Search Results for author: Qiang Qiu

Found 66 papers, 16 papers with code

Large Convolutional Model Tuning via Filter Subspace

no code implementations1 Mar 2024 Wei Chen, Zichen Miao, Qiang Qiu

Furthermore, each filter atom can be recursively decomposed as a combination of another set of atoms, which naturally expands the number of tunable parameters in the filter subspace.

Training Bayesian Neural Networks with Sparse Subspace Variational Inference

1 code implementation16 Feb 2024 Junbo Li, Zichen Miao, Qiang Qiu, Ruqi Zhang

Bayesian neural networks (BNNs) offer uncertainty quantification but come with the downside of substantially increased training and inference costs.

Uncertainty Quantification Variational Inference

Binary Latent Diffusion

no code implementations CVPR 2023 Ze Wang, Jiang Wang, Zicheng Liu, Qiang Qiu

In this paper, we show that a binary latent space can be explored for compact yet expressive image representations.

Image Generation Quantization +1

Learning to Retain while Acquiring: Combating Distribution-Shift in Adversarial Data-Free Knowledge Distillation

no code implementations CVPR 2023 Gaurav Patel, Konda Reddy Mopuri, Qiang Qiu

To this end, at every generator update, we aim to maintain the student's performance on previously encountered examples while acquiring knowledge from samples of the current distribution.

Data-free Knowledge Distillation Meta-Learning +1

Energy-Inspired Self-Supervised Pretraining for Vision Models

no code implementations2 Feb 2023 Ze Wang, Jiang Wang, Zicheng Liu, Qiang Qiu

In the proposed framework, we model energy estimation and data restoration as the forward and backward passes of a single network without any auxiliary components, e. g., an extra decoder.

Colorization Denoising +2

Seq-UPS: Sequential Uncertainty-aware Pseudo-label Selection for Semi-Supervised Text Recognition

no code implementations31 Aug 2022 Gaurav Patel, Jan Allebach, Qiang Qiu

To this end, we propose a pseudo-label generation and an uncertainty-based data selection framework for semi-supervised text recognition.

Pseudo Label

Learning to Learn Dense Gaussian Processes for Few-Shot Learning

no code implementations NeurIPS 2021 Ze Wang, Zichen Miao, XianTong Zhen, Qiang Qiu

In contrast to sparse Gaussian processes, we define a set of dense inducing variables to be of a much larger size than the support set in each task, which collects prior knowledge from experienced tasks.

Few-Shot Learning Gaussian Processes +2

Image Generation using Continuous Filter Atoms

no code implementations NeurIPS 2021 Ze Wang, Seunghyun Hwang, Zichen Miao, Qiang Qiu

In this paper, we model the subspace of convolutional filters with a neural ordinary differential equation (ODE) to enable gradual changes in generated images.

Image-to-Image Translation Navigate +1

Spatiotemporal Joint Filter Decomposition in 3D Convolutional Neural Networks

no code implementations NeurIPS 2021 Zichen Miao, Ze Wang, Xiuyuan Cheng, Qiang Qiu

In this paper, we introduce spatiotemporal joint filter decomposition to decouple spatial and temporal learning, while preserving spatiotemporal dependency in a video.

Action Recognition

TESSERACT: Gradient Flip Score to Secure Federated Learning Against Model Poisoning Attacks

no code implementations19 Oct 2021 Atul Sharma, Wei Chen, Joshua Zhao, Qiang Qiu, Somali Chaterji, Saurabh Bagchi

The attack uses the intuition that simply by changing the sign of the gradient updates that the optimizer is computing, for a set of malicious clients, a model can be diverted from the optima to increase the test error rate.

Federated Learning Model Poisoning

Meta-OLE: Meta-learned Orthogonal Low-Rank Embedding

no code implementations29 Sep 2021 Ze Wang, Yue Lu, Qiang Qiu

We introduce Meta-OLE, a new geometry-regularized method for fast adaptation to novel tasks in few-shot image classification.

Classification Few-Shot Image Classification +1

Continual Learning with Filter Atom Swapping

1 code implementation ICLR 2022 Zichen Miao, Ze Wang, Wei Chen, Qiang Qiu

In this paper, we first enforce a low-rank filter subspace by decomposing convolutional filters within each network layer over a small set of filter atoms.

Continual Learning

A Joint Subspace View to Convolutional Neural Networks

no code implementations29 Sep 2021 Ze Wang, Xiuyuan Cheng, Guillermo Sapiro, Qiang Qiu

In other words, a CNN is now reduced to layers of filter atoms, typically a few hundred of parameters per layer, with a common block of subspace coefficients shared across layers.

Adaptive Convolutions with Per-pixel Dynamic Filter Atom

no code implementations ICCV 2021 Ze Wang, Zichen Miao, Jun Hu, Qiang Qiu

Applying feature dependent network weights have been proved to be effective in many fields.

Translation

Cirrus: A Long-range Bi-pattern LiDAR Dataset

no code implementations5 Dec 2020 Ze Wang, Sihao Ding, Ying Li, Jonas Fenn, Sohini Roychowdhury, Andreas Wallin, Lane Martin, Scott Ryvola, Guillermo Sapiro, Qiang Qiu

Point density varies significantly across such a long range, and different scanning patterns further diversify object representation in LiDAR.

3D Object Detection Autonomous Driving +2

Using Text to Teach Image Retrieval

no code implementations19 Nov 2020 Haoyu Dong, Ze Wang, Qiang Qiu, Guillermo Sapiro

Image retrieval relies heavily on the quality of the data modeling and the distance measurement in the feature space.

Image Retrieval Retrieval +2

Learning to Learn Variational Semantic Memory

1 code implementation NeurIPS 2020 XianTong Zhen, Yingjun Du, Huan Xiong, Qiang Qiu, Cees G. M. Snoek, Ling Shao

The variational semantic memory accrues and stores semantic information for the probabilistic inference of class prototypes in a hierarchical Bayesian framework.

Few-Shot Learning General Knowledge +1

ACDC: Weight Sharing in Atom-Coefficient Decomposed Convolution

no code implementations4 Sep 2020 Ze Wang, Xiuyuan Cheng, Guillermo Sapiro, Qiang Qiu

We then explicitly regularize CNN kernels by enforcing decomposed coefficients to be shared across sub-structures, while leaving each sub-structure only its own dictionary atoms, a few hundreds of parameters typically, which leads to dramatic model reductions.

Image Classification

Graph Convolution with Low-rank Learnable Local Filters

2 code implementations ICLR 2021 Xiuyuan Cheng, Zichen Miao, Qiang Qiu

Recent deep models using graph convolutions provide an appropriate framework to handle such non-Euclidean data, but many of them, particularly those based on global graph Laplacians, lack expressiveness to capture local features required for representation of signals lying on the non-Euclidean grid.

Action Recognition Facial Expression Recognition +2

Low to High Dimensional Modality Hallucination using Aggregated Fields of View

1 code implementation13 Jul 2020 Kausic Gunasekar, Qiang Qiu, Yezhou Yang

While hallucinating data from a modality with richer information, e. g., RGB to depth, has been researched extensively, we investigate the more challenging low-to-high modality hallucination with interesting use cases in robotics and autonomous systems.

Hallucination Vocal Bursts Intensity Prediction

Differential 3D Facial Recognition: Adding 3D to Your State-of-the-Art 2D Method

no code implementations3 Apr 2020 J. Matias Di Martino, Fernando Suzacq, Mauricio Delbracio, Qiang Qiu, Guillermo Sapiro

Active illumination is a prominent complement to enhance 2D face recognition and make it more robust, e. g., to spoofing attacks and low-light conditions.

3D Reconstruction Face Recognition

SalGaze: Personalizing Gaze Estimation Using Visual Saliency

no code implementations23 Oct 2019 Zhuoqing Chang, Matias Di Martino, Qiang Qiu, Steven Espinosa, Guillermo Sapiro

Traditional gaze estimation methods typically require explicit user calibration to achieve high accuracy.

Gaze Estimation

Range Adaptation for 3D Object Detection in LiDAR

no code implementations26 Sep 2019 Ze Wang, Sihao Ding, Ying Li, Minming Zhao, Sohini Roychowdhury, Andreas Wallin, Guillermo Sapiro, Qiang Qiu

To the best of our knowledge, this paper is the first attempt to study cross-range LiDAR adaptation for object detection in point clouds.

3D Object Detection Autonomous Driving +2

A Dictionary Approach to Domain-Invariant Learning in Deep Networks

no code implementations NeurIPS 2020 Ze Wang, Xiuyuan Cheng, Guillermo Sapiro, Qiang Qiu

In this paper, we consider domain-invariant deep learning by explicitly modeling domain shifts with only a small amount of domain-specific parameters in a Convolutional Neural Network (CNN).

Domain Adaptation

Scale-Equivariant Neural Networks with Decomposed Convolutional Filters

no code implementations25 Sep 2019 Wei Zhu, Qiang Qiu, Robert Calderbank, Guillermo Sapiro, Xiuyuan Cheng

Encoding the input scale information explicitly into the representation learned by a convolutional neural network (CNN) is beneficial for many vision tasks especially when dealing with multiscale input signals.

Image Classification

Scaling-Translation-Equivariant Networks with Decomposed Convolutional Filters

no code implementations24 Sep 2019 Wei Zhu, Qiang Qiu, Robert Calderbank, Guillermo Sapiro, Xiuyuan Cheng

Encoding the scale information explicitly into the representation learned by a convolutional neural network (CNN) is beneficial for many computer vision tasks especially when dealing with multiscale inputs.

Image Classification Translation

Single-shot 3D shape reconstruction using deep convolutional neural networks

no code implementations17 Sep 2019 Hieu Nguyen, Hui Li, Qiang Qiu, Yuzeng Wang, Zhao-Yang Wang

A robust single-shot 3D shape reconstruction technique integrating the fringe projection profilometry (FPP) technique with the deep convolutional neural networks (CNNs) is proposed in this letter.

3D Shape Reconstruction

Learning data-derived privacy preserving representations from information metrics

no code implementations ICLR 2019 Martin Bertran, Natalia Martinez, Afroditi Papadaki, Qiang Qiu, Miguel Rodrigues, Guillermo Sapiro

We study space-preserving transformations where the utility provider can use the same algorithm on original and sanitized data, a critical and novel attribute to help service providers accommodate varying privacy requirements with a single set of utility algorithms.

Attribute Face Recognition +1

In Defense of Single-column Networks for Crowd Counting

no code implementations18 Aug 2018 Ze Wang, Zehao Xiao, Kai Xie, Qiang Qiu, Xian-Tong Zhen, Xian-Bin Cao

Crowd counting usually addressed by density estimation becomes an increasingly important topic in computer vision due to its widespread applications in video surveillance, urban planning, and intelligence gathering.

Crowd Counting Data Augmentation +1

LaneNet: Real-Time Lane Detection Networks for Autonomous Driving

2 code implementations4 Jul 2018 Ze Wang, Weiqiang Ren, Qiang Qiu

Lane detection is to detect lanes on the road and provide the accurate location and shape of each lane.

Autonomous Driving Edge Classification +1

Stop memorizing: A data-dependent regularization framework for intrinsic pattern learning

no code implementations ICLR 2019 Wei Zhu, Qiang Qiu, Bao Wang, Jianfeng Lu, Guillermo Sapiro, Ingrid Daubechies

Deep neural networks (DNNs) typically have enough capacity to fit random data by brute force even when conventional data-dependent regularizations focusing on the geometry of the features are imposed.

Learning to Collaborate for User-Controlled Privacy

no code implementations18 May 2018 Martin Bertran, Natalia Martinez, Afroditi Papadaki, Qiang Qiu, Miguel Rodrigues, Guillermo Sapiro

As such, users and utility providers should collaborate in data privacy, a paradigm that has not yet been developed in the privacy research community.

RotDCF: Decomposition of Convolutional Filters for Rotation-Equivariant Deep Networks

no code implementations ICLR 2019 Xiuyuan Cheng, Qiang Qiu, Robert Calderbank, Guillermo Sapiro

Explicit encoding of group actions in deep features makes it possible for convolutional neural networks (CNNs) to handle global deformations of images, which is critical to success in many vision tasks.

Liveness Detection Using Implicit 3D Features

no code implementations18 Apr 2018 J. Matias Di Martino, Qiang Qiu, Trishul Nagenalli, Guillermo Sapiro

Spoofing attacks are a threat to modern face recognition systems.

3D Reconstruction Face Recognition

Virtual CNN Branching: Efficient Feature Ensemble for Person Re-Identification

1 code implementation15 Mar 2018 Albert Gong, Qiang Qiu, Guillermo Sapiro

In this paper we introduce an ensemble method for convolutional neural network (CNN), called "virtual branching," which can be implemented with nearly no additional parameters and computation on top of standard CNNs.

Person Re-Identification

DCFNet: Deep Neural Network with Decomposed Convolutional Filters

1 code implementation ICML 2018 Qiang Qiu, Xiuyuan Cheng, Robert Calderbank, Guillermo Sapiro

In this paper, we suggest to decompose convolutional filters in CNN as a truncated expansion with pre-fixed bases, namely the Decomposed Convolutional Filters network (DCFNet), where the expansion coefficients remain learned from data.

General Classification Image Classification

OLÉ: Orthogonal Low-rank Embedding, A Plug and Play Geometric Loss for Deep Learning

1 code implementation5 Dec 2017 José Lezama, Qiang Qiu, Pablo Musé, Guillermo Sapiro

Deep neural networks trained using a softmax layer at the top and the cross-entropy loss are ubiquitous tools for image classification.

General Classification Metric Learning +2

ForestHash: Semantic Hashing With Shallow Random Forests and Tiny Convolutional Networks

no code implementations ECCV 2018 Qiang Qiu, Jose Lezama, Alex Bronstein, Guillermo Sapiro

In this paper, we introduce a random forest semantic hashing scheme that embeds tiny convolutional neural networks (CNN) into shallow random forests, with near-optimal information-theoretic code aggregation among trees.

General Classification Image Classification +2

LDMNet: Low Dimensional Manifold Regularized Neural Networks

no code implementations CVPR 2018 Wei Zhu, Qiang Qiu, Jiaji Huang, Robert Calderbank, Guillermo Sapiro, Ingrid Daubechies

To resolve this, we propose a new framework, the Low-Dimensional-Manifold-regularized neural Network (LDMNet), which incorporates a feature regularization method that focuses on the geometry of both the input data and the output features.

Face Recognition Small Data Image Classification

Learning to Succeed while Teaching to Fail: Privacy in Closed Machine Learning Systems

no code implementations23 May 2017 Jure Sokolic, Qiang Qiu, Miguel R. D. Rodrigues, Guillermo Sapiro

Confronted with this challenge, in this paper we open a new line of research, where the security, privacy, and fairness is learned and used in a closed environment.

BIG-bench Machine Learning Blocking +1

Oriented Response Networks

1 code implementation CVPR 2017 Yanzhao Zhou, Qixiang Ye, Qiang Qiu, Jianbin Jiao

DCNNs using ARFs, referred to as Oriented Response Networks (ORNs), can produce within-class rotation-invariant deep features while maintaining inter-class discrimination for classification tasks.

Ranked #83 on Image Classification on CIFAR-100 (using extra training data)

General Classification Image Classification

Self-learning Scene-specific Pedestrian Detectors using a Progressive Latent Model

no code implementations CVPR 2017 Qixiang Ye, Tianliang Zhang, Qiang Qiu, Baochang Zhang, Jie Chen, Guillermo Sapiro

In this paper, a self-learning approach is proposed towards solving scene-specific pedestrian detection problem without any human' annotation involved.

Object Object Discovery +5

Not Afraid of the Dark: NIR-VIS Face Recognition via Cross-spectral Hallucination and Low-rank Embedding

no code implementations CVPR 2017 Jose Lezama, Qiang Qiu, Guillermo Sapiro

We observe that it is often equally effective to perform hallucination to input NIR images or low-rank embedding to output deep features for a VIS deep model for cross-spectral recognition.

Face Recognition Hallucination

GraphConnect: A Regularization Framework for Neural Networks

no code implementations21 Dec 2015 Jiaji Huang, Qiang Qiu, Robert Calderbank, Guillermo Sapiro

The new method encourages the relationships between the learned decisions to resemble a graph representing the manifold structure.

Discriminative Robust Transformation Learning

no code implementations NeurIPS 2015 Jiaji Huang, Qiang Qiu, Guillermo Sapiro, Robert Calderbank

This paper proposes a framework for learning features that are robust to data variation, which is particularly important when only a limited number of trainingsamples are available.

Geometry-aware Deep Transform

no code implementations ICCV 2015 Jiaji Huang, Qiang Qiu, Robert Calderbank, Guillermo Sapiro

Many recent efforts have been devoted to designing sophisticated deep learning structures, obtaining revolutionary results on benchmark datasets.

Metric Learning

The Role of Principal Angles in Subspace Classification

no code implementations15 Jul 2015 Jiaji Huang, Qiang Qiu, Robert Calderbank

Subspace models play an important role in a wide range of signal processing tasks, and this paper explores how the pairwise geometry of subspaces influences the probability of misclassification.

Classification General Classification

Random Forests Can Hash

no code implementations16 Dec 2014 Qiang Qiu, Guillermo Sapiro, Alex Bronstein

Traditional random forest fails to enforce the consistency of hashes generated from each tree for the same class data, i. e., to preserve the underlying similarity, and it also lacks a principled way for code aggregation across trees.

Retrieval

Learning Transformations for Classification Forests

no code implementations19 Dec 2013 Qiang Qiu, Guillermo Sapiro

This work introduces a transformation-based learner model for classification forests.

Classification General Classification

Learning Transformations for Clustering and Classification

no code implementations9 Sep 2013 Qiang Qiu, Guillermo Sapiro

A low-rank transformation learning framework for subspace clustering and classification is here proposed.

Classification Clustering +1

Domain-invariant Face Recognition using Learned Low-rank Transformation

no code implementations1 Aug 2013 Qiang Qiu, Guillermo Sapiro, Ching-Hui Chen

We present a low-rank transformation approach to compensate for face variations due to changes in visual domains, such as pose and illumination.

Face Recognition Object Recognition

Compositional Dictionaries for Domain Adaptive Face Recognition

no code implementations1 Aug 2013 Qiang Qiu, Rama Chellappa

This approach has three advantages: first, the extracted sparse representation for a subject is consistent across domains and enables pose and illumination insensitive face recognition.

Dictionary Learning Face Recognition

Learning Robust Subspace Clustering

no code implementations1 Aug 2013 Qiang Qiu, Guillermo Sapiro

This proposed learned robust subspace clustering framework significantly enhances the performance of existing subspace clustering methods.

Clustering

Sparse Dictionary-based Attributes for Action Recognition and Summarization

no code implementations1 Aug 2013 Qiang Qiu, Zhuolin Jiang, Rama Chellappa

We unify the class distribution and appearance information into an objective function for learning a sparse dictionary of action attributes.

Action Recognition Dictionary Learning +1

Subspace Interpolation via Dictionary Learning for Unsupervised Domain Adaptation

no code implementations CVPR 2013 Jie Ni, Qiang Qiu, Rama Chellappa

Domain adaptation addresses the problem where data instances of a source domain have different distributions from that of a target domain, which occurs frequently in many real life scenarios.

Dictionary Learning Face Recognition +2

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