arXiv subjects

Shu Liu

Publications and source records attributed to Shu Liu.

At least 181 records · Page 10Linked to original sources

arXiv2020

PointGroup: Dual-Set Point Grouping for 3D Instance Segmentation

Instance segmentation is an important task for scene understanding. Compared to the fully-developed 2D, 3D instance segmentation for point clouds have much room to improve. In this paper, we present PointGroup, a new end-to-end bottom-up architecture, specifically focused on better grouping the points by exploring the void space between objects. We design a two-branch network to extract point features and predict semantic labels and offsets, for shifting each point towards its respective instance centroid. A clustering component is followed to utilize both the original and offset-shifted point coordinate sets, taking advantage of their complementary strength. Further, we formulate the ScoreNet to evaluate the candidate instances, followed by the Non-Maximum Suppression (NMS) to remove duplicates. We conduct extensive experiments on two challenging datasets, ScanNet v2 and S3DIS, on which our method achieves the highest performance, 63.6% and 64.0%, compared to 54.9% and 54.4% achieved by former best solutions in terms of mAP with IoU threshold 0.5.

cs.CV↗

arXiv2020

3DSSD: Point-based 3D Single Stage Object Detector

Currently, there have been many kinds of voxel-based 3D single stage detectors, while point-based single stage methods are still underexplored. In this paper, we first present a lightweight and effective point-based 3D single stage object detector, named 3DSSD, achieving a good balance between accuracy and efficiency. In this paradigm, all upsampling layers and refinement stage, which are indispensable in all existing point-based methods, are abandoned to reduce the large computation cost. We novelly propose a fusion sampling strategy in downsampling process to make detection on less representative points feasible. A delicate box prediction network including a candidate generation layer, an anchor-free regression head with a 3D center-ness assignment strategy is designed to meet with our demand of accuracy and speed. Our paradigm is an elegant single stage anchor-free framework, showing great superiority to other existing methods. We evaluate 3DSSD on widely used KITTI dataset and more challenging nuScenes dataset. Our method outperforms all state-of-the-art voxel-based single stage methods by a large margin, and has comparable performance to two stage point-based methods as well, with inference speed more than 25 FPS, 2x faster than former state-of-the-art point-based methods.

cs.CV↗

arXiv2019

Push for Center Learning via Orthogonalization and Subspace Masking for Person Re-Identification

Person re-identification aims to identify whether pairs of images belong to the same person or not. This problem is challenging due to large differences in camera views, lighting and background. One of the mainstream in learning CNN features is to design loss functions which reinforce both the class separation and intra-class compactness. In this paper, we propose a novel Orthogonal Center Learning method with Subspace Masking for person re-identification. We make the following contributions: (i) we develop a center learning module to learn the class centers by simultaneously reducing the intra-class differences and inter-class correlations by orthogonalization; (ii) we introduce a subspace masking mechanism to enhance the generalization of the learned class centers; and (iii) we devise to integrate the average pooling and max pooling in a regularizing manner that fully exploits their powers. Extensive experiments show that our proposed method consistently outperforms the state-of-the-art methods on the large-scale ReID datasets including Market-1501, DukeMTMC-ReID, CUHK03 and MSMT17.

cs.CV↗

arXiv2019

Explicit Constructions of Two-Dimensional Reed-Solomon Codes in High Insertion and Deletion Noise Regime

Insertion and deletion (insdel for short) errors are synchronization errors in communication systems caused by the loss of positional information in the message. Reed-Solomon codes have gained a lot of interest due to its encoding simplicity, well structuredness and list-decoding capability in the classical setting. This interest also translates to the insdel metric setting, as the Guruswami-Sudan decoding algorithm can be utilized to provide a deletion correcting algorithm in the insdel metric. Nevertheless, there have been few studies on the insdel error-correcting capability of Reed-Solomon codes. Our main contributions in this paper are explicit constructions of two families of 2-dimensional Reed-Solomon codes with insdel error-correcting capabilities asymptotically reaching those provided by the Singleton bound. The first construction gives a family of Reed-Solomon codes with insdel error-correcting capability asymptotic to its length. The second construction provides a family of Reed Solomon codes with an exact insdel error-correcting capability up to its length. Both our constructions improve the previously known construction of 2-dimensional Reed-Solomon codes whose insdel error-correcting capability is only logarithmic on the code length.

cs.IT↗

arXiv2019

EmbedMask: Embedding Coupling for One-stage Instance Segmentation

Current instance segmentation methods can be categorized into segmentation-based methods that segment first then do clustering, and proposal-based methods that detect first then predict masks for each instance proposal using repooling. In this work, we propose a one-stage method, named EmbedMask, that unifies both methods by taking advantages of them. Like proposal-based methods, EmbedMask builds on top of detection models making it strong in detection capability. Meanwhile, EmbedMask applies extra embedding modules to generate embeddings for pixels and proposals, where pixel embeddings are guided by proposal embeddings if they belong to the same instance. Through this embedding coupling process, pixels are assigned to the mask of the proposal if their embeddings are similar. The pixel-level clustering enables EmbedMask to generate high-resolution masks without missing details from repooling, and the existence of proposal embedding simplifies and strengthens the clustering procedure to achieve high speed with higher performance than segmentation-based methods. Without any bells and whistles, EmbedMask achieves comparable performance as Mask R-CNN, which is the representative two-stage method, and can produce more detailed masks at a higher speed. Code is available at github.com/yinghdb/EmbedMask.

cs.CV↗

arXiv2019

Pilot HI Survey of Planck Galactic Cold Clumps with FAST

We present a pilot HI survey of 17 Planck Galactic Cold Clumps (PGCCs) with the Five-hundred-meter Aperture Spherical radio Telescope (FAST). HI Narrow Self-Absorption (HINSA) is an effective method to detect cold HI being mixed with molecular hydrogen H$_2$ and improves our understanding of the atomic to molecular transition in the interstellar medium. HINSA was found in 58\% PGCCs that we observed. The column density of HINSA was found to have an intermediate correlation with that of $^{13}$CO, following $\rm log( N(HINSA)) = (0.52\pm 0.26) log(N_{^{13}CO}) + (10 \pm 4.1) $. HI abundance relative to total hydrogen [HI]/[H] has an average value of $4.4\times 10^{-3}$, which is about 2.8 times of the average value of previous HINSA surveys toward molecular clouds. For clouds with total column density N$\rm_H >5 \times 10^{20}$ cm$^{-2}$, an inverse correlation between HINSA abundance and total hydrogen column density is found, confirming the depletion of cold HI gas during molecular gas formation in more massive clouds. Nonthermal line width of $^{13}$CO is about 0-0.5 km s$^{-1}$ larger than that of HINSA. One possible explanation of narrower nonthermal width of HINSA is that HINSA region is smaller than that of $^{13}$CO. Based on an analytic model of H$_2$ formation and H$_2$ dissociation by cosmic ray, we found the cloud ages to be within 10$^{6.7}$-10$^{7.0}$ yr for five sources.

astro-ph.GA↗

arXiv2019

Hierarchical Point-Edge Interaction Network for Point Cloud Semantic Segmentation

We achieve 3D semantic scene labeling by exploring semantic relation between each point and its contextual neighbors through edges. Besides an encoder-decoder branch for predicting point labels, we construct an edge branch to hierarchically integrate point features and generate edge features. To incorporate point features in the edge branch, we establish a hierarchical graph framework, where the graph is initialized from a coarse layer and gradually enriched along the point decoding process. For each edge in the final graph, we predict a label to indicate the semantic consistency of the two connected points to enhance point prediction. At different layers, edge features are also fed into the corresponding point module to integrate contextual information for message passing enhancement in local regions. The two branches interact with each other and cooperate in segmentation. Decent experimental results on several 3D semantic labeling datasets demonstrate the effectiveness of our work.

cs.CV↗

arXiv2019

Driver Identification via the Steering Wheel

Driver identification has emerged as a vital research field, where both practitioners and researchers investigate the potential of driver identification to enable a personalized driving experience. Within recent years, a selection of studies have reported that individuals could be perfectly identified based on their driving behavior under controlled conditions. However, research investigating the potential of driver identification under naturalistic conditions claim accuracies only marginally higher than random guess. The paper at hand provides a comprehensive summary of the recent work, highlighting the main discrepancies in the design of the machine learning approaches, primarily the window length parameter that was considered. Key findings further indicate that the longitudinal vehicle control information is particularly useful for driver identification, leaving the research gap on the extent to which the lateral vehicle control can be used for reliable identification. Building upon existing work, we provide a novel approach for the design of the window length parameter that provides evidence that reliable driver identification can be achieved with data limited to the steering wheel only. The results and insights in this paper are based on data collected from the largest naturalistic driving study conducted in this field. Overall, a neural network based on GRUs was found to provide better identification performance than traditional methods, increasing the prediction accuracy from under 15\% to over 65\% for 15 drivers. When leveraging the full field study dataset, comprising 72 drivers, the accuracy of identification prediction of the approach improved a random guess approach by a factor of 25.

stat.ML↗

arXiv2019

Fast Point R-CNN

We present a unified, efficient and effective framework for point-cloud based 3D object detection. Our two-stage approach utilizes both voxel representation and raw point cloud data to exploit respective advantages. The first stage network, with voxel representation as input, only consists of light convolutional operations, producing a small number of high-quality initial predictions. Coordinate and indexed convolutional feature of each point in initial prediction are effectively fused with the attention mechanism, preserving both accurate localization and context information. The second stage works on interior points with their fused feature for further refining the prediction. Our method is evaluated on KITTI dataset, in terms of both 3D and Bird's Eye View (BEV) detection, and achieves state-of-the-arts with a 15FPS detection rate.

cs.CV↗

arXiv2019

STD: Sparse-to-Dense 3D Object Detector for Point Cloud

We present a new two-stage 3D object detection framework, named sparse-to-dense 3D Object Detector (STD). The first stage is a bottom-up proposal generation network that uses raw point cloud as input to generate accurate proposals by seeding each point with a new spherical anchor. It achieves a high recall with less computation compared with prior works. Then, PointsPool is applied for generating proposal features by transforming their interior point features from sparse expression to compact representation, which saves even more computation time. In box prediction, which is the second stage, we implement a parallel intersection-over-union (IoU) branch to increase awareness of localization accuracy, resulting in further improved performance. We conduct experiments on KITTI dataset, and evaluate our method in terms of 3D object and Bird's Eye View (BEV) detection. Our method outperforms other state-of-the-arts by a large margin, especially on the hard set, with inference speed more than 10 FPS.

cs.CV↗

arXiv2019

Soil and soil CO2 magnify greenhouse effect

Soil has been recognized as an indirect driver of global warming by regulating atmospheric greenhouse gases. However, in view of the higher heat capacity and CO2 concentration in soil than those in atmosphere, the direct contributions of soil to greenhouse effect may be non-ignorable. Through field manipulation of CO2 concentration both in soil and atmosphere, we demonstrated that the soil-retained heat and its slow transmission process within soil may cause slower heat leaking from the earth. Furthermore, soil air temperature was non-linearly affected by soil CO2 concentration with the highest value under 7500 ppm CO2. This study indicates that the soil and soil CO2, together with atmospheric CO2, play indispensable roles in fueling the greenhouse effect. We proposed that anthropogenic changes in soils should be focused in understanding drivers of the globe warming.

physics.geo-ph↗

arXiv2019

Planning the scientific applications of the Five-hundred-meter Aperture Spherical radio Telescope

The Five-hundred-meter Aperture Spherical radio Telescope (FAST) is by far the largest telescope of any kind ever built. FAST produced its first light in September 2016 and it is now under commissioning, with normal operation to commence in late 2019. During testing and early science operation, FAST has started making astronomical discoveries, particularly pulsars of various kinds, including millisecond pulsars, binaries, gamma-ray pulsars, etc. The papers in this mini-volume propose ambitious observational projects to advance our knowledge of astronomy, astrophysics and fundamental physics in many ways. Although it may take FAST many years to achieve all the goals explained in these papers, taken together they define a powerful strategic vision for the next decade.

astro-ph.IM↗

arXiv2019

The Interstellar Medium: The Key Component in Galactic Evolution and Modern Cosmology

The gases of the interstellar medium (ISM) possess orders of magnitude more mass than those of all the stars combined and are thus the prime component of the baryonic universe. With L-band surface sensitivity even better than the planned phase one Square-Kilometer-Array (SKA1), the Five-hundred-meter Aperture Spherical radio Telescope (FAST) promises unprecedented insights into two of the primary components of ISM, namely, the atomic hydrogen (HI) and the hydroxyl molecule (OH). We discuss here the evolving landscape of our understanding of ISM, particularly, its complex phases, the magnetic fields within, the so-called dark molecular gas (DMG), high velocity clouds, and the connection between local and distant ISM. We lay out, in broad strokes, several expected FAST projects, including an all northern-sky high-resolution HI survey (22,000 deg$^2$, 3\arcmin\ FWHM beam, 0.2 km/s), targeted OH mapping, searching for absorption or masing signals, and etc. Currently under commissioning, the commensal observing mode of FAST will be capable of simultaneously obtaining HI and pulsar data streams, making large-scale surveys in both science areas more efficient.

astro-ph.IM↗

arXiv2019

Associatively Segmenting Instances and Semantics in Point Clouds

A 3D point cloud describes the real scene precisely and intuitively.To date how to segment diversified elements in such an informative 3D scene is rarely discussed. In this paper, we first introduce a simple and flexible framework to segment instances and semantics in point clouds simultaneously. Then, we propose two approaches which make the two tasks take advantage of each other, leading to a win-win situation. Specifically, we make instance segmentation benefit from semantic segmentation through learning semantic-aware point-level instance embedding. Meanwhile, semantic features of the points belonging to the same instance are fused together to make more accurate per-point semantic predictions. Our method largely outperforms the state-of-the-art method in 3D instance segmentation along with a significant improvement in 3D semantic segmentation. Code has been made available at: https://github.com/WXinlong/ASIS.

cs.CV↗

arXiv2019

Human Pose Estimation with Spatial Contextual Information

We explore the importance of spatial contextual information in human pose estimation. Most state-of-the-art pose networks are trained in a multi-stage manner and produce several auxiliary predictions for deep supervision. With this principle, we present two conceptually simple and yet computational efficient modules, namely Cascade Prediction Fusion (CPF) and Pose Graph Neural Network (PGNN), to exploit underlying contextual information. Cascade prediction fusion accumulates prediction maps from previous stages to extract informative signals. The resulting maps also function as a prior to guide prediction at following stages. To promote spatial correlation among joints, our PGNN learns a structured representation of human pose as a graph. Direct message passing between different joints is enabled and spatial relation is captured. These two modules require very limited computational complexity. Experimental results demonstrate that our method consistently outperforms previous methods on MPII and LSP benchmark.

cs.CV↗

arXiv2018

IPOD: Intensive Point-based Object Detector for Point Cloud

We present a novel 3D object detection framework, named IPOD, based on raw point cloud. It seeds object proposal for each point, which is the basic element. This paradigm provides us with high recall and high fidelity of information, leading to a suitable way to process point cloud data. We design an end-to-end trainable architecture, where features of all points within a proposal are extracted from the backbone network and achieve a proposal feature for final bounding inference. These features with both context information and precise point cloud coordinates yield improved performance. We conduct experiments on KITTI dataset, evaluating our performance in terms of 3D object detection, Bird's Eye View (BEV) detection and 2D object detection. Our method accomplishes new state-of-the-art , showing great advantage on the hard set.

cs.CV↗

arXiv2018

Sequential Context Encoding for Duplicate Removal

Duplicate removal is a critical step to accomplish a reasonable amount of predictions in prevalent proposal-based object detection frameworks. Albeit simple and effective, most previous algorithms utilize a greedy process without making sufficient use of properties of input data. In this work, we design a new two-stage framework to effectively select the appropriate proposal candidate for each object. The first stage suppresses most of easy negative object proposals, while the second stage selects true positives in the reduced proposal set. These two stages share the same network structure, \ie, an encoder and a decoder formed as recurrent neural networks (RNN) with global attention and context gate. The encoder scans proposal candidates in a sequential manner to capture the global context information, which is then fed to the decoder to extract optimal proposals. In our extensive experiments, the proposed method outperforms other alternatives by a large margin.

cs.CV↗

arXiv2018

Deep Anticipation: Light Weight Intelligent Mobile Sensing in IoT by Recurrent Architecture

The rapid growth of IoT era is shaping the future of mobile services. Advanced communication technology enables a heterogeneous connectivity where mobile devices broadcast information to everything. Mobile applications such as robotics and vehicles connecting to cloud and surroundings transfer the short-range on-board sensor perception system to long-range mobile-sensing perception system. However, the mobile sensing perception brings new challenges for how to efficiently analyze and intelligently interpret the deluge of IoT data in mission- critical services. In this article, we model the challenges as latency, packet loss and measurement noise which severely deteriorate the reliability and quality of IoT data. We integrate the artificial intelligence into IoT to tackle these challenges. We propose a novel architecture that leverages recurrent neural networks (RNN) and Kalman filtering to anticipate motions and interac- tions between objects. The basic idea is to learn environment dynamics by recurrent networks. To improve the robustness of IoT communication, we use the idea of Kalman filtering and deploy a prediction and correction step. In this way, the architecture learns to develop a biased belief between prediction and measurement in the different situation. We demonstrate our approach with synthetic and real-world datasets with noise that mimics the challenges of IoT communications. Our method brings a new level of IoT intelligence. It is also lightweight compared to other state-of-the-art convolutional recurrent architecture and is ideally suitable for the resource-limited mobile applications.

cs.CV↗