MPEG-7, Swedish leaf, ETH-80, and Flavia leaf. The results confirm that our method is better than comparable-complexity approaches based on Fourier descriptors and does not perform unfavorably with respect to more complex, state-of-the-art, shape descriptors. In particular, our method is far superior to the complex shape description methods in terms of retrieval efficiency and computational complexity.