Search Results for author: Kai-Zhu Huang

Found 21 papers, 5 papers with code

Reliability Does Matter: An End-to-End Weakly Supervised Semantic Segmentation Approach

1 code implementation19 Nov 2019 Bingfeng Zhang, Jimin Xiao, Yunchao Wei, Ming-Jie Sun, Kai-Zhu Huang

Such reliable regions are then directly served as ground-truth labels for the parallel segmentation branch, where a newly designed dense energy loss function is adopted for optimization.

Image Classification Segmentation +2

Stochastic Conjugate Gradient Algorithm with Variance Reduction

1 code implementation27 Oct 2017 Xiao-Bo Jin, Xu-Yao Zhang, Kai-Zhu Huang, Guang-Gang Geng

Conjugate gradient (CG) methods are a class of important methods for solving linear equations and nonlinear optimization problems.

Computational Efficiency

A Fast Projected Fixed-Point Algorithm for Large Graph Matching

1 code implementation3 Jul 2012 Yao Lu, Kai-Zhu Huang, Cheng-Lin Liu

In particular, with high accuracy, our algorithm takes only a few seconds (in a PC) to match two graphs of 1, 000 nodes.

Graph Matching

Segmentation Mask Guided End-to-End Person Search

1 code implementation27 Aug 2019 Dingyuan Zheng, Jimin Xiao, Kai-Zhu Huang, Yao Zhao

Person search aims to search for a target person among multiple images recorded by multiple surveillance cameras, which faces various challenges from both pedestrian detection and person re-identification.

Pedestrian Detection Person Re-Identification +2

IAN: The Individual Aggregation Network for Person Search

no code implementations16 May 2017 Jimin Xiao, Yanchun Xie, Tammam Tillo, Kai-Zhu Huang, Yunchao Wei, Jiashi Feng

In addition, to relieve the negative effect caused by varying visual appearances of the same individual, IAN introduces a novel center loss that can increase the intra-class compactness of feature representations.

object-detection Object Detection +1

A Unified Gradient Regularization Family for Adversarial Examples

no code implementations19 Nov 2015 Chunchuan Lyu, Kai-Zhu Huang, Hai-Ning Liang

In this paper, we propose a unified framework to build robust machine learning models against adversarial examples.

BIG-bench Machine Learning Data Augmentation

Robust Text Detection in Natural Scene Images

no code implementations11 Jan 2013 Xu-Cheng Yin, Xuwang Yin, Kai-Zhu Huang, Hong-Wei Hao

Text detection in natural scene images is an important prerequisite for many content-based image analysis tasks.

Clustering Metric Learning +2

Manifold Adversarial Learning

no code implementations16 Jul 2018 Shufei Zhang, Kai-Zhu Huang, Jianke Zhu, Yang Liu

All the existing adversarial training methods consider only how the worst perturbed examples (i. e., adversarial examples) could affect the model output.

Long Short-Term Attention

no code implementations30 Oct 2018 Guoqiang Zhong, Xin Lin, Kang Chen, Qingyang Li, Kai-Zhu Huang

Attention is an important cognition process of humans, which helps humans concentrate on critical information during their perception and learning.

Beyond Attributes: Adversarial Erasing Embedding Network for Zero-shot Learning

no code implementations19 Nov 2018 Xiao-Bo Jin, Kai-Zhu Huang, Jianyu Miao

AEEN-HOA consists of two branches, i. e., the upper stream is capable of erasing some initially discovered regions, then the high-order attribute supervision is incorporated to characterize the relationship between the class attributes.

Attribute Zero-Shot Learning

Sparse Metric Learning via Smooth Optimization

no code implementations NeurIPS 2009 Yiming Ying, Kai-Zhu Huang, Colin Campbell

From this saddle representation, we develop an efficient smooth optimization approach for sparse metric learning although the learning model is based on a non-differential loss function.

Dimensionality Reduction Metric Learning

LEARNING ADVERSARIAL EXAMPLES WITH RIEMANNIAN GEOMETRY

no code implementations ICLR 2019 Shufei Zhang, Kai-Zhu Huang, Rui Zhang, Amir Hussain

In this paper, we propose a generalized framework that addresses the learning problem of adversarial examples with Riemannian geometry.

Generative Adversarial Classifier for Handwriting Characters Super-Resolution

no code implementations18 Jan 2019 Zhuang Qian, Kai-Zhu Huang, Qiufeng Wang, Jimin Xiao, Rui Zhang

Generative Adversarial Networks (GAN) receive great attentions recently due to its excellent performance in image generation, transformation, and super-resolution.

Classification General Classification +2

Joint Multi-Label Attention Networks for Social Text Annotation

no code implementations NAACL 2019 Hang Dong, Wei Wang, Kai-Zhu Huang, Frans Coenen

To better utilise this information, we design a framework that separates the title from the content of a document and apply a title-guided attention mechanism over each sentence in the content.

Sentence text annotation

On Model Robustness Against Adversarial Examples

no code implementations15 Nov 2019 Shufei Zhang, Kai-Zhu Huang, Zenglin Xu

We propose to exploit an energy function to describe the stability and prove that reducing such energy guarantees the robustness against adversarial examples.

Deep Minimax Probability Machine

no code implementations20 Nov 2019 Lirong He, Ziyi Guo, Kai-Zhu Huang, Zenglin Xu

In a worst-case scenario, MPM tries to minimize an upper bound of misclassification probabilities, considering the global information (i. e., mean and covariance information of each class).

Hybrid Channel Based Pedestrian Detection

no code implementations28 Dec 2019 Fiseha B. Tesema, Hong Wu, Mingjian Chen, Junpeng Lin, William Zhu, Kai-Zhu Huang

When using a more advanced RPN in our framework, our approach can be further improved and get competitive results on both benchmarks.

Pedestrian Detection

Robust Generative Adversarial Network

no code implementations ICLR 2020 Shufei Zhang, Zhuang Qian, Kai-Zhu Huang, Jimin Xiao, Yuan He

Generative adversarial networks (GANs) are powerful generative models, but usually suffer from instability and generalization problem which may lead to poor generations.

Generative Adversarial Network

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