This paper proposes a cross-modal distillation frame-work, PartDistill, which transfers 2D knowledge from vision-language models (VLMs) to facilitate 3D shape part segmentation. PartDistill addresses three major challenges in this task: the lack of 3D segmentation in invisible or undetected regions in the 2D projections, inconsistent 2D predictions by VLMs, and the lack of knowledge accumu-lation across different 3D shapes. PartDistill consists of a teacher network that uses a VLM to make 2D predictions and a student network that learns from the 2D pre-dictions while extracting geometrical features from multi-ple 3D shapes to carry out 3D part segmentation. A bi-directional distillation, including forward and backward distillations, is carried out within the framework, where the former forward distills the 2D predictions to the student net-work, and the latter improves the quality of the 2D predictions, which subsequently enhances the final 3D segmen-tation. Moreover, PartDistill can exploit generative mod-els that facilitate effortless 3D shape creation for generating knowledge sources to be distilled. Through extensive experiments, PartDistill boosts the existing methods with substantial margins on widely used ShapeNetPart and Part-NetE datasets, by more than 15% and 12% higher mIoU scores, respectively. The code for this work is available at https://github.com/ardianumam/PartDistill.

Previous methods have demonstrated remarkable performance in single image super-resolution (SISR) tasks with known and fixed degradation (e.g., bicubic downsampling). However, when the actual degradation deviates from these assumptions, these methods may experience significant declines in performance. In this paper, we propose a Dual Branch Degradation Extractor Network to address the blind SR problem. While some blind SR methods assume noisefree degradation and others do not explicitly consider the presence of noise in the degradation model, our approach predicts two unsupervised degradation embeddings that represent blurry and noisy information. The SR network can then be adapted to blur embedding and noise embedding in distinct ways. Furthermore, we treat the degradation extractor as a regularizer to capitalize on differences between SR and HR images. Extensive experiments on several benchmarks demonstrate our method achieves SOTA performance in the blind SR problem.

We propose a co-part segmentation method that takes a set of point clouds of the same category as input where neither a ground truth label nor a prior network is required. With difficulties caused by the label absence, we formulate the co-part segmentation task into two subtasks, including superpoint generation and part aggregation. In the first subtask, our superpoint generation network divides each point cloud into homogeneous partitions, each called superpoint, while in the second subtask, these superpoints are further aggregated into a few semantic parts via our part aggregation network. We introduce the coupled attention blocks in the part aggregation network to explicitly enforce semantic consistency in the segmentation by exploiting intra-, inter-, and paired-cloud geometrical information by minimizing the devised intra-, inter-, and paired-cloud losses, respectively. The intra-cloud loss triggers a semantic segmentation in each point cloud, while the inter-cloud loss considers all clouds to enforce their semantic consistency. The paired-cloud loss is designed to ensure that each part of one point cloud can be discriminatively reconstructed from the superpoints of another point cloud. We perform experiments on two benchmark datasets, ShapeNet part and COSEG, and provide quantitative and qualitative results to demonstrate the superiority of our method over existing methods. We also show that the proposed method can help several downstream tasks, including semi-supervised part segmentation and data augmentation for shape classification. The code for this work will be publicly available upon the paper's publication.

A new research topic, multi-source domain adaptive object detection (MSDAOD) under privacy-preserving constraint is explored in this paper, where the clients can only access their own source data while the server can only access unlabeled target data. Accordingly, a novel MSDAOD framework is proposed wherein the clients employ a source-only Probabilistic Faster R-CNN (PFRCNN) to generate models with localization uncertainty, while a Multi-teacher Pseudo-label Ensemble Network (MPEN) is developed on the server side. In MPEN, FedMA-based algorithm aggregates the above models to a domain-invariant backbone while a novel pseudo-label ensemble (PLE) scheme is employed to reduce false positives arising from domain specific parts, and enhance the overall system performance using target domain information. Experiments demonstrate that our method outperforms other state-of-the-art MSDAOD and privacy-preserving methods by 10%~16% in average precision (AP).

An efficient domain adaptation scheme is presented in this paper for nighttime pedestrian detection using Faster R-CNN. First, we adopt Fourier domain adaptation on training data by replacing low-frequency spectrum of source data (RGB images) with that of target data (infrared images). Such approach is more efficient compared with existing state-of-the-art methods of domain adaptation for object detection, as it does not require adversarial learning, or adding extra components to the Faster R-CNN. In addition, a simple preprocessing of intensity scaling is empirically selected among several image enhancement algorithms for testing data. Experimental results demonstrate that performance improvements of up to 30% and 10% can be achieved with the above processes for training data and testing data, respectively, for an indoor scene with poor illumination condition (while other processes may actually lower the performance).

In advanced multi-view video surveillance systems, people localization is usually a crucial part of the complete system and need to be accomplished in a short time to reserve sufficient processing time for subsequent high-level analysis. As the surveillance area increases, it is required to install a large number of cameras for multi-view people localization. To lower the equipment cost and setup time, we incorporate fisheye (or wide-angle) camera to an efficient vanishing point-based line sampling scheme for people localization, by ensuring the fisheye camera is looking downward so that its principal point becomes the vanishing point of vertical lines. Experimental results show that the utilization of fisheye-camera can (i) achieve localization accuracy comparable or superior to that using ordinary cameras, (ii) reduce the camera count by 75% on the average while covering the same or larger size of a monitored area, and (iii) greatly simplify the camera installation process.

Camera calibration is a crucial prerequisite in many applications of computer vision. In this paper, a new geometry-based camera calibration technique is proposed, which resolves two main issues associated with the widely used Zhang's method: (i) the lack of guidelines to avoid outliers in the computation and (ii) the assumption of fixed camera focal length. The proposed approach is based on the closed-form solution of principal lines with their intersection being the principal point while each principal line can concisely represent relative orientation/position (up to one degree of freedom for both) between a special pair of coordinate systems of image plane and calibration pattern. With such analytically tractable image features, computations associated with the calibration are greatly simplified, while the guidelines in (i) can be established intuitively. Experimental results for synthetic and real data show that the proposed approach does compare favorably with Zhang's method, in terms of correctness, robustness, and flexibility, and addresses issues (i) and (ii) satisfactorily.

Obstacle avoidance and environment sensing are crucial applications in autonomous driving and robotics. Among all types of sensors, RGB camera is widely used in these applications as it can offer rich visual contents with relatively low-cost, and using a single image to perform depth estimation has become one of the main focuses in resent research works. However, prior works usually rely on highly complicated computation and power-consuming GPU to achieve such task; therefore, we focus on developing a real-time light-weight system for depth prediction in this paper. Based on the well-known encoder-decoder architecture, we propose a supervised learning-based CNN with detachable decoders that produce depth predictions with different scales. We also formulate a novel log-depth loss function that computes the difference of predicted depth map and ground truth depth map in log space, so as to increase the prediction accuracy for nearby locations. To train our model efficiently, we generate depth map and semantic segmentation with complex teacher models. Via a series of ablation studies and experiments, it is validated that our model can efficiently performs real-time depth prediction with only 0.32M parameters, with the best trained model outperforms previous works on KITTI dataset for various evaluation matrices.

This paper proposed a modified YOLOv3 which has an extra object depth prediction module for obstacle detection and avoidance. We use a pre-processed KITTI dataset to train the proposed, unified model for (i) object detection and (ii) depth prediction and use the AirSim flight simulator to generate synthetic aerial images to verify that our model can be applied in different data domains. Experimental results show that the proposed model compares favorably with other depth map prediction methods in terms of accuracy in the prediction of object depth for pre-processed KITTI dataset, while the unified approach can actually improve both (i) and (ii) at the same time.

While fast imaging in low-light condition is crucial for surveillance and robot applications, it is still a formidable challenge to resolve the seemingly inevitable high noise level and low photon count issues. A variety of image enhancement methods such as de-blurring and de-noising have been proposed in the past. However, limitations can still be found in these methods under extreme low-light condition. To overcome such difficulty, a learning-based image enhancement approach is proposed in this paper. In order to support the development of learning-based methodology, we collected a new low-lighting dataset (<;0.1 lux) of raw short-exposure (6.67 ms) images, as well as the corresponding long-exposure reference images. Based on such dataset, we develop a light-weight convolutional network structure which is involved with fewer parameters and has lower computation cost compared with a regular-size network. The presented work is expected to make possible the implementation of more advanced edge devices, and their applications.

Drones are well-liked nowadays. However, deep learning models for object detection still cannot have high detection rates for pedestrians in aerial images even though they already show high precision on PASCAL VOC 2007. The main challenges of aerial image analysis include: (i) the size of an object in aerial images can be very small, and (ii) the objects in aerial images are tilted outward due to perspective projection deformation, which make the pedestrians hard to recognize in aerial images. In this paper, we utilize image partition and vanishing point transformation to overcome the above challenges. Experimental results demonstrate that such pre-processing methods can indeed increase the detection rates significantly for some deep learning models.

In this paper, a novel, fully automatic camera calibration procedure is proposed to estimate the camera principal point, i.e., the intersection of the optical axis and the image plane. The basic idea is to derive the orthogonal projection of the camera optical axis onto the flat-screen monitor via changing/analyzing simple edge patterns displayed on the flat monitor. The principal point can then be estimated as the intersection of images of line features thus derived on the screen, called calibration lines, from multiple monitor configurations. Thus, the calibration can be performed automatically using fixed camera and multiple monitors, or using multiple poses of a movable monitor. Several experiments are developed to evaluate the proposed method for accuracy as well as robustness, and the results show that the proposed approach compares favorably with some previous works, including the classic Zhang's algorithm that derives the principal point together with other camera parameters at the same time.

Fisheye camera is widely used in various applications because of its wider field-of-view. However, high distortion of images captured by fisheye cameras make it difficult for certain tasks which are based on traditional epipolar geometry (using epipolar lines) and stereo correspondence, such as depth map estimation. While most of existing depth map estimation methods use perspective-projection-based camera model, considering fisheye camera for depth map estimation will be beneficial because of its wider FOV. The availability of epipolar curves will be truly helpful for correspondence matching performed in depth map estimation. We propose a novel way of deriving epipolar curves using an auxiliary screen having benefits include: (i) deterministic results instead of statistical ones with accuracy within 1.5 pixels and (ii) the two epipoles as well as the epipolar curves in fisheye images can be obtained automatically, while unlimited number of the latter can be generated if needed.

An automatic calibration procedure for a fisheye camera is presented in this paper by employing a flat panel monitor. The procedure does not require precise camera-monitor alignment, and any manual input of data or commands, making it useful for factory automation for mass production of such cameras. The fully automatic calibration procedure, which requires the generation of various test patterns on the display, and analysis of fisheye images of these patterns, consists of the following steps: (i) estimate the image center of the camera, (ii) identify the line on the monitor which intersects optical axis of the camera perpendicularly, and (iii) along the above line, obtain calibration data needed in de-warping the fisheye image. Experimental results demonstrate that the proposed approach performs satisfactorily in terms of effectiveness and accuracy.

A full stage bank serial number (SN) recognition system is proposed in this paper. We introduce Block-wise Prediction Networks (BPN) to treat the localization of an SN as block-wise binary classification, which can be considered as a coarse version of dense/pixel-wise prediction used in semantic segmentation. The benefits include short execution time, which is equal to 85.22 ms in CPU, and the use of global features instead of local features to improve the segmentation. Our system then separates the localized Region of Interest (RoI) into individual characters, and feeds them into softmax CNN classifier. Experimental results show that the proposed method can achieve 99.92% and 99.24% accuracy for character and SN of Renminbi (RMB), respectively, tested with 2,368 two sides images of 1,184 RMB bills.

In this paper, we propose a novel, egocentric pointing system based on Google Glass which is equipped with an optical head-mounted display (OHMD) and a near-eye camera, with the eye-pointing line passing through the lower left corner of the display. For a pointed target, the pointing (or ranging) algorithm is based on a distance-pixel curve established from the camera-eye (epipolar) geometry. Additional pointing algorithms for estimating gazing point on a planar surface are also developed by establishing another distance-pixel curve along the same epipolar line. Experiments show that less than 0.32° angular error in the egocentric pointing can be achieved for object distance ranging from 80cm to 178cm by the best estimation scheme, with slightly less accurate results (i.e. 0.58°) achievable by simpler estimation schemes.

In advanced video surveillance systems, people localization is usually a part of the complete system and should be accomplished in a short time so as to reserve sufficient processing time for subsequent high-level analysis, such as abnormal event/behavior detection and intruder detection. Hence, in addition to localization accuracy, computational efficiency is of critical importance as well. In this paper, we adopt a vanishing point-based line sampling scheme and propose a fast multicamera people localization approach capable of locating a crowd of dense people and estimating their heights in a fairly short time with high accuracy. For each camera view, sample lines, originated from a vanishing point, of foreground objects are projected onto the ground plane. Then, people locations are estimated by detecting the ground regions containing a high density of the projected lines. Enhanced from some previous works, the proposed approach does not require processing steps of high computation cost, such as projecting all foreground pixels of all views to multiple reference planes or computing pairwise intersections of projected sample lines at different heights. In addition, some novel acceleration modules, such as torso validation and physical rule-based filtering, are developed to further reduce the computation time of people localization. The experiments on real surveillance scenes validate that the proposed approach achieves significant speedup (up to 186%) over state-of-the-art techniques while still ensure a comparably high localization accuracy, even for crowded scenes with serious occlusions.

In a video surveillance system with static cameras, object segmentation often fails when part of the object has similar color with the background, resulting in poor performance of the subsequent object tracking. Multiple kernels have been utilized in object tracking to deal with occlusion, but the performance still highly depends on segmentation. This paper presents an innovative system, named Multiple-kernel Adaptive Segmentation and Tracking (MAST), which dynamically controls the decision thresholds of background subtraction and shadow removal around the adaptive kernel regions based on the preliminary tracking results. Then the objects are tracked for the second time according to the adaptively segmented foreground. Evaluations of both segmentation and tracking on benchmark datasets and our own recorded video sequences demonstrate that the proposed method can successfully track objects in similar-color background and/or shadow areas with favorable segmentation performance.

In a video surveillance system with a single static camera, tracking results of moving persons can be effectively used for camera self-calibration. However, the current methods need to depend on robustness of both tracking and segmentation procedures. RANSAC has been widely used to remove outliers in finding the vertical vanishing point and the horizon line, but the performance is degraded when the proportion of outliers is high. Last but not least, all of them require excessive simplifications in the algorithmic procedures resulting in increasing reprojection error. In this paper, a robust segmentation and tracking system is applied to provide accurate estimation of head and foot locations of moving persons. The noise in the computation of vanishing points is handled by mean shift clustering and Laplace linear regression through convex optimization. We also propose to use the estimation of distribution algorithm (EDA) to search for the local optimal solution for camera calibration that minimizes average reprojection error on the ground plane, while relaxing the assumptions on camera parameters. Promising evaluations of the performance of our proposed method on real scenes are presented.

MV-HEVC and 3D-HEVC are the extensions of High Efficiency Video Coding (HEVC) originally designed for Multi-view and 3D video coding respectively. In this paper we reveal a new application of using these video coding standards, i.e., to compress the Multi-Intensity Illuminated Infrared (MIIR) video, a new type of infrared video developed to overcome the limitation of improper illumination conditions for nighttime surveillance. With the proposed encoding method for the MIIR video, the coding structures of MV- and 3D-HEVC can compress such video in a more efficient way, i.e., up to 13.2%/17.6% bitrate reductions with MV/3D-HEVC compared with standalone HEVC compression and 64.4%/68.7% bitrate reductions compared with simulcast HEVC coding. Experiment results also reveal that weighted prediction algorithm in HEVC can be improved to deal with large change of brightness in the MIIR video.

An efficient calibration method for multi-plane homography is proposed in this paper. Two laser levels are used to cast laser lines to construct virtual poles in the environment without deploying real objects. HSV color model, Hough transform, and least squares method are applied to locate the laser lines in the captured images. Using the features of vanishing line and the view-invariant cross-ratio model, the 3-D coordinate of the camera can be estimated. The multi-plane homography between camera image and the world space can be derived based on two layers homography. The first layer homography relates the ground and the image, whereas the second layer homography can be efficiently obtained using the first layer homography and the virtual poles. Experimental results show that the virtual poles and the derived homography are both accurately estimated.

Vision-based video surveillance systems have gained increasing popularity. However, their functionality is substantially limited under nighttime conditions due to the poor visibility caused by improper illumination. Equipped on night vision cameras, ordinary infrared (IR) illuminators of fixed-intensity usually lead to the imaging problem of overexposure (or underexposure) when the object is too close to (or too far from) the camera. To overcome this limitation, we use a novel multi-intensity IR illuminator to extend the effective range of distance of camera surveillance, and establish in this paper the MI3 (Multi-Intensity Infrared Illumination) database based on such an illuminator. The database contains intensity varying video sequences of several indoor and outdoor scenes. Ground truths including people counting and foreground labelling are provided for different research usages. Performances of related algorithms are tested for demonstration and evaluation.

The widespread use of vision-based video surveillance systems has inspired many research efforts on people localization. One of the current main trends in this field is based on probabilistic occupancy map (POM) obtained from multiple camera views. Although the POM-based approaches are robust against noisy foregrounds and can achieve great localization accuracy, they require high computation complexity. In this paper, two enhancement schemes are proposed to improve the efficiency of the POM-based people localization: (i) quick screening of potential people locations, and (ii) timely termination of iterations for occupancy probability estimation. Experimental results show that the proposed approach achieves up to 7.25 times speed-up compared to the standard POM-based approach, while delivering comparable people localization accuracy.

Recently, an unsupervised image segmentation framework, Segmentation by Aggregating Superpixels (SAS) is proposed and shown to be very promising. However, the texture cues, which have been shown to be very effective in many researches, are not used. In this paper, we propose an effective method for incorporating texture information into the SAS framework, using superpixels. To extract texture information, our algorithm first uses texture filtering and subsequently GMM clustering. Then, we develop an edge-aware low-pass filtering to generate multiple-scale textural superpixels (TXSPs) from the clustering results. Finally, by joining TXSPs with the superpixel set originally used in SAS, the incorporation of texture information is accomplished. Our method achieves superior performance on the Berkeley Segmentation Dataset (BSDS300) under several evaluation criteria when compared to other benchmark algorithms.

The widespread use of vision-based surveillance systems has inspired many research efforts on people localization. In this paper, a series of novel image transforms based on the vanishing point of vertical lines is proposed for enhancement of the probabilistic occupancy map (POM)-based people localization scheme. Utilizing the characteristic that the extensions of vertical lines intersect at a vanishing point, the proposed transforms, based on image or ground plane coordinate system, aims at producing transformed images wherein each standing/walking person will have an upright appearance. Thus, the degradation in localization accuracy due to the deviation of camera configuration constraint specified can be alleviated, while the computation efficiency resulted from the applicability of integral image can be retained. Experimental results show that significant improvement in POM-based people localization for more general camera configurations can indeed be achieved with the proposed image transforms.

In nighttime video surveillance, proper illumination plays a key role for the image quality. For ordinary IR-illuminators with fixed intensity, faraway objects are often hard to identify due to insufficient illumination while nearby objects may suffer from over-exposure, resulting in image foreground/background of poor quality. In this paper we proposed a novel video summarization method which utilizes a novel multi-intensity IR-illuminator to generate images of human activities with different illumination levels. By adopting GMM-based foreground extraction procedure for images acquired for each illumination level, foreground objects with most plausible quality can be selected and merged with a preselected representation for still background. The result brings out a reasonable video summary for moving foreground, which is generally unachievable for nighttime surveillance videos.

The widespread use of vision-based surveillance systems has inspires many research efforts on people localization. In this paper, a novel nonlinear image transform based on the vanishing point of vertical lines, entitled VP-transform, is proposed for enhancement of the people localization scheme proposed in [1]. Utilizing the characteristic that the extensions of vertical lines intersect at a vanishing point, the proposed transformation aims at producing rectified images wherein each standing/walking person will have an upright appearance. Thus, the degradation in localization accuracy, due to deviation of the constraint of camera configuration specified in [1], can be alleviated. Experimental results validate the effectiveness of the proposed image transform that significant improvement in people localization can be achieved.

In this paper, we propose a real-time multicamera people localization method based on line sampling of image foregrounds. For each view, these line samples are originated from the vanishing point of lines perpendicular to the ground plane. With these line samples, vertical line samples in the 3-D scene can be reconstructed for potential human locations. After some efficient geometric refinement and filtering procedures, the remaining qualified 3-D line samples are clustered and integrated for the identification of locations and heights of people in the scene. Both indoor and outdoor scenarios are examined to demonstrate the effectiveness of our approach in handling serious occlusion in crowed scenes. The average localization error of less than 15 cm for average viewing distance of 15m suggests that our method can be applied to a broad range of surveillance applications that require the real-time computation of localization without using special hardware for acceleration.

A correspondence measure of 2D line segments in two different views is proposed in this paper. Such a quantitative measure is view-invariant and can handle line segment of arbitrary configuration in the 3D scene. A line-based people localization scheme is proposed by applying such a measure to improve the efficiency of [1]. By verifying whether 2D line samples from different views belong to the same person, computations associated with incorrectly reconstructed 3D line samples of people can be avoided. Experimental results show that people localization results, in terms of correctness and accuracy, comparable to [1] can be obtained with the proposed scheme, but with more than three times in computation speed.

In this paper, we propose an efficient people localization approach using multiple cameras based on axial representations of foreground regions. Unlike many previous methods that need to project all foreground pixels of all views to multiple reference planes via homography, we instead apply vanishing point-based line sampling to reduce the large amount of pixel processing so that computational efficiency can be greatly enhanced. Experimental simulations show that the proposed approach is more than 40 times faster than the compared, pixel-based localization method on average, without sacrificing the localization accuracy.

To model a scene for background subtraction, Gaussian mixture modeling (GMM) is a popular choice for its capability of adaptation to background variations. However, GMM often suffers from a tradeoff between robustness to background changes and sensitivity to foreground abnormalities and is inefficient in managing the tradeoff for various surveillance scenarios. By reviewing the formulations of GMM, we identify that such a tradeoff can be easily controlled by adaptive adjustments of the GMM's learning rates for image pixels at different locations and of distinct properties. A new rate control scheme based on high-level feedback is then developed to provide better regularization of background adaptation for GMM and to help resolving the tradeoff. Additionally, to handle lighting variations that change too fast to be caught by GMM, a heuristic rooting in frame difference is proposed to assist the proposed rate control scheme for reducing false foreground alarms. Experiments show the proposed learning rate control scheme, together with the heuristic for adaptation of over-quick lighting change, gives better performance than conventional GMM approaches.

In this paper, we propose a probabilistic method to model the dynamic traffic flow across non-overlapping camera views. By assuming the transition time of object movement follows a certain global model, we may infer the time-varying traffic status in the unseen region without performing explicit object correspondence between camera views. In this paper, we model object correspondence and parameter estimation as a unified problem under the proposed Expectation-Maximization (EM) based framework. By treating object correspondence as a latent random variable, the proposed framework can iteratively search for the optimal model parameters with the implicit consideration of object correspondence.

Local learning for classification is useful in dealing with various vision problems. One key factor for such approaches to be effective is to find good neighbors for the learning procedure. In this work, we describe a novel method to rank neighbors by learning a local distance function, and meanwhile to derive the local distance function by focusing on the high-ranked neighbors. The two aspects of considerations can be elegantly coupled through a well-defined objective function, motivated by a supervised ranking method called P-Norm Push. While the local distance functions are learned independently, they can be reshaped altogether so that their values can be directly compared. We apply the proposed method to the Caltech-101 dataset, and demonstrate the use of proper neighbors can improve the performance of classification techniques based on nearest-neighbor selection.

Human activity recognition is a popular topic in the field of computer vision. While most analysis algorithms take into consideration of the whole human body, the movements of merely the head and limbs are often informative enough in many practical applications. In this paper, a novel approach is proposed to track these extruding parts of a human body in consecutive images. Accordingly, a simplified torso-less pattern of gesture is proposed to represent human activities, and with its effectiveness verified subjectively by some experimental results. Such a representation can not only ensure the privacy of the person being watched, but is also suitable for real-time surveillance based on bandwidth-limited communication since only a very small amount of data are used compared to conventional approaches.

A novel approach to simultaneous scaling and enhancing of a digital image is proposed, where efficient code matching is utilized to classify various edge/corner patterns. Adaptive schemes of inter-/extra-polations are adopted to maintain a balance of smoothness and sharpness of the scaled image, with unlimited scaling factors. Unlike many complex methods for image scaling and enhancement, our approach is very simple, making hardware implementation feasible.

In order to handle device matching for analog circuits, some pairs of modules need to be placed symmetrically with respect to a common axis. In this paper, we deal with the module placement with symmetry constraints for analog design using the Transitive Closure Graph-Sequence (TCG-S) representation. Since the geometric relationships of modules are transparent to TCG-S and its induced operations, TCG-S has better flexibility than previous works in dealing with symmetry constraints. We first propose the necessary and sufficient conditions of TCG-S for symmetry modules. Then, we propose a polynomial-time packing algorithm for a TCG-S with symmetry constraints. Experimental results show that the TCG-S based algorithm results in the best area utilization.

A more efficient coding scheme for H.264 by heuristically assign macroblock partition types for video foreground and background coding is proposed. High visual quality of foreground regions are retained while low bit-rate background coding is achieved. More importantly, the encoding time is reduced significantly owing to the elimination of the exhausted searches over all partition types during the RD optimization

This paper proposes a potential-based path planning algorithm of articulated robots with 2-DOF joints. The algorithm is an extension of a previous algorithm developed for 3-DOF joints. While 3-DOF joints result in a very straightforward potential minimization algorithm, 2-DOF joints are obviously more practical for active operations. The proposed approach computes repulsive force and torque between charged objects by using generalized potential model. A collision-free path can be obtained by locally adjusting the robot configuration to search for minimum potential config urations using these force and torque. The optimization of path safeness, through the innovative potential minimization algorithm, makes the proposed approach unique. In order to speedup the computation, a sequential planning strategy is adopted. Simulation results show that the proposed algorithm works well compared with the algorithm for 3-DOF joints, in terms of collision avoidance and computation efficiency.

A novel collision avoidance algorithm is proposed to solve the path-planning problem of a high DOF robot manipulators in 3-D workspace. The algorithm is based on a generalized potential field model of 3-D workspace. The approach computes, similar to that done in electrostatics, repulsive force and torque between manipulator and obstacles using the workspace information directly. Using these force and torque, a collision-free path of a manipulator can be obtained by locally adjusting the manipulator configuration for minimum potential. The proposed approach is efficient since these potential gradients are analytically tractable. Simulation results show that the proposed algorithm works well, in terms of computation time and collision avoidance, for manipulators up to 6 links.

This paper proposes a novel path planning algorithm of 3-D articulated robots with moving bases based on a generalized potential field model. The approach computes, similar to that done in electrostatics, repulsive force and torque between charged objects. A collision-free path can be obtained by locally adjusting the robot configuration to search for minimum potential configurations using these force and torque. The proposed approach is efficient since these potential gradients are analytically tractable. In order to speedup the computation, a sequential planning strategy is adopted. Simulation results show that the proposed algorithm works well, in terms of collision avoidance and computation efficiency.

Visual tracking systems using background subtraction have been very popular largely due to their efficiency in extracting moving objects. However, such systems often compute the reference background by assuming no moving objects are present during the initialization stage, though the assumption may not be realistic. We propose an automatic way to perform background initialization using a probabilistic SVM (support vector machine). By formulating the problem as an on-line classification one, our approach has the potential to be real-time. SVM classification is carried out for all elements of each image frame by computing the output probabilities. Newly found background elements are evaluated and determined if they should be added to the solution. The process of background initialization continues until there are no more new background elements to be considered. As the features used in an SVM dictate the outcome of classification, we find that optical flow value and inter-frame difference are the two most important ones. Experimental results are included to demonstrate the efficiency of our method.

The paper proposes a collision avoidance algorithm to solve the problem of (local) path planning for a three-dimensional (3D) object moving among polyhedral obstacles. The algorithm is based on a generalized potential model of workspace (J.-H. Chuang, 1998) which assumes that the boundary of every 3D object is uniformly charged. According to the proposed approach, the repulsive force and torque between the moving object and the obstacles due to the above model is used to adjust the position and orientation of the object so as to keep it away from the obstacles while passing through a bottleneck in the free space. Simulation results demonstrate that the path of a 3D object thus obtained is indeed safe and spatially smooth. The adopted potential field is analytically tractable which makes the path planning efficient.

In this paper, the potential-based skeletonization approach for 2D medial axis transform (MAT), which identifies object skeleton as potential valleys using a Newtonian potential model in place of the distance function, is generalized to three dimensions. The generalized potential functions given by Chung (1998), which decay faster with distance than the Newtonian potential, is used for the 3D case. The efficiency of the proposed approach results from the fact that these functions and their gradients can be obtained in closed forms for polyhedral surfaces. According to the simulation results, the skeletons obtained with the proposed approach are closely related to the corresponding MAT skeletons. While the medial axis (surface) is 2D in general for a 3D object, the potential valleys, being one-dimensional, form a more realistic skeleton. Other desirable attributes of the algorithm include stability against perturbations of the object boundary, the flexibility to obtain partial skeleton directly, and low time complexity.

One of the existing approaches to path planning problems uses a potential function to represent the topological structure of the free space. Newtonian potential was used in Chuang and Ahuja (1998) to represent object and obstacles in the two-dimensional (2-D) workspace wherein their boundaries are assumed to be uniformly charged. In this paper, more general, nonuniform distributions are considered. It is shown that for linear or quadratic source distributions, the repulsion between two polygonal objects can be evaluated analytically. Simulation results show that by properly adjusting the charge distribution along obstacle/object boundaries, path planning results ran be improved in terms of collision avoidance, path length, etc.

A potential-based model of three-dimensional workspace is proposed for ensuring obstacle avoidance in path planning. It is assumed that the workspace boundary is uniformly distributed with generalized charges. The potential due to a point charge is inversely proportional to the distance to the power of an integer, the order of the potential function. It is shown that such potential functions and their gradients due to polyhedral surfaces can be derived analytically, and thus can facilitate efficient collision avoidance. Intuitively, the potential fields and their effects on object paths should be spatially continuous and smooth. The continuity and differentiability properties of a particular potential function are investigated. In theory, by minimizing the repulsion between object and obstacles, the approach completely eliminates the possibility of a collision between them if the dynamics of the moving object is ignored.

An analytically tractable potential field model of free space is presented. The model assumes that the border of every two dimensional (2D) region is uniformly charged. It is shown that the potential and the resulting repulsion (force and torque) between polygonal regions can he calculated in closed form. By using the Newtonian potential function, collision avoidance between object and obstacle thus modeled is guaranteed in a path planning problem. A local planner is developed for finding object paths going through narrow areas of free space where the obstacle avoidance is most important. Simulation results show that not only does individual object configuration of a path obtained with the proposed approach avoid obstacles effectively, the configurations also connect smoothly into a path.

This paper is concerned with efficient derivation of the medial axis transform of a 2D polygonal region. Instead of using the shortest distance to the region border, a potential field model is used for computational efficiency. The region border is assumed to be charged and the valleys of the resulting potential field are used to estimate the axes for the medial axis transform. The potential valleys are found by following the force field, thus, avoiding 2D search. The potential field is computed in closed form using equations of the border segments. The simple Newtonian potential is shown to be inadequate for this purpose. A higher order potential is defined which decays faster with distance than the inverse of distance. It is shown that as the potential order becomes arbitrarily large, the axes approach those computed using the shortest distance to the border. Algorithms are given for the computation of axes, which can run in linear parallel time for part of the axes having initial guesses. Experimental results are presented for a number of examples.

The paper proposes a new interconnection network topology, called varietal hypercube for large scale multicomputer systems. An n-dimensional varietal hypercube is constructed by two (n-1)-dimensional varietal hypercubes in a way similar to that for the hypercube except for some minor modifications. The resulting network has the same number of nodes and links as the hypercube, and has most of the desirable properties of the hypercube, including recursive structure, partionability, strong connectivity, and the ability to embed other architectures such as ring and mesh. The diameter of the varietal hypercube is about two thirds of the diameter of the hypercube. The average distance of the varietal hypercube is also smaller than that of the hypercube. Optimal routing and broadcasting algorithms which guarantee the shortest path communication are developed. Comparisons with other variations of the hypercube, such as twisted cube, folded hypercube, and crossed cube, are also included.

Newtonian potential function is used to represent polygonal objects and obstacles. The closed-form expression of this potential field and other gradient-related quantities are derived. Such results not only eliminate the problems associated with the discretization of the object and obstacles in evaluating the risk of collision, but also make the search for the optimal object configurations efficient. The object skeleton, a shape description of the moving object, is introduced to guide the moving object through narrow regions while the search is done at different stages. The free space can then be divided by the narrow regions where the path planning takes place-a very simple free space decomposition scheme. Successful global strategies are developed to connect the local plans into a safe and smooth global path.>