Search Results for author: Hong Zhang

Found 69 papers, 18 papers with code

Optimizing SLAM Evaluation Footprint Through Dynamic Range Coverage Analysis of Datasets

no code implementations13 Sep 2022 Islam Ali, Hong Zhang

Simultaneous Localization and Mapping (SLAM) is considered an ever-evolving problem due to its usage in many applications.

Simultaneous Localization and Mapping

Prognostic Significance of Tumor-Infiltrating Lymphocytes Using Deep Learning on Pathology Images in Colorectal Cancers

no code implementations23 Aug 2022 Anran Liu, Xingyu Li, Hongyi Wu, Bangwei Guo, Jitendra Jonnagaddala, Hong Zhang, Xu Steven Xu

Methods We developed an automated, multiscale LinkNet workflow for quantifying cellular-level TILs for CRC tumors using H&E-stained images.

Predicting microsatellite instability and key biomarkers in colorectal cancer from H&E-stained images: Achieving SOTA predictive performance with fewer data using Swin Transformer

no code implementations22 Aug 2022 Bangwei Guo, Xingyu Li, Jitendra Jonnagaddala, Hong Zhang, Xu Steven Xu

In this study, based on the latest Hierarchical Vision Transformer using Shifted Windows (Swin-T), we developed an efficient workflow for biomarkers in CRC (MSI, hypermutation, chromosomal instability, CpG island methylator phenotype, BRAF, and TP53 mutation) that only required relatively small datasets, but achieved the state-of-the-art (SOTA) predictive performance.

Accelerating Numerical Solvers for Large-Scale Simulation of Dynamical System via NeurVec

1 code implementation7 Aug 2022 Zhongzhan Huang, Senwei Liang, Hong Zhang, Haizhao Yang, Liang Lin

Conventional numerical solvers used in the simulation are significantly limited by the step size for time integration, which hampers efficiency and feasibility especially when high accuracy is desired.

Perspective Phase Angle Model for Polarimetric 3D Reconstruction

1 code implementation20 Jul 2022 Guangcheng Chen, Li He, Yisheng Guan, Hong Zhang

Current polarimetric 3D reconstruction methods, including those in the well-established shape from polarization literature, are all developed under the orthographic projection assumption.

3D Reconstruction Surface Normal Estimation

Highlight Specular Reflection Separation based on Tensor Low-rank and Sparse Decomposition Using Polarimetric Cues

no code implementations7 Jul 2022 Moein Shakeri, Hong Zhang

This regularization boosts the performance of the method to recover an accurate diffuse image by handling the color distortion, a common problem of chromaticity-based methods, especially in case of strong specular reflection.

Reflection Removal

Diagnostic Communication and Visual System based on Vehicle UDS Protocol

no code implementations25 Jun 2022 Hong Zhang, Ding Li

Unified Diagnostic Services (UDS) is a diagnostic communication protocol used in electronic control units (ECUs) within automotive electronics, which is specified in the ISO 14229-1.

Provable Benefit of Multitask Representation Learning in Reinforcement Learning

no code implementations13 Jun 2022 Yuan Cheng, Songtao Feng, Jing Yang, Hong Zhang, Yingbin Liang

To the best of our knowledge, this is the first theoretical study that characterizes the benefit of representation learning in exploration-based reward-free multitask RL for both upstream and downstream tasks.

Offline RL reinforcement-learning +1

PNODE: A memory-efficient neural ODE framework based on high-level adjoint differentiation

1 code implementation2 Jun 2022 Hong Zhang, Wenjun Zhao

On the image classification problems, PNODE is up to two times faster than the vanilla neural ODE and up to 2. 3 times faster than the best existing reverse-accurate method.

Image Classification

A robust and lightweight deep attention multiple instance learning algorithm for predicting genetic alterations

no code implementations31 May 2022 Bangwei Guo, Xingyu Li, Miaomiao Yang, Hong Zhang, Xu Steven Xu

In addition, compared to the published models for genetic alterations, AMIML provided a significant improvement for predicting a wide range of genes (e. g., KMT2C, TP53, and SETD2 for KIRC; ERBB2, BRCA1, and BRCA2 for BRCA; JAK1, POLE, and MTOR for UCEC) as well as produced outstanding predictive models for other clinically relevant gene mutations, which have not been reported in the current literature.

Deep Attention Multiple Instance Learning

FinBERT-MRC: financial named entity recognition using BERT under the machine reading comprehension paradigm

1 code implementation31 May 2022 Yuzhe Zhang, Hong Zhang

We conduct experiments on a publicly available Chinese financial dataset ChFinAnn and a real-word bussiness dataset AdminPunish.

Machine Reading Comprehension named-entity-recognition +1

FedDUAP: Federated Learning with Dynamic Update and Adaptive Pruning Using Shared Data on the Server

no code implementations25 Apr 2022 Hong Zhang, Ji Liu, Juncheng Jia, Yang Zhou, Huaiyu Dai, Dejing Dou

Despite achieving remarkable performance, Federated Learning (FL) suffers from two critical challenges, i. e., limited computational resources and low training efficiency.

Federated Learning

Colorectal cancer survival prediction using deep distribution based multiple-instance learning

no code implementations24 Apr 2022 Xingyu Li, Jitendra Jonnagaddala, Min Cen, Hong Zhang, Xu Steven Xu

Several deep learning algorithms have been developed to predict survival of cancer patients using whole slide images (WSIs). However, identification of image phenotypes within the WSIs that are relevant to patient survival and disease progression is difficult for both clinicians, and deep learning algorithms.

Multiple Instance Learning Survival Prediction +1

Condition-Invariant and Compact Visual Place Description by Convolutional Autoencoder

1 code implementation15 Apr 2022 Hanjing Ye, Weinan Chen, Jingwen Yu, Li He, Yisheng Guan, Hong Zhang

We employ a high-level layer of a pre-trained CNN to generate features, and train a CAE to map the features to a low-dimensional space to improve the condition invariance property of the descriptor and reduce its dimension at the same time.

Visual Place Recognition

Sparse Optical Flow-Based Line Feature Tracking

no code implementations7 Apr 2022 Qiang Fu, Hongshan Yu, Islam Ali, Hong Zhang

To achieve this goal, an efficient two endpoint tracking (TET) method is presented: first, describe a given line feature with its two endpoints; next, track the two endpoints based on SOF to obtain two new tracked endpoints by minimizing a pixel-level grayscale residual function; finally, connect the two tracked endpoints to generate a new line feature.

Optical Flow Estimation Pose Estimation

Optimize Deep Learning Models for Prediction of Gene Mutations Using Unsupervised Clustering

no code implementations31 Mar 2022 Zihan Chen, Xingyu Li, Miaomiao Yang, Hong Zhang, Xu Steven Xu

We showed that unsupervised clustering of image patches could help identify predictive patches, exclude patches lack of predictive information, and therefore improve prediction on gene mutations in all three different cancer types, compared with the WSI based method without selection of image patches and models based on only tumor regions.

Multiple Instance Learning

Deep Polarimetric HDR Reconstruction

no code implementations27 Mar 2022 Juiwen Ting, Moein Shakeri, Hong Zhang

This paper proposes a novel learning based high-dynamic-range (HDR) reconstruction method using a polarization camera.

HDR Reconstruction

DSRRTracker: Dynamic Search Region Refinement for Attention-based Siamese Multi-Object Tracking

no code implementations21 Mar 2022 JiaXu Wan, Hong Zhang, Jin Zhang, Yuan Ding, Yifan Yang, Yan Li, Xuliang Li

Many multi-object tracking (MOT) methods follow the framework of "tracking by detection", which associates the target objects-of-interest based on the detection results.

Multi-Object Tracking

DePS: An improved deep learning model for de novo peptide sequencing

no code implementations16 Mar 2022 Cheng Ge, Yi Lu, Jia Qu, Liangxu Xie, Feng Wang, Hong Zhang, Ren Kong, Shan Chang

De novo peptide sequencing from mass spectrometry data is an important method for protein identification.

Are We Ready for Robust and Resilient SLAM? A Framework For Quantitative Characterization of SLAM Datasets

no code implementations23 Feb 2022 Islam Ali, Hong Zhang

In order to fill this void, characterization of the operating conditions of SLAM systems is essential in order to provide an environment for quantitative measurement of robustness and resilience.

Improving Feature Extraction from Histopathological Images Through A Fine-tuning ImageNet Model

no code implementations3 Jan 2022 Xingyu Li, Min Cen, Jinfeng Xu, Hong Zhang, Xu Steven Xu

The extracted features from the finetuned FTX2048 exhibited significantly higher accuracy for predicting tisue types of CRC compared to the off the shelf feature directly from Xception based on ImageNet database.

Transfer Learning

Online Mutual Adaptation of Deep Depth Prediction and Visual SLAM

no code implementations7 Nov 2021 Shing Yan Loo, Moein Shakeri, Sai Hong Tang, Syamsiah Mashohor, Hong Zhang

In addition, we compare our online adaptation framework against the state-of-the-art pre-trained depth prediction CNNs to show that our online adapted depth prediction CNN outperforms the depth prediction CNNs that have been trained on a large collection of datasets.

Depth Estimation Depth Prediction +1

A Retrospective Analysis using Deep-Learning Models for Prediction of Survival Outcome and Benefit of Adjuvant Chemotherapy in Stage II/III Colorectal Cancer

no code implementations5 Nov 2021 Xingyu Li, Jitendra Jonnagaddala, Shuhua Yang, Hong Zhang, Xu Steven Xu

We developed a novel deep-learning algorithm (CRCNet) using whole-slide images from Molecular and Cellular Oncology (MCO) to predict survival benefit of adjuvant chemotherapy in stage II/III CRC.

whole slide images

Foresight of Graph Reinforcement Learning Latent Permutations Learnt by Gumbel Sinkhorn Network

no code implementations23 Oct 2021 Tianqi Shen, Hong Zhang, Ding Yuan, Jiaping Xiao, Yifan Yang

Vital importance has necessity to be attached to cooperation in multi-agent environments, as a result of which some reinforcement learning algorithms combined with graph neural networks have been proposed to understand the mutual interplay between agents.

Graph Attention reinforcement-learning

Stiffness-aware neural network for learning Hamiltonian systems

no code implementations ICLR 2022 Senwei Liang, Zhongzhan Huang, Hong Zhang

We propose stiffness-aware neural network (SANN), a new method for learning Hamiltonian dynamical systems from data.

Rethinking Training from Scratch for Object Detection

1 code implementation6 Jun 2021 Yang Li, Hong Zhang, Yu Zhang

The ImageNet pre-training initialization is the de-facto standard for object detection.

object-detection Object Detection

Deep Snapshot HDR Reconstruction Based on the Polarization Camera

no code implementations12 May 2021 Juiwen Ting, Xuesong Wu, Kangkang Hu, Hong Zhang

In our approach, we first study the relationship among polarizer orientation, degree and angle of polarization of light to the exposure time of a pixel in the polarization image.

HDR Reconstruction

GPNAS: A Neural Network Architecture Search Framework Based on Graphical Predictor

no code implementations19 Mar 2021 Dige Ai, Hong Zhang

In this paper, we propose a framework to decouple network structure from operator search space, and use two BOHBs to search alternatively.

Neural Architecture Search

Polarimetric Monocular Dense Mapping Using Relative Deep Depth Prior

no code implementations10 Feb 2021 Moein Shakeri, Shing Yan Loo, Hong Zhang

This paper is concerned with polarimetric dense map reconstruction based on a polarization camera with the help of relative depth information as a prior.

Mass-Conserving Implicit-Explicit Methods for Coupled Compressible Navier-Stokes Equations

no code implementations22 Jan 2021 Shinhoo Kang, Emil M. Constantinescu, Hong Zhang, Robert L. Jacob

Earth system models are composed of coupled components that separately model systems such as the global atmosphere, ocean, and land surface.

Numerical Analysis Numerical Analysis Computational Physics Fluid Dynamics 76N06, 65M08, 65L04

Direct Depth Learning Network for Stereo Matching

no code implementations10 Dec 2020 Hong Zhang, Haojie Li, Shenglun Chen, Tiantian Yan, Zhihui Wang, Guo Lu, Wanli Ouyang

To make the Adaptive-Grained Depth Refinement stage robust to the coarse depth and adaptive to the depth range of the points, the Granularity Uncertainty is introduced to Adaptive-Grained Depth Refinement stage.

Autonomous Driving Depth Estimation +1

Full Matching on Low Resolution for Disparity Estimation

no code implementations10 Dec 2020 Hong Zhang, Shenglun Chen, Zhihui Wang, Haojie Li, Wanli Ouyang

To this end, we first propose to decompose the full matching task into multiple stages of the cost aggregation module.

Disparity Estimation

Object-aware Feature Aggregation for Video Object Detection

no code implementations23 Oct 2020 Qichuan Geng, Hong Zhang, Na Jiang, Xiaojuan Qi, Liangjun Zhang, Zhong Zhou

As a consequence, augmenting features with such prior knowledge can effectively improve the classification and localization performance.

object-detection Object Recognition +1

A Comprehensive Review for MRF and CRF Approaches in Pathology Image Analysis

no code implementations29 Sep 2020 Yixin Li, Chen Li, Xiaoyan Li, Kai Wang, Md Mamunur Rahaman, Changhao Sun, Hao Chen, Xinran Wu, Hong Zhang, Qian Wang

In this review, we present a comprehensive overview of pathology image analysis based on the markov random fields (MRFs) and conditional random fields (CRFs), which are two popular random field models.

PL-VINS: Real-Time Monocular Visual-Inertial SLAM with Point and Line Features

1 code implementation16 Sep 2020 Qiang Fu, Jialong Wang, Hongshan Yu, Islam Ali, Feng Guo, Yijia He, Hong Zhang

This paper presents PL-VINS, a real-time optimization-based monocular VINS method with point and line features, developed based on the state-of-the-art point-based VINS-Mono \cite{vins}.

Pose Estimation

Fast ORB-SLAM without Keypoint Descriptors

no code implementations22 Aug 2020 Qiang Fu, Hongshan Yu, Xiaolong Wang, Zhengeng Yang, Hong Zhang, Ajmal Mian

ORB-SLAM2 \cite{orbslam2} is a benchmark method in this domain, however, it consumes significant time for computing descriptors that never get reused unless a frame is selected as a keyframe.

Robotics Computational Geometry I.4.0; I.4.9

To Filter Prune, or to Layer Prune, That Is The Question

1 code implementation11 Jul 2020 Sara Elkerdawy, Mostafa Elhoushi, Abhineet Singh, Hong Zhang, Nilanjan Ray

LayerPrune presents a set of layer pruning methods based on different criteria that achieve higher latency reduction than filter pruning methods on similar accuracy.

DeepRelativeFusion: Dense Monocular SLAM using Single-Image Relative Depth Prediction

no code implementations7 Jun 2020 Shing Yan Loo, Syamsiah Mashohor, Sai Hong Tang, Hong Zhang

To this end, we use a visual SLAM algorithm to reliably recover the camera poses and semi-dense depth maps of the keyframes, and then use relative depth prediction to densify the semi-dense depth maps and refine the keyframe pose-graph.

Depth Estimation Depth Prediction +2

Gated Path Selection Network for Semantic Segmentation

no code implementations19 Jan 2020 Qichuan Geng, Hong Zhang, Xiaojuan Qi, Ruigang Yang, Zhong Zhou, Gao Huang

Semantic segmentation is a challenging task that needs to handle large scale variations, deformations and different viewpoints.

Semantic Segmentation

Implicit Semantic Data Augmentation for Deep Networks

1 code implementation NeurIPS 2019 Yulin Wang, Xuran Pan, Shiji Song, Hong Zhang, Cheng Wu, Gao Huang

Our work is motivated by the intriguing property that deep networks are surprisingly good at linearizing features, such that certain directions in the deep feature space correspond to meaningful semantic transformations, e. g., adding sunglasses or changing backgrounds.

Image Augmentation

Improved Techniques for Training Adaptive Deep Networks

1 code implementation ICCV 2019 Hao Li, Hong Zhang, Xiaojuan Qi, Ruigang Yang, Gao Huang

Adaptive inference is a promising technique to improve the computational efficiency of deep models at test time.

Knowledge Distillation

Semi-Supervised Monocular Depth Estimation with Left-Right Consistency Using Deep Neural Network

2 code implementations18 May 2019 Ali Jahani Amiri, Shing Yan Loo, Hong Zhang

In general, semi-supervised training is preferred as it does not suffer from the weaknesses of either supervised training, resulting from the difference in the cameras and the LiDARs field of view, or unsupervised training, resulting from the poor depth accuracy that can be recovered from a stereo pair.

Depth Prediction Monocular Depth Estimation

Lightweight Monocular Depth Estimation Model by Joint End-to-End Filter pruning

1 code implementation13 May 2019 Sara Elkerdawy, Hong Zhang, Nilanjan Ray

This is achieved by removing the least important features with a novel joint end-to-end filter pruning.

Monocular Depth Estimation

Moving Object Detection under Discontinuous Change in Illumination Using Tensor Low-Rank and Invariant Sparse Decomposition

no code implementations CVPR 2019 Moein Shakeri, Hong Zhang

We are interested in moving object detection in applications involving time-lapse image sequences for which current methods mistakenly group moving objects and illumination changes into foreground.

Moving Object Detection object-detection

Single-step Options for Adversary Driving

no code implementations20 Mar 2019 Nazmus Sakib, Hengshuai Yao, Hong Zhang, Shangling Jui

In this paper, we use reinforcement learning for safety driving in adversary settings.


Fine-Grained Vehicle Classification with Unsupervised Parts Co-occurrence Learning

no code implementations23 Jan 2019 Sara Elkerdawy, Nilanjan Ray, Hong Zhang

In addition, we achieve 95. 5% and 93. 19% on CompCars on both train-test splits, 70-30 and 50-50, outperforming the other methods by 4. 5% and 8% respectively.

Classification Fine-Grained Image Classification +3

Human Pose Estimation with Spatial Contextual Information

no code implementations7 Jan 2019 Hong Zhang, Hao Ouyang, Shu Liu, Xiaojuan Qi, Xiaoyong Shen, Ruigang Yang, Jiaya Jia

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.

Pose Estimation

Part-level Car Parsing and Reconstruction from Single Street View

no code implementations27 Nov 2018 Qichuan Geng, Hong Zhang, Xinyu Huang, Sen Wang, Feixiang Lu, Xinjing Cheng, Zhong Zhou, Ruigang Yang

As it is labor-intensive to annotate semantic parts on real street views, we propose a specific approach to implicitly transfer part features from synthesized images to real street views.

Car Pose Estimation Domain Adaptation +1

CNN-SVO: Improving the Mapping in Semi-Direct Visual Odometry Using Single-Image Depth Prediction

2 code implementations1 Oct 2018 Shing Yan Loo, Ali Jahani Amiri, Syamsiah Mashohor, Sai Hong Tang, Hong Zhang

Reliable feature correspondence between frames is a critical step in visual odometry (VO) and visual simultaneous localization and mapping (V-SLAM) algorithms.

Depth Estimation Depth Prediction +3

Theoretical and Empirical Study of Adversarial Examples

no code implementations27 Sep 2018 Fuchen Liu, Hongwei Shang, Hong Zhang

We show that under some symmetrical assumptions, label smoothing, logit squeezing, weight decay, mix up and feature smoothing all produce an unbiased estimation of the decision boundary with smaller estimated variance.

Data Augmentation

Unsupervised Learning of Stereo Matching

no code implementations ICCV 2017 Chao Zhou, Hong Zhang, Xiaoyong Shen, Jiaya Jia

However, due to the limitations of these datasets and the difficulty of collecting new stereo data, current methods fail in real-life cases.

Stereo Matching Stereo Matching Hand

Fast Shadow Detection from a Single Image Using a Patched Convolutional Neural Network

1 code implementation26 Sep 2017 Sepideh Hosseinzadeh, Moein Shakeri, Hong Zhang

In recent years, various shadow detection methods from a single image have been proposed and used in vision systems; however, most of them are not appropriate for the robotic applications due to the expensive time complexity.

Shadow Detection

Face Recognition using Multi-Modal Low-Rank Dictionary Learning

no code implementations15 Mar 2017 Homa Foroughi, Moein Shakeri, Nilanjan Ray, Hong Zhang

Face recognition has been widely studied due to its importance in different applications; however, most of the proposed methods fail when face images are occluded or captured under illumination and pose variations.

Dictionary Learning Face Recognition +1

Convolutional Neural Network-Based Block Up-sampling for Intra Frame Coding

no code implementations22 Feb 2017 Yue Li, Dong Liu, Houqiang Li, Li Li, Feng Wu, Hong Zhang, Haitao Yang

A block can be down-sampled before being compressed by normal intra coding, and then up-sampled to its original resolution.


Object Classification with Joint Projection and Low-rank Dictionary Learning

no code implementations5 Dec 2016 Homa Foroughi, Nilanjan Ray, Hong Zhang

To address these issues, we propose a joint projection and low-rank dictionary learning method using dual graph constraints (JP-LRDL).

Classification Dictionary Learning +1

On The Stability of Video Detection and Tracking

no code implementations20 Nov 2016 Hong Zhang, Naiyan Wang

Lastly, based on this metric, we evaluate several existing methods for video detection and show how they affect accuracy and stability.

Multi-Scale Patch Aggregation (MPA) for Simultaneous Detection and Segmentation

no code implementations CVPR 2016 Shu Liu, Xiaojuan Qi, Jianping Shi, Hong Zhang, Jiaya Jia

Aiming at simultaneous detection and segmentation (SDS), we propose a proposal-free framework, which detect and segment object instances via mid-level patches.

Object Proposal Generation

Joint Projection and Dictionary Learning using Low-rank Regularization and Graph Constraints

no code implementations24 Mar 2016 Homa Foroughi, Nilanjan Ray, Hong Zhang

To address this issue, we propose a joint projection and dictionary learning using low-rank regularization and graph constraints (JPDL-LR).

Dictionary Learning Dimensionality Reduction +2

COROLA: A Sequential Solution to Moving Object Detection Using Low-rank Approximation

no code implementations13 May 2015 Moein Shakeri, Hong Zhang

Extracting moving objects from a video sequence and estimating the background of each individual image are fundamental issues in many practical applications such as visual surveillance, intelligent vehicle navigation, and traffic monitoring.

Moving Object Detection object-detection

Convolutional Neural Network-Based Image Representation for Visual Loop Closure Detection

no code implementations20 Apr 2015 Yi Hou, Hong Zhang, Shilin Zhou

Deep convolutional neural networks (CNN) have recently been shown in many computer vision and pattern recog- nition applications to outperform by a significant margin state- of-the-art solutions that use traditional hand-crafted features.

Loop Closure Detection Simultaneous Localization and Mapping

Stopping Rules for Bag-of-Words Image Search and Its Application in Appearance-Based Localization

no code implementations28 Dec 2013 Kiana Hajebi, Hong Zhang

We introduce a notion of difficulty for the image matching problems and propose methods that reduce the amount of computations required for the feature vector-quantization task in BoW by exploiting the fact that easier queries need less computational resources.

Image Retrieval Quantization

An Efficient Index for Visual Search in Appearance-based SLAM

no code implementations27 Sep 2013 Kiana Hajebi, Hong Zhang

Vector-quantization can be a computationally expensive step in visual bag-of-words (BoW) search when the vocabulary is large.


Cannot find the paper you are looking for? You can Submit a new open access paper.