Search Results for author:
Found 54 papers, 15 papers with code
RhyRNN: Rhythmic RNN for Recognizing Events in Long and Complex Videos
We study the behavior of RhyRNN and empirically show that our method works well even when mph{only event-level labels are available} in the training stage (compared to algorithms requiring sub-activity labels for recognition), and thus is more practical when the sub-activity labels are missing or difficult to obtain.
Domain Adaptation Using Pseudo Labels
In the absence of labeled target data, unsupervised domain adaptation approaches seek to align the marginal distributions of the source and target domains in order to train a classifier for the target.
Transformer-based Selective Super-Resolution for Efficient Image Refinement
Conventional super-resolution methods suffer from two drawbacks: substantial computational cost in upscaling an entire large image, and the introduction of extraneous or potentially detrimental information for downstream computer vision tasks during the refinement of the background.
Patch-based Selection and Refinement for Early Object Detection
Early object detection (OD) is a crucial task for the safety of many dynamic systems.
Ordinal Classification with Distance Regularization for Robust Brain Age Prediction
However, these methods are subject to an inherent regression to the mean effect, which causes a systematic bias resulting in an overestimation of brain age in young subjects and underestimation in old subjects.
Ranked #1 on
Ordinal Classification
on OASIS+NACC+ICBM+ABIDE+IXI
Instance Adaptive Prototypical Contrastive Embedding for Generalized Zero Shot Learning
Generalized zero-shot learning(GZSL) aims to classify samples from seen and unseen labels, assuming unseen labels are not accessible during training.
Deep-Learning-based Fast and Accurate 3D CT Deformable Image Registration in Lung Cancer
Conclusion: A patient-specific vision-transformer-based network was developed and shown to be accurate and efficient to reconstruct 3D CT images from kV images.
Brainomaly: Unsupervised Neurologic Disease Detection Utilizing Unannotated T1-weighted Brain MR Images
Harnessing the power of deep neural networks in the medical imaging domain is challenging due to the difficulties in acquiring large annotated datasets, especially for rare diseases, which involve high costs, time, and effort for annotation.
Ranked #1 on
Anomaly Detection
on ADNI
PatchRot: A Self-Supervised Technique for Training Vision Transformers
Vision transformers require a huge amount of labeled data to outperform convolutional neural networks.
HealthyGAN: Learning from Unannotated Medical Images to Detect Anomalies Associated with Human Disease
Therefore, this paper poses the research question of how to improve unsupervised anomaly detection by utilizing (1) an unannotated set of mixed images, in addition to (2) the set of healthy images as being used in the literature.
CoVaxNet: An Online-Offline Data Repository for COVID-19 Vaccine Hesitancy Research
Despite the astonishing success of COVID-19 vaccines against the virus, a substantial proportion of the population is still hesitant to be vaccinated, undermining governmental efforts to control the virus.
Optical Flow for Video Super-Resolution: A Survey
Video super-resolution is currently one of the most active research topics in computer vision as it plays an important role in many visual applications.
Deep residual inception encoder-decoder network for amyloid PET harmonization
Multiple positron emission tomography (PET) tracers are available for amyloid imaging, posing a significant challenge to consensus interpretation and quantitative analysis.
A2B-GAN: Utilizing Unannotated Anomalous Images for Anomaly Detection in Medical Image Analysis
This paper poses the research question of how to improve anomaly detection by using an unannotated set of mixed images of both normal and anomalous samples (in addition to a set of normal images from healthy subjects).
PAT: Pseudo-Adversarial Training For Detecting Adversarial Videos
Extensive research has demonstrated that deep neural networks (DNNs) are prone to adversarial attacks.
FedNS: Improving Federated Learning for collaborative image classification on mobile clients
Federated Learning (FL) is a paradigm that aims to support loosely connected clients in learning a global model collaboratively with the help of a centralized server.
Partial Domain Adaptation Using Selective Representation Learning For Class-Weight Computation
Driven by the motivation that image styles are private to each domain, in this work, we develop a method that identifies outlier classes exclusively from image content information and train a label classifier exclusively on class-content from source images.
Learning Latent Topology for Graph Matching
Graph matching (GM) has been traditionally modeled as a deterministic optimization problem characterized by an affinity matrix under pre-defined graph topology.
Evaluating a Simple Retraining Strategy as a Defense Against Adversarial Attacks
Though deep neural networks (DNNs) have shown superiority over other techniques in major fields like computer vision, natural language processing, robotics, recently, it has been proven that they are vulnerable to adversarial attacks.
AdvFoolGen: Creating Persistent Troubles for Deep Classifiers
Researches have shown that deep neural networks are vulnerable to malicious attacks, where adversarial images are created to trick a network into misclassification even if the images may give rise to totally different labels by human eyes.
Hardware Acceleration of Sparse and Irregular Tensor Computations of ML Models: A Survey and Insights
Machine learning (ML) models are widely used in many important domains.
Deep Learning of Determinantal Point Processes via Proper Spectral Sub-gradient
Determinantal point processes (DPPs) is an effective tool to deliver diversity on multiple machine learning and computer vision tasks.
VSEC-LDA: Boosting Topic Modeling with Embedded Vocabulary Selection
When applying a topic model, a relatively standard pre-processing step is to first build a vocabulary of frequent words.
Recognizing Video Events with Varying Rhythms
In this paper, we investigate the problem of recognizing long and complex events with varying action rhythms, which has not been considered in the literature but is a practical challenge.
Deep Residual Dense U-Net for Resolution Enhancement in Accelerated MRI Acquisition
Accelerated MRI scan may be achieved by acquiring less amount of k-space data (down-sampling in the k-space).
Learning deep graph matching with channel-independent embedding and Hungarian attention
Graph matching aims to establishing node-wise correspondence between two graphs, which is a classic combinatorial problem and in general NP-complete.
Ranked #15 on
Graph Matching
on PASCAL VOC
(matching accuracy metric)
Improving Batch Normalization with Skewness Reduction for Deep Neural Networks
As normalizing based on mean and variance does not necessarily make the features to have the same distribution, we propose a new normalization scheme: Batch Normalization with Skewness Reduction (BNSR).
Plan-Recognition-Driven Attention Modeling for Visual Recognition
We build an event recognition system, ER-PRN, which takes Pixel Dynamics Network as a subroutine, to recognize events based on observations augmented by plan-recognition-driven attention.
Generalizing Graph Matching beyond Quadratic Assignment Model
Graph matching has received persistent attention over decades, which can be formulated as a quadratic assignment problem (QAP).
Mean Local Group Average Precision (mLGAP): A New Performance Metric for Hashing-based Retrieval
In this paper, we introduce a new metric named Mean Local Group Average Precision (mLGAP) for better evaluation of the performance of hashing-based retrieval.
Incremental Multi-graph Matching via Diversity and Randomness based Graph Clustering
In this paper, we present an incremental multi-graph matching approach, which deals with the arriving graph utilizing the previous matching results under the global consistency constraint.
Weakly Supervised Deep Image Hashing through Tag Embeddings
We utilize the information contained in the user-generated tags associated with the images to learn the hash codes.
Training Neural Networks by Using Power Linear Units (PoLUs)
Experimental results show that our proposed activation function outperforms other state-of-the-art models with most networks.
Recognizing Plans by Learning Embeddings from Observed Action Distributions
The first involves resampling the distribution sequences to single action sequences, from which we could learn an action affinity model based on learned action (word) embeddings for plan recognition.
Joint Cuts and Matching of Partitions in One Graph
As two fundamental problems, graph cuts and graph matching have been investigated over decades, resulting in vast literature in these two topics respectively.
Capturing Localized Image Artifacts through a CNN-based Hyper-image Representation
Thus some image-based small-data applications first train their framework on a collection of patches (instead of the entire image) to better learn the representation of localized artifacts.
A Strategy for an Uncompromising Incremental Learner
Using an implementation based on deep neural networks, we demonstrate that phantom sampling dramatically avoids catastrophic forgetting.
A Computational Approach to Relative Aesthetics
To this end, we formulate a novel problem of ranking images with respect to their aesthetic quality.
A Structured Approach to Predicting Image Enhancement Parameters
This paper presents a novel approach to predicting the enhancement parameters given a new image using only its features, without using any training images.
Classification of Diabetic Retinopathy Images Using Multi-Class Multiple-Instance Learning Based on Color Correlogram Features
All people with diabetes have the risk of developing diabetic retinopathy (DR), a vision-threatening complication.
Supporting Navigation of Outdoor Shopping Complexes for Visually-impaired Users through Multi-modal Data Fusion
Outdoor shopping complexes (OSC) are extremely difficult for people with visual impairment to navigate.
Relative Learning from Web Images for Content-adaptive Enhancement
Then we carry out subjective tests, which show that users prefer images enhanced by our approach over other existing methods.
Investigating Human Factors in Image Forgery Detection
We compare the performance of an automated algorithm and humans for forgery detection problem.
Joint Regression and Ranking for Image Enhancement
Research on automated image enhancement has gained momentum in recent years, partially due to the need for easy-to-use tools for enhancing pictures captured by ubiquitous cameras on mobile devices.
Improving Vision-based Self-positioning in Intelligent Transportation Systems via Integrated Lane and Vehicle Detection
In the framework, the number of lanes, the vehicle's position in those lanes and the presence of other vehicles are considered as parameters.
Hierarchical Attention Network for Action Recognition in Videos
Understanding human actions in wild videos is an important task with a broad range of applications.
PPP: Joint Pointwise and Pairwise Image Label Prediction
However, pointwise labels in image classification and tag annotation are inherently related to the pairwise labels.
Neural Dataset Generality
It is curious that while the filters learned by these CNNs are related to the atomic structures of the images from which they are learnt, all datasets learn similar looking low-level filters.
Diving deeper into mentee networks
We studied various characteristics of such networks and found some interesting behaviors.
Simpler non-parametric methods provide as good or better results to multiple-instance learning.
Multiple-instance learning (MIL) is a unique learning problem in which training data labels are available only for collections of objects (called bags) instead of individual objects (called instances).
Simpler Non-Parametric Methods Provide as Good or Better Results to Multiple-Instance Learning
Multiple-instance learning (MIL) is a unique learning problem in which training data labels are available only for collections of objects (called bags) instead of individual objects (called instances).
Unsupervised Video Analysis Based on a Spatiotemporal Saliency Detector
Recently, the spectrum analysis based visual saliency approach has attracted a lot of interest due to its simplicity and good performance, where the phase information of the image is used to construct the saliency map.
Predicting Multiple Attributes via Relative Multi-task Learning
Relative attributes learning aims to learn ranking functions describing the relative strength of attributes.
Relative Hidden Markov Models for Evaluating Motion Skill
The proposed algorithm effectively learns a model from the training data so that the attribute under consideration is linked to the likelihood of the inputs under the learned model.
