We propose a multi-view weakly-supervised 3D human pose estimation network. In this network, we fuse multi-view features and generate the final 3D pose, supervising the 3D pose using human body segmentation generated by a human parsing network. Using human body segmentation provides powerful supervision for the network. We further propose the Pose2Seg algorithm to transform 3D pose into simple simulated segmentation for all views, which allows the network to utilize the human body segmentation to supervise the 3D pose generated by the network, without the need for complex rendering processes and estimation of human body shape. We demonstrate the effectiveness of our method on various datasets and compare our method to other state-of-the-art algorithms, showing its advantages in terms of both quality and ability to generalize.

Accurately representing 3D faces is critical in computer vision and graphics. Semantic information, i.e., the relative positional information on facial regions (e.g., "nose", "eyes"), is an essential attribute in facial geometry modeling. Existing methods often rely on geometry representations with fixed and discrete facial semantics, such as topology-uniformed mesh to model 3D faces, which limit their flexibility in tackling tasks requiring fine-grained semantic representation, such as representing facial meshes with topological structures different from the training samples. This paper introduces cm-3dmm, a novel nonlinear 3D morphable face model, which leverages implicit neural functions to map coordinates from a continuous semantic domain (represented by UV coordinates) to the 3D geometry domain. The model achieves flexible and continuous semantic representation of facial geometry, with identity and expression embeddings explicitly providing corresponding deformation controls. We also propose a data processing pipeline to construct training facial geometry datasets with continuous semantics. Experimental results demonstrate the superiority of cm-3dmm.

Source code is essential for researchers to reproduce the methods and replicate the results of artificial intelligence (AI) papers. Some organizations and researchers manually collect AI papers with available source code to contribute to the AI community. However, manual collection is a labor-intensive and time-consuming task. To address this issue, we propose a method to automatically identify papers with available source code and extract their source code repository URLs. With this method, we find that 20.5% of regular papers of 10 top AI conferences published from 2010 to 2019 are identified as papers with available source code and that 8.1% of these source code repositories are no longer accessible. We also create the XMU NLP Lab README Dataset, the largest dataset of labeled README files for source code document research. Through this dataset, we have discovered that quite a few README files have no installation instructions or usage tutorials provided. Further, a large-scale comprehensive statistical analysis is made for a general picture of the source code of AI conference papers. The proposed solution can also go beyond AI conference papers to analyze other scientific papers from both journals and conferences to shed light on more domains.

We present a novel approach for person images synthesis in this paper, that can generate person images in arbitrary poses, shapes and views. Unlike existing methods just using keypoints' locations in heatmaps format, we propose to render SMPL model to UV maps, which can provide human structural information about poses and shapes. Thus, by varying the parameters of poses, shapes and camera in SMPL model, we can generate different person images with various attributions in a simple way, while in most cases we can only obtain new shapes of people by computer graphics methods. We train an end to end generative adversarial network with unlabeled data. As our SMPL parameters come from a pretrained model, we call our overall network semi- supervised. Our network keeps a global appearance during the fine-tuning stage of the target person, thus we can get a complete appearance of the target person, rather than the inaccurate appearance caused by inferencing without enough information. Experiments on Human3.6M Dataset and a self-collected dataset demonstrate the excellent effectiveness of our approach on person images synthesis for different applications.

In this paper, an image-based walkthrough system is presented for navigating real-world outdoor scenes based on only three uncalibrated images without reconstructing 3D model. An image-based rendering operation is performed on these sample images and generates photorealistic in-between views in real time. We extend the traditional two-step view morphing to real-time tri-view morphing based on epipolar constraint. Compared with other tri-view morphing methods, our scheme copes well with both complex outdoor scenes and wide baseline scenes, which also poses a qualified alternative to state-of-the-art dense 3D reconstruction. Putting on a head mount display (HMD) like Oculus Rift, user can experience realistic and immersive exploration of real-world scenes.

In this paper, we propose a novel approach to incorporate structure knowledge into Convolutional Neural Networks (CNNs) for articulated human pose estimation from a single still image. Recent research on pose estimation adopt CNNs as base blocks to combine with other graphical models. Different from existing methods using features from CNNs to model the tree structure, we directly use the structure pose prior to guide the learning of CNN. First, we introduce a deep CNN with effective receptive fields which capture the holistic context of the whole image. Second, limb loss is used as intermediate supervision of CNN to learn the correlations of joints. Both parts and joints features are extracted in the middle of neural network and then are used to guide the following network learning. The proposed framework can exploit an implicit structure model of human body. Only using one stage and without any complex post processing, our method achieves state-of-art results on both FLIC and LSP benchmarks.

Fine-grained vehicle model recognition is a challenging problem in intelligent transportation systems due to the subtle intra-category appearance variation. In this paper, we demonstrate that this problem can be addressed by locating discriminative parts, where the most significant appearance variation appears, based on the large-scale training set. We also propose a corresponding coarse-to-fine method to achieve this, in which these discriminative regions are detected automatically based on feature maps extracted by convolutional neural network. A mapping from feature maps to the input image is established to locate the regions, and these regions are repeatedly refined until there are no more qualified ones. The global and local features are then extracted from the whole vehicle images and the detected regions, respectively. Based upon the holistic cues and the subordinate-level variation within these global and local features, an one-versus-all support vector machine classifier is applied for classification. The experimental results show that our framework outperforms most of the state-of-the-art approaches, achieving 98.29% accuracy over 281 vehicle makes and models.

Given 3D outdoor scenes acquired by a LIDAR sensor, we address the problem of semantic segmentation of 3D point clouds involving simultaneously segmenting and classifying the data. The capability of semantic segmentation is essential for several applications, such as autonomous robot navigation and 3D reconstruction of point clouds. In this paper, we present a higher-order class-specific CRF model according to the discriminative priors of object classes to capture the rich statistics of natural scenes. Consequently we cast this model in an energy minimization framework and propose the detailed energy potentials based on the class-specific priors of higher-order cliques. Then the SOSPD algorithm is adapted to optimize the energy function. To evaluate the performance of our method, we provide both quantitative and qualitative results on a challenging dataset. The results show an average Fl-score of 0.82 compared to the state-of-the-art Fl-score of 0.73.

This study discusses new frequency reversal orthogonal frequency‐division multiplexing schemes for cooperative networks in order to minimise the effect of multiple carrier frequency offsets. Theorem 1 proposed and proved in this study tells that zero‐forcing (ZF) decoding is equivalent to maximum‐likelihood (ML) decoding if the equivalent channel matrix is complex orthogonal. Frequency reversal Alamouti code for two‐relay scenario and new space‐frequency codes for more relay nodes scenario are designed under the guidance of Theorem 1. The reversal operations of these codes make the equivalent channel matrix between the relays and the destination complex orthogonalised. Thus, ZF decoding achieves the advantages of ML decoding, and simulation results also confirm that. Benefiting from the simplified ZF decoding method in Theorem 1, the decoding complexity becomes really low.

This paper aims at improving the distance-based location management scheme for mobile communication networks. In location management, a mobile terminal (MT) is tracked based on its location-update area (LA). The improvement is brought about by joint optimization of LA center and LA size. For LA center optimization (LCO), we determine the optimal center position of the LA given the initial position of the MT upon each location update. The investigation of optimal LA center has eluded research to date. Based on the popular continuous-time random walk (CTRW) mobility model, we propose an analytical framework that uses a diffusion equation to determine the optimal LA center that minimizes the total cost of location management, consisting of the location update cost and terminal paging cost. This framework allows us to easily model the non-Markovian movement of the MT and evaluate the impact of various measurable physical parameters (such as length of road section, angle between road sections, and road section crossing time) and LA center. In particular, we show that proper LA center can significantly reduce the total cost. For example, for the circular LA and low Poisson call-arrival rate, optimizing the LA center alone has the potential of reducing the cost by up to 37 percent. Joint optimization of the LA center and terminal paging scheme can reduce the cost even further. Simulations results match the theoretical analysis to a gap within 3 percent, indicating that our theoretical model is very accurate.

P2P live streaming systems are developed in two major approaches: tree-push versus mesh-pull. The hybrid push-pull streaming, as an emerging and promising approach, offers a good tradeoff between traffic overhead and system throughput. In this paper, we demonstrate that video redundancy is a large c