no code implementations • ECCV 2020 • Tianshu Yu, Yikang Li, Baoxin Li
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.
no code implementations • 9 Feb 2024 • Sachin Chhabra, Hemanth Venkateswara, Baoxin Li
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.
1 code implementation • 10 Dec 2023 • Tianyi Zhang, Kishore Kasichainula, Yaoxin Zhuo, Baoxin Li, Jae-sun Seo, Yu Cao
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.
1 code implementation • 3 Nov 2023 • Tianyi Zhang, Kishore Kasichainula, Yaoxin Zhuo, Baoxin Li, Jae-sun Seo, Yu Cao
Early object detection (OD) is a crucial task for the safety of many dynamic systems.
1 code implementation • IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) 2023 • Jay Shah, Md Mahfuzur Rahman Siddiquee, Yi Su, Teresa Wu, Baoxin Li
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
no code implementations • 13 Sep 2023 • Riti Paul, Sahil Vora, Baoxin Li
Generalized zero-shot learning(GZSL) aims to classify samples from seen and unseen labels, assuming unseen labels are not accessible during training.
no code implementations • 21 Apr 2023 • Yuzhen Ding, Hongying Feng, Yunze Yang, Jason Holmes, Zhengliang Liu, David Liu, William W. Wong, Nathan Y. Yu, Terence T. Sio, Steven E. Schild, Baoxin Li, Wei Liu
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.
1 code implementation • 18 Feb 2023 • Md Mahfuzur Rahman Siddiquee, Jay Shah, Teresa Wu, Catherine Chong, Todd J. Schwedt, Gina Dumkrieger, Simona Nikolova, Baoxin Li
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
1 code implementation • 27 Oct 2022 • Sachin Chhabra, Prabal Bijoy Dutta, Hemanth Venkateswara, Baoxin Li
Vision transformers require a huge amount of labeled data to outperform convolutional neural networks.
1 code implementation • 5 Sep 2022 • Md Mahfuzur Rahman Siddiquee, Jay Shah, Teresa Wu, Catherine Chong, Todd Schwedt, Baoxin Li
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.
1 code implementation • 30 Jun 2022 • Bohan Jiang, Paras Sheth, Baoxin Li, Huan Liu
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.
no code implementations • 20 Mar 2022 • Zhigang Tu, Hongyan Li, Wei Xie, Yuanzhong Liu, Shifu Zhang, Baoxin Li, Junsong Yuan
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.
1 code implementation • Alzheimer's and Dementia 2022 • Jay Shah, Fei Gao, Baoxin Li, Valentina Ghisays, Ji Luo, Yinghua Chen, Wendy Lee, Yuxiang Zhou, Tammie L.S. Benzinger, Eric M. Reiman, Kewei Chen, Yi Su, Teresa Wu
Multiple positron emission tomography (PET) tracers are available for amyloid imaging, posing a significant challenge to consensus interpretation and quantitative analysis.
no code implementations • 29 Sep 2021 • Md Mahfuzur Rahman Siddiquee, Teresa Wu, Baoxin Li
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).
no code implementations • 13 Sep 2021 • Nupur Thakur, Baoxin Li
Extensive research has demonstrated that deep neural networks (DNNs) are prone to adversarial attacks.
no code implementations • 20 Jan 2021 • Yaoxin Zhuo, Baoxin Li
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.
no code implementations • 6 Jan 2021 • Sandipan Choudhuri, Riti Paul, Arunabha Sen, Baoxin Li, Hemanth Venkateswara
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.
no code implementations • 1 Jan 2021 • Tianshu Yu, Runzhong Wang, Junchi Yan, Baoxin Li
Graph matching (GM) has been traditionally modeled as a deterministic optimization problem characterized by an affinity matrix under pre-defined graph topology.
no code implementations • 20 Jul 2020 • Nupur Thakur, Yuzhen Ding, Baoxin Li
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.
no code implementations • 20 Jul 2020 • Yuzhen Ding, Nupur Thakur, Baoxin Li
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.
no code implementations • 2 Jul 2020 • Shail Dave, Riyadh Baghdadi, Tony Nowatzki, Sasikanth Avancha, Aviral Shrivastava, Baoxin Li
Machine learning (ML) models are widely used in many important domains.
no code implementations • ICLR 2020 • Tianshu Yu, Yikang Li, Baoxin Li
Determinantal point processes (DPPs) is an effective tool to deliver diversity on multiple machine learning and computer vision tasks.
no code implementations • 15 Jan 2020 • Yuzhen Ding, Baoxin Li
When applying a topic model, a relatively standard pre-processing step is to first build a vocabulary of frequent words.
1 code implementation • 14 Jan 2020 • Yikang Li, Tianshu Yu, Baoxin Li
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.
no code implementations • 13 Jan 2020 • Pak Lun Kevin Ding, Zhiqiang Li, Yuxiang Zhou, Baoxin Li
Accelerated MRI scan may be achieved by acquiring less amount of k-space data (down-sampling in the k-space).
no code implementations • ICLR 2020 • Tianshu Yu, Runzhong Wang, Junchi Yan, Baoxin Li
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)
no code implementations • ICLR 2020 • Pak Lun Kevin Ding, Sarah Martin, Baoxin Li
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).
no code implementations • 2 Dec 2018 • Yantian Zha, Yikang Li, Tianshu Yu, Subbarao Kambhampati, Baoxin Li
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.
no code implementations • NeurIPS 2018 • Tianshu Yu, Junchi Yan, Yilin Wang, Wei Liu, Baoxin Li
Graph matching has received persistent attention over decades, which can be formulated as a quadratic assignment problem (QAP).
no code implementations • 24 Nov 2018 • Pak Lun Kevin Ding, Yikang Li, Baoxin Li
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.
no code implementations • ECCV 2018 • Tianshu Yu, Junchi Yan, Wei Liu, Baoxin Li
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.
1 code implementation • CVPR 2019 • Vijetha Gattupalli, Yaoxin Zhuo, Baoxin Li
We utilize the information contained in the user-generated tags associated with the images to learn the hash codes.
1 code implementation • 1 Feb 2018 • Yikang Li, Pak Lun Kevin Ding, Baoxin Li
Experimental results show that our proposed activation function outperforms other state-of-the-art models with most networks.
no code implementations • 5 Dec 2017 • Yantian Zha, Yikang Li, Sriram Gopalakrishnan, Baoxin Li, Subbarao Kambhampati
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.
no code implementations • CVPR 2018 • Tianshu Yu, Junchi Yan, Jieyi Zhao, Baoxin Li
As two fundamental problems, graph cuts and graph matching have been investigated over decades, resulting in vast literature in these two topics respectively.
no code implementations • 14 Nov 2017 • Parag Shridhar Chandakkar, Baoxin Li
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.
1 code implementation • 2 May 2017 • Ragav Venkatesan, Hemanth Venkateswara, Sethuraman Panchanathan, Baoxin Li
Using an implementation based on deep neural networks, we demonstrate that phantom sampling dramatically avoids catastrophic forgetting.
no code implementations • 5 Apr 2017 • Parag S. Chandakkar, Vijetha Gattupalli, Baoxin Li
To this end, we formulate a novel problem of ranking images with respect to their aesthetic quality.
no code implementations • 5 Apr 2017 • Parag S. Chandakkar, Baoxin Li
This paper presents a novel approach to predicting the enhancement parameters given a new image using only its features, without using any training images.
no code implementations • 5 Apr 2017 • Ragav Venkatesan, Parag S. Chandakkar, Baoxin Li
All people with diabetes have the risk of developing diabetic retinopathy (DR), a vision-threatening complication.
no code implementations • 5 Apr 2017 • Archana Paladugu, Parag S. Chandakkar, Peng Zhang, Baoxin Li
Outdoor shopping complexes (OSC) are extremely difficult for people with visual impairment to navigate.
no code implementations • 5 Apr 2017 • Parag S. Chandakkar, Qiongjie Tian, Baoxin Li
Then we carry out subjective tests, which show that users prefer images enhanced by our approach over other existing methods.
no code implementations • 5 Apr 2017 • Parag S. Chandakkar, Baoxin Li
We compare the performance of an automated algorithm and humans for forgery detection problem.
no code implementations • 5 Apr 2017 • Parag S. Chandakkar, Baoxin Li
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.
no code implementations • 5 Apr 2017 • Parag S. Chandakkar, Yilin Wang, Baoxin Li
In the framework, the number of lanes, the vehicle's position in those lanes and the presence of other vehicles are considered as parameters.
no code implementations • 21 Jul 2016 • Yilin Wang, Suhang Wang, Jiliang Tang, Neil O'Hare, Yi Chang, Baoxin Li
Understanding human actions in wild videos is an important task with a broad range of applications.
no code implementations • CVPR 2016 • Yilin Wang, Suhang Wang, Jiliang Tang, Huan Liu, Baoxin Li
However, pointwise labels in image classification and tag annotation are inherently related to the pairwise labels.
1 code implementation • 14 May 2016 • Ragav Venkatesan, Vijetha Gattupalli, Baoxin Li
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.
1 code implementation • 27 Apr 2016 • Ragav Venkatesan, Baoxin Li
We studied various characteristics of such networks and found some interesting behaviors.
1 code implementation • IEEE International Conference on Computer Vision 2015 • Ragav Venkatesan, Parag Chandakkar, Baoxin Li
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).
no code implementations • ICCV 2015 • Ragav Venkatesan, Parag Chandakkar, Baoxin Li
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).
no code implementations • 24 Mar 2015 • Qiang Zhang, Yilin Wang, Baoxin Li
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.
no code implementations • CVPR 2014 • Lin Chen, Qiang Zhang, Baoxin Li
Relative attributes learning aims to learn ranking functions describing the relative strength of attributes.
no code implementations • CVPR 2013 • Qiang Zhang, Baoxin Li
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.