arXiv subjects

Hao Chen

Publications and source records attributed to Hao Chen.

At least 1,315 records · Page 73Linked to original sources

arXiv2020

BlendMask: Top-Down Meets Bottom-Up for Instance Segmentation

Instance segmentation is one of the fundamental vision tasks. Recently, fully convolutional instance segmentation methods have drawn much attention as they are often simpler and more efficient than two-stage approaches like Mask R-CNN. To date, almost all such approaches fall behind the two-stage Mask R-CNN method in mask precision when models have similar computation complexity, leaving great room for improvement. In this work, we achieve improved mask prediction by effectively combining instance-level information with semantic information with lower-level fine-granularity. Our main contribution is a blender module which draws inspiration from both top-down and bottom-up instance segmentation approaches. The proposed BlendMask can effectively predict dense per-pixel position-sensitive instance features with very few channels, and learn attention maps for each instance with merely one convolution layer, thus being fast in inference. BlendMask can be easily incorporated with the state-of-the-art one-stage detection frameworks and outperforms Mask R-CNN under the same training schedule while being 20% faster. A light-weight version of BlendMask achieves $ 34.2% $ mAP at 25 FPS evaluated on a single 1080Ti GPU card. Because of its simplicity and efficacy, we hope that our BlendMask could serve as a simple yet strong baseline for a wide range of instance-wise prediction tasks. Code is available at https://git.io/AdelaiDet

cs.CV↗

arXiv2020

Transport of Finite Size Self-Propelled Particles Confined in a 2D Zigzag Channel with Gaussian Colored Noise

The directional transport of finite size self-propelled Brownian particles confined in a 2D zigzag channel with colored noise is investigated. The noises(noise parallel to x-axis and y-axis), the asymmetry parameter Δk, the ratio f(ratio of the particle radius and the bottleneck half width), the selfpropelled speed v0 have joint effect on the particles. The average velocity of self-propelled particles is significantly different from passive particles. The average velocity exhibits complicated behavior with increasing self-propelled speed v0

cond-mat.stat-mech↗

arXiv2020

Integrating Combined Task and Motion Planning with Compliant Control

Planning a motion for inserting pegs remains an open problem. The difficulty lies in both the inevitable errors in the grasps of a robotic hand and absolute precision problems in robot joint motors. This paper proposes an integral method to solve the problem. The method uses combined task and motion planning to plan the grasps and motion for a dual-arm robot to pick up the objects and move them to assembly poses. Then, it controls the dual-arm robot using a compliant strategy (a combination of linear search, spiral search, and impedance control) to finish up the insertion. The method is implemented on a dual-arm Universal Robots 3 robot. Six objects, including a connector with fifteen peg-in-hole pairs for detailed analysis and other five objects with different contours of pegs and holes for additional validation, were tested by the robot. Experimental results show reasonable force-torque signal changes and end-effector position changes. The proposed method exhibits high robustness and high fidelity in successfully conducting planned peg-in-hole tasks.

cs.RO↗

arXiv2020

Privacy-preserving Incremental ADMM for Decentralized Consensus Optimization

The alternating direction method of multipliers (ADMM) has been recently recognized as a promising optimizer for large-scale machine learning models. However, there are very few results studying ADMM from the aspect of communication costs, especially jointly with privacy preservation, which are critical for distributed learning. We investigate the communication efficiency and privacy-preservation of ADMM by solving the consensus optimization problem over decentralized networks. Since walk algorithms can reduce communication load, we first propose incremental ADMM (I-ADMM) based on the walk algorithm, the updating order of which follows a Hamiltonian cycle instead. However, I-ADMM cannot guarantee the privacy for agents against external eavesdroppers even if the randomized initialization is applied. To protect privacy for agents, we then propose two privacy-preserving incremental ADMM algorithms, i.e., PI-ADMM1 and PI-ADMM2, where perturbation over step sizes and primal variables is adopted, respectively. Through theoretical analyses, we prove the convergence and privacy preservation for PI-ADMM1, which are further supported by numerical experiments. Besides, simulations demonstrate that the proposed PI-ADMM1 and PI-ADMM2 algorithms are communication efficient compared with state-of-the-art methods.

math.OC↗

arXiv2020

Cardiovascular risk and work stress in biomedical researchers in China: An observational, big data study protocol

Introduction: Internet technologies could strengthen data collection and integration and have been used extensively in public health research. It is necessary to apply this technology to further investigate the behaviour and health of biomedical researchers. A browser-based extension was developed by researchers and clinicians to promote the collection and analysis of researchers' behavioural and psychological data. This protocol illustrates an observational study aimed at (1) characterising the health status of biomedical researchers in China and assessing work stress, job satisfaction, role conflict, role ambiguity, and family support; (2) identifying the association between work, behaviour, and health; and (3) investigating the association between behaviour and mental status. Our findings will contribute to the understanding of the influences of job, work environment, and family support on the mental and physical health of biomedical researchers. Methods and analysis: This is a prospective observational study; all candidates will be recruited from China. Participants will install an extension on their Internet browsers, which will collect data when they are accessing PubMed. A web-based survey will be sent to the user interfaces every 6 months that will involve sociodemographic variables, perceived stress scale, job satisfaction scale, role conflict and ambiguity scale, and family support scale. Machine-learning algorithms will analyse the data generated during daily access. Ethics and dissemination: This study received ethical approval from the ethics committee of the Shanghai Children's Medical Centre (reference number SCMCIRB-K2018082). Study results will be disseminated through peer-reviewed publications and conference presentations.

cs.CY↗

arXiv2020

SANNS: Scaling Up Secure Approximate k-Nearest Neighbors Search

The $k$-Nearest Neighbor Search ($k$-NNS) is the backbone of several cloud-based services such as recommender systems, face recognition, and database search on text and images. In these services, the client sends the query to the cloud server and receives the response in which case the query and response are revealed to the service provider. Such data disclosures are unacceptable in several scenarios due to the sensitivity of data and/or privacy laws. In this paper, we introduce SANNS, a system for secure $k$-NNS that keeps client's query and the search result confidential. SANNS comprises two protocols: an optimized linear scan and a protocol based on a novel sublinear time clustering-based algorithm. We prove the security of both protocols in the standard semi-honest model. The protocols are built upon several state-of-the-art cryptographic primitives such as lattice-based additively homomorphic encryption, distributed oblivious RAM, and garbled circuits. We provide several contributions to each of these primitives which are applicable to other secure computation tasks. Both of our protocols rely on a new circuit for the approximate top-$k$ selection from $n$ numbers that is built from $O(n + k^2)$ comparators. We have implemented our proposed system and performed extensive experimental results on four datasets in two different computation environments, demonstrating more than $18-31\times$ faster response time compared to optimally implemented protocols from the prior work. Moreover, SANNS is the first work that scales to the database of 10 million entries, pushing the limit by more than two orders of magnitude.

cs.DS↗

arXiv2020

Planning to Build Soma Blocks Using a Dual-arm Robot

This paper presents a planner that can automatically find an optimal assembly sequence for a dual-arm robot to assemble the soma blocks. The planner uses the mesh model of objects and the final state of the assembly to generate all possible assembly sequence and evaluate the optimal assembly sequence by considering the stability, graspability, assemblability, as well as the need for a second arm. Especially, the need for a second arm is considered when supports from worktables and other workpieces are not enough to produce a stable assembly. The planner will refer to an assisting grasp to additionally hold and support the unstable components so that the robot can further assemble new workpieces and finally reach a stable state. The output of the planner is the optimal assembly orders, candidate grasps, assembly directions, and the assisting grasps if any. The output of the planner can be used to guide a dual-arm robot to perform the assembly task. The planner is verified in both simulations and real-world executions.

cs.RO↗

arXiv2020

NAS-FCOS: Fast Neural Architecture Search for Object Detection

The success of deep neural networks relies on significant architecture engineering. Recently neural architecture search (NAS) has emerged as a promise to greatly reduce manual effort in network design by automatically searching for optimal architectures, although typically such algorithms need an excessive amount of computational resources, e.g., a few thousand GPU-days. To date, on challenging vision tasks such as object detection, NAS, especially fast versions of NAS, is less studied. Here we propose to search for the decoder structure of object detectors with search efficiency being taken into consideration. To be more specific, we aim to efficiently search for the feature pyramid network (FPN) as well as the prediction head of a simple anchor-free object detector, namely FCOS, using a tailored reinforcement learning paradigm. With carefully designed search space, search algorithms and strategies for evaluating network quality, we are able to efficiently search a top-performing detection architecture within 4 days using 8 V100 GPUs. The discovered architecture surpasses state-of-the-art object detection models (such as Faster R-CNN, RetinaNet and FCOS) by 1.5 to 3.5 points in AP on the COCO dataset, with comparable computation complexity and memory footprint, demonstrating the efficacy of the proposed NAS for object detection.

cs.CV↗

arXiv2020

ABCNet: Real-time Scene Text Spotting with Adaptive Bezier-Curve Network

Scene text detection and recognition has received increasing research attention. Existing methods can be roughly categorized into two groups: character-based and segmentation-based. These methods either are costly for character annotation or need to maintain a complex pipeline, which is often not suitable for real-time applications. Here we address the problem by proposing the Adaptive Bezier-Curve Network (ABCNet). Our contributions are three-fold: 1) For the first time, we adaptively fit arbitrarily-shaped text by a parameterized Bezier curve. 2) We design a novel BezierAlign layer for extracting accurate convolution features of a text instance with arbitrary shapes, significantly improving the precision compared with previous methods. 3) Compared with standard bounding box detection, our Bezier curve detection introduces negligible computation overhead, resulting in superiority of our method in both efficiency and accuracy. Experiments on arbitrarily-shaped benchmark datasets, namely Total-Text and CTW1500, demonstrate that ABCNet achieves state-of-the-art accuracy, meanwhile significantly improving the speed. In particular, on Total-Text, our realtime version is over 10 times faster than recent state-of-the-art methods with a competitive recognition accuracy. Code is available at https://tinyurl.com/AdelaiDet

cs.CV↗

arXiv2020

Robust Multimodal Brain Tumor Segmentation via Feature Disentanglement and Gated Fusion

Accurate medical image segmentation commonly requires effective learning of the complementary information from multimodal data. However, in clinical practice, we often encounter the problem of missing imaging modalities. We tackle this challenge and propose a novel multimodal segmentation framework which is robust to the absence of imaging modalities. Our network uses feature disentanglement to decompose the input modalities into the modality-specific appearance code, which uniquely sticks to each modality, and the modality-invariant content code, which absorbs multimodal information for the segmentation task. With enhanced modality-invariance, the disentangled content code from each modality is fused into a shared representation which gains robustness to missing data. The fusion is achieved via a learning-based strategy to gate the contribution of different modalities at different locations. We validate our method on the important yet challenging multimodal brain tumor segmentation task with the BRATS challenge dataset. With competitive performance to the state-of-the-art approaches for full modality, our method achieves outstanding robustness under various missing modality(ies) situations, significantly exceeding the state-of-the-art method by over 16% in average for Dice on whole tumor segmentation.

cs.CV↗

arXiv2020

TLFW: A Three-layer Framework in Wireless Rechargeable Sensor Network with a Mobile Base Station

Wireless sensor networks as the base support for the Internet of things has been a large number of popularity and application. Such as intelligent agriculture, we have to use the sensor network to obtain the growth environmental data of crops, etc.. However, the difficulty of power supply of wireless nodes has seriously hindered the application and development of Internet of things. In order to solve this problem, people use low-power, sleep scheduling and other energy-saving methods on the nodes. Although these methods can prolong the working time of nodes, they will eventually become invalid because of the exhaustion of energy. The use of solar energy, wind energy, and wireless signals in the environment to obtain energy is another way to solve the energy problem of nodes. However, these methods are affected by weather, environment and other factors, and are unstable. Thus, the discontinuity work of the node is caused. In recent years, the development of wireless power transfer (WPT) has brought another solution to this problem. In this paper, a three-layer framework is proposed for mobile station data collection in rechargeable wireless sensor networks to keep the node running forever, named TLFW which includes the sensor layer, cluster head layer, and mobile station layer. And the framework can minimize the total energy consumption of the system. The simulation results show that the scheme can reduce the energy consumption of the entire system, compared with a Mobile Station in a Rechargeable Sensor Network(MSiRSN).

eess.SP↗

arXiv2020

Fifth-degree elastic potential for predictive stress-strain relations and elastic instabilities under large strain and complex loading in Si

Materials under complex loading develop large strains and often transition via an elastic instability, as observed in both simple and complex systems. Here, we present Si I under large strain in terms of Lagrangian strain by an $5^{th}$-order elastic potential found by minimizing error relative to density functional theory (DFT) results. The Cauchy stress-Lagrangian strain curves for arbitrary complex loadings are in excellent correspondence with DFT results, including elastic instability driving Si I$\rightarrow$II phase transformation (PT) and the shear instabilities. PT conditions for Si I$\rightarrow$II under action of cubic axial stresses are linear in Cauchy stresses in agreement with DFT predictions. Such elastic potential permits study of elastic instabilities and orientational dependence leading to different PTs, slip, twinning, or fracture, providing a fundamental basis for continuum simulations of crystal behavior under extreme loading.

cond-mat.mtrl-sci↗

arXiv2020

Progressive Object Transfer Detection

Recent development of object detection mainly depends on deep learning with large-scale benchmarks. However, collecting such fully-annotated data is often difficult or expensive for real-world applications, which restricts the power of deep neural networks in practice. Alternatively, humans can detect new objects with little annotation burden, since humans often use the prior knowledge to identify new objects with few elaborately-annotated examples, and subsequently generalize this capacity by exploiting objects from wild images. Inspired by this procedure of learning to detect, we propose a novel Progressive Object Transfer Detection (POTD) framework. Specifically, we make three main contributions in this paper. First, POTD can leverage various object supervision of different domains effectively into a progressive detection procedure. Via such human-like learning, one can boost a target detection task with few annotations. Second, POTD consists of two delicate transfer stages, i.e., Low-Shot Transfer Detection (LSTD), and Weakly-Supervised Transfer Detection (WSTD). In LSTD, we distill the implicit object knowledge of source detector to enhance target detector with few annotations. It can effectively warm up WSTD later on. In WSTD, we design a recurrent object labelling mechanism for learning to annotate weakly-labeled images. More importantly, we exploit the reliable object supervision from LSTD, which can further enhance the robustness of target detector in the WSTD stage. Finally, we perform extensive experiments on a number of challenging detection benchmarks with different settings. The results demonstrate that, our POTD outperforms the recent state-of-the-art approaches.

cs.CV↗

arXiv2020

Deep Learning for Source Code Modeling and Generation: Models, Applications and Challenges

Deep Learning (DL) techniques for Natural Language Processing have been evolving remarkably fast. Recently, the DL advances in language modeling, machine translation and paragraph understanding are so prominent that the potential of DL in Software Engineering cannot be overlooked, especially in the field of program learning. To facilitate further research and applications of DL in this field, we provide a comprehensive review to categorize and investigate existing DL methods for source code modeling and generation. To address the limitations of the traditional source code models, we formulate common program learning tasks under an encoder-decoder framework. After that, we introduce recent DL mechanisms suitable to solve such problems. Then, we present the state-of-the-art practices and discuss their challenges with some recommendations for practitioners and researchers as well.

cs.SE↗

arXiv2020

Automatic detection and counting of retina cell nuclei using deep learning

The ability to automatically detect, classify, calculate the size, number, and grade of retinal cells and other biological objects is critically important in eye disease like age-related macular degeneration (AMD). In this paper, we developed an automated tool based on deep learning technique and Mask R-CNN model to analyze large datasets of transmission electron microscopy (TEM) images and quantify retinal cells with high speed and precision. We considered three categories for outer nuclear layer (ONL) cells: live, intermediate, and pyknotic. We trained the model using a dataset of 24 samples. We then optimized the hyper-parameters using another set of 6 samples. The results of this research, after applying to the test datasets, demonstrated that our method is highly accurate for automatically detecting, categorizing, and counting cell nuclei in the ONL of the retina. Performance of our model was tested using general metrics: general mean average precision (mAP) for detection; and precision, recall, F1-score, and accuracy for categorizing and counting.

cs.CV↗

arXiv2020

DS-PASS: Detail-Sensitive Panoramic Annular Semantic Segmentation through SwaftNet for Surrounding Sensing

Semantically interpreting the traffic scene is crucial for autonomous transportation and robotics systems. However, state-of-the-art semantic segmentation pipelines are dominantly designed to work with pinhole cameras and train with narrow Field-of-View (FoV) images. In this sense, the perception capacity is severely limited to offer higher-level confidence for upstream navigation tasks. In this paper, we propose a network adaptation framework to achieve Panoramic Annular Semantic Segmentation (PASS), which allows to re-use conventional pinhole-view image datasets, enabling modern segmentation networks to comfortably adapt to panoramic images. Specifically, we adapt our proposed SwaftNet to enhance the sensitivity to details by implementing attention-based lateral connections between the detail-critical encoder layers and the context-critical decoder layers. We benchmark the performance of efficient segmenters on panoramic segmentation with our extended PASS dataset, demonstrating that the proposed real-time SwaftNet outperforms state-of-the-art efficient networks. Furthermore, we assess real-world performance when deploying the Detail-Sensitive PASS (DS-PASS) system on a mobile robot and an instrumented vehicle, as well as the benefit of panoramic semantics for visual odometry, showing the robustness and potential to support diverse navigational applications.

cs.CV↗

arXiv2020

Memory-Efficient Hierarchical Neural Architecture Search for Image Denoising

Recently, neural architecture search (NAS) methods have attracted much attention and outperformed manually designed architectures on a few high-level vision tasks. In this paper, we propose HiNAS (Hierarchical NAS), an effort towards employing NAS to automatically design effective neural network architectures for image denoising. HiNAS adopts gradient based search strategies and employs operations with adaptive receptive field to build an flexible hierarchical search space. During the search stage, HiNAS shares cells across different feature levels to save memory and employ an early stopping strategy to avoid the collapse issue in NAS, and considerably accelerate the search speed. The proposed HiNAS is both memory and computation efficient, which takes only about 4.5 hours for searching using a single GPU. We evaluate the effectiveness of our proposed HiNAS on two different datasets, namely an additive white Gaussian noise dataset BSD500, and a realistic noise dataset SIM1800. Experimental results show that the architecture found by HiNAS has fewer parameters and enjoys a faster inference speed, while achieving highly competitive performance compared with state-of-the-art methods. We also present analysis on the architectures found by NAS. HiNAS also shows good performance on experiments for image de-raining.

cs.CV↗

arXiv2020

Unsupervised Bidirectional Cross-Modality Adaptation via Deeply Synergistic Image and Feature Alignment for Medical Image Segmentation

Unsupervised domain adaptation has increasingly gained interest in medical image computing, aiming to tackle the performance degradation of deep neural networks when being deployed to unseen data with heterogeneous characteristics. In this work, we present a novel unsupervised domain adaptation framework, named as Synergistic Image and Feature Alignment (SIFA), to effectively adapt a segmentation network to an unlabeled target domain. Our proposed SIFA conducts synergistic alignment of domains from both image and feature perspectives. In particular, we simultaneously transform the appearance of images across domains and enhance domain-invariance of the extracted features by leveraging adversarial learning in multiple aspects and with a deeply supervised mechanism. The feature encoder is shared between both adaptive perspectives to leverage their mutual benefits via end-to-end learning. We have extensively evaluated our method with cardiac substructure segmentation and abdominal multi-organ segmentation for bidirectional cross-modality adaptation between MRI and CT images. Experimental results on two different tasks demonstrate that our SIFA method is effective in improving segmentation performance on unlabeled target images, and outperforms the state-of-the-art domain adaptation approaches by a large margin.

eess.IV↗