Tianfu (Matt) Wu
Computer Vision, Deep Learning, Trustworthy AI, Statistical Learning and Inference
Works (51)
Updated: April 16th, 2024 05:02
2024 article
Sign-Based Gradient Descent With Heterogeneous Data: Convergence and Byzantine Resilience
Jin, R., Liu, Y., Huang, Y., He, X., Wu, T., & Dai, H. (2024, January 12). IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, Vol. 1.
author keywords: Byzantine resilience; communication efficiency; data heterogeneity; federated learning (FL); sign-based gradient descent
TL;DR:
A novel magnitude-driven stochastic-sign-based gradient compressor is proposed to address the non-convergence issue of signSGD, and the convergence of the proposed method is established in the presence of arbitrary data heterogeneity.
(via Semantic Scholar)
arxiv.org (repository)
Sources: Web Of Science, ORCID
Added: March 11, 2024
2023 article
CGBA: Curvature-aware Geometric Black-box Attack
2023 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION, ICCV, pp. 124–133.
UN Sustainable Development Goal Categories
16. Peace, Justice and Strong Institutions
(OpenAlex)
Source: Web Of Science
Added: April 15, 2024
2023 conference paper
CGBA: Curvature-aware Geometric Black-box Attack
Proceedings of the IEEE/CVF International Conference on Computer Vision, 124–133.
Source: ORCID
Added: October 11, 2023
2023 journal article
Holistically-Attracted Wireframe Parsing: From Supervised to Self-Supervised Learning
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 45(12), 14727–14744.
author keywords: Wireframe parsing; line segment detection; holistic attraction field representation; self-supervised learning
Sources: ORCID, Web Of Science
Added: October 11, 2023
2023 article
LEARNING SPATIALLY-ADAPTIVE SQUEEZE-EXCITATION NETWORKS FOR FEW SHOT IMAGE SYNTHESIS
2023 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, ICIP, pp. 2855–2859.
author keywords: data-specificity; spatially-adaptive; attention; low-shot
TL;DR:
A spatially-adaptive squeeze-excitation module for image synthesis task that is tested in low-shot image generative learning task, and shows better performance than prior arts.
(via Semantic Scholar)
Sources: Web Of Science, ORCID
Added: March 11, 2024
2023 article
LEARNING SPATIALLY-ADAPTIVE STYLE-MODULATION NETWORKS FOR SINGLE IMAGE SYNTHESIS
2023 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, ICIP, pp. 1455–1459.
author keywords: single image; generative learning; spatially-adaptive
TL;DR:
A spatially adaptive style-modulation (SASM) module that learns to preserve realistic spatial configuration of images and learns to generate samples with better fidelity than prior works is proposed.
(via Semantic Scholar)
Sources: Web Of Science, ORCID
Added: March 11, 2024
2023 article
Level-S<SUP>2</SUP>fM: Structure from Motion on Neural Level Set of Implicit Surfaces
2023 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), pp. 17205–17214.
Sources: ORCID, Web Of Science
Added: August 30, 2023
2023 conference paper
Monocular 3D Object Detection with Bounding Box Denoising in 3D by Perceiver
Proceedings of the IEEE/CVF International Conference on Computer Vision, 6436–6446.
Source: ORCID
Added: October 11, 2023
2023 journal article
NOPE-SAC: Neural One-Plane RANSAC for Sparse-View Planar 3D Reconstruction
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 45(12), 15233–15248.
author keywords: Deep learning; planar 3 d reconstruction; sparse-view 3 d reconstruction; two-view camera pose estimation
TL;DR:
A novel Neural One-PlanE RANSAC framework (termed NOPE-SAC) that exerts excellent capability of neural networks to learn one-plane pose hypotheses from 3D plane correspondences that significantly improves the camera pose estimation for the two-view inputs with severe viewpoint changes.
(via Semantic Scholar)
Sources: ORCID, Web Of Science
Added: October 11, 2023
2023 conference paper
PaCa-ViT: Learning Patch-to-Cluster Attention in Vision Transformers
2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).
TL;DR:
The proposed PaCa module is used in designing efficient and interpretable ViT backbones and semantic segmentation head networks and is significantly more efficient than PVT models in MS-COCO and MIT-ADE20k due to the linear complexity.
(via Semantic Scholar)
Source: ORCID
Added: August 30, 2023
2023 journal article
QuadAttacK: A Quadratic Programming Approach to Learning Ordered Top-K Adversarial Attacks
Advances in Neural Information Processing Systems.
Source: ORCID
Added: October 11, 2023
2023 article
Thermal Estimation for 3D-ICs through Generative Networks
2023 IEEE INTERNATIONAL 3D SYSTEMS INTEGRATION CONFERENCE, 3DIC.
author keywords: 3DIC; thermal; generative; GAN; hybrid-bonding
TL;DR:
This paper presents a generative approach for modeling the power to heat dissipation for a 3DIC and shows that, given the power map, the model can generate the resultant heat for the bulk, opening the door for thermally aware floorplanning.
(via Semantic Scholar)
Sources: Web Of Science, ORCID
Added: August 14, 2023
2023 article
Volumetric Wireframe Parsing from Neural Attraction Fields
Source: ORCID
Added: August 30, 2023
2022 article
HoW-3D: Holistic 3DWireframe Perception from a Single Image
2022 INTERNATIONAL CONFERENCE ON 3D VISION, 3DV, pp. 596–605.
TL;DR:
This paper presents a novel Deep Spatial Gestalt (DSG) model, which outperforms the previous wire-frame detectors in detecting the invisible line geometry in single-view images and is even very competitive with prior arts that take high-fidelity PointCloud as inputs on reconstructing 3D wireframes.
(via Semantic Scholar)
Sources: Web Of Science, ORCID
Added: June 5, 2023
2022 journal article
Learning Layout and Style Reconfigurable GANs for Controllable Image Synthesis
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 44(9), 5070–5087.
author keywords: Layout; Image synthesis; Generators; Snow; Task analysis; Training; Fasteners; Image synthesis; layout-to-image; layout-to-mask-to-image; deep generative learning; GAN; ISLA-norm
MeSH headings : Algorithms; Deep Learning; Image Processing, Computer-Assisted / methods
TL;DR:
An intuitive paradigm for the task, layout-to-mask- to-image, which learns to unfold object masks in a weakly-supervised way based on an input layout and object style codes is proposed and a method built on Generative Adversarial Networks (GANs) is presented.
(via Semantic Scholar)
UN Sustainable Development Goal Categories
3. Good Health and Well-being
(Web of Science)
Sources: Web Of Science, ORCID
Added: August 22, 2022
2022 article
Learning Local-Global Contextual Adaptation for Multi-Person Pose Estimation
2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), pp. 13055–13064.
TL;DR:
This paper proposes a multi-person pose estimation approach, dubbed as LOGO-CAP, by learning the LOcal-GlObal Contextual Adaptation for human Pose, which is end-to-end trainable with near real-time inference speed in a single forward pass, obtaining state-of-the-art performance on the COCO keypoint benchmark for bottom-up human pose estimation.
(via Semantic Scholar)
UN Sustainable Development Goal Categories
3. Good Health and Well-being
(Web of Science)
Sources: Web Of Science, ORCID
Added: December 19, 2022
2022 conference paper
Learning auxiliary monocular contexts helps monocular 3D object detection
Proceedings of the AAAI Conference on Artificial Intelligence, 36(2), 1810–1818.
Source: ORCID
Added: October 11, 2023
2022 article
Modeling of Adaptive Receiver Performance Using Generative Adversarial Networks
IEEE 72ND ELECTRONIC COMPONENTS AND TECHNOLOGY CONFERENCE (ECTC 2022), pp. 1958–1963.
author keywords: SerDes; receiver; behavior modeling; adaptive; generative; GAN; DFE; IBIS-AMI
TL;DR:
A data-driven approach to modeling a high-speed serializer/deserializer (SerDes) receiver through generative adversarial networks (GANs) through the use of a discriminator structure that improves the training to generate a contour plot that makes it difficult to distinguish the ground truth.
(via Semantic Scholar)
UN Sustainable Development Goal Categories
3. Good Health and Well-being
(Web of Science)
10. Reduced Inequalities
(OpenAlex)
Sources: Web Of Science, ORCID
Added: September 19, 2022
2022 journal article
Preliminary Evaluation of a System with On-Body and Aerial Sensors for Monitoring Working Dogs
SENSORS, 22(19).
author keywords: ECG; heart rate variability; electrodes; machine learning; 3D printing; wearable; drone; UAV; embedded system
MeSH headings : Algorithms; Animals; Dogs; Electrocardiography; Heart Rate; Monitoring, Physiologic; Working Dogs
TL;DR:
An initial effort towards deployment of on-body and aerial sensors to monitor the working dogs and their environments during scent detection and search and rescue tasks in order to ensure their welfare, enable novel dog-machine interfaces, and allow for higher success rate of remote and automated task performance is presented.
(via Semantic Scholar)
UN Sustainable Development Goal Categories
3. Good Health and Well-being
(Web of Science)
Sources: Web Of Science, ORCID
Added: October 24, 2022
2022 article
REFINING SELF-SUPERVISED LEARNING IN IMAGING: BEYOND LINEAR METRIC
2022 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, ICIP, pp. 76–80.
author keywords: Self-Supervised learning; Contrastive Learning; Jaccard Index; Non-linearity
TL;DR:
A new statistical perspective is introduced, exploiting the Jaccard similarity metric, as a measure-based metric to effectively invoke non-linear features in the loss of self-supervised contrastive learning.
(via Semantic Scholar)
UN Sustainable Development Goal Categories
3. Good Health and Well-being
(Web of Science)
Sources: Web Of Science, ORCID
Added: October 23, 2023
2022 journal article
Revisiting Non-Parametric Matching Cost Volumes for Robust and Generalizable Stereo Matching
Advances in Neural Information Processing Systems, 35, 16305–16318.
Source: ORCID
Added: October 11, 2023
2022 article
RxGAN: Modeling High-Speed Receiver through Generative Adversarial Networks
MLCAD '22: PROCEEDINGS OF THE 2022 ACM/IEEE 4TH WORKSHOP ON MACHINE LEARNING FOR CAD (MLCAD), pp. 167–172.
Contributors: P. Kashyap n
author keywords: SerDes; receiver; behavior modeling; adaptive; generative; measurement; GAN; DFE; IBIS-AMI
TL;DR:
This work proposes a data-driven approach using generative adversarial training to model a real-world receiver with varying DFE and CTLE configurations while handling different channel conditions and bitstreams.
(via Semantic Scholar)
UN Sustainable Development Goal Categories
7. Affordable and Clean Energy
(OpenAlex)
9. Industry, Innovation and Infrastructure
(Web of Science)
Sources: Web Of Science, ORCID
Added: October 31, 2022
2021 journal article
A Bottom-Up and Top-Down Integration Framework for Online Object Tracking
IEEE TRANSACTIONS ON MULTIMEDIA, 23, 105–119.
author keywords: Target tracking; Correlation; Encoding; Object tracking; Benchmark testing; Visualization; Online object tracking; bottom-up and top-down; graph regularized sparse coding; alternating direction method of multipliers
TL;DR:
A graph regularized sparse coding scheme as the long-term memory based tracker and the outputs from the top-down sparse coding are potentially useful for downstream tasks such as action recognition, multiple-object tracking, and object re-identification.
(via Semantic Scholar)
UN Sustainable Development Goal Categories
10. Reduced Inequalities
(OpenAlex)
11. Sustainable Cities and Communities
(Web of Science)
Sources: Web Of Science, ORCID
Added: January 19, 2021
2021 journal article
Event driven sensor fusion
SIGNAL PROCESSING, 188.
author keywords: Sensor fusion; Multi-modal fusion; Event driven classification
TL;DR:
This paper addresses the issue of coping with damaged sensors when using the model, by learning a hidden space between sensor modalities which can be exploited to safeguard detection performance.
(via Semantic Scholar)
UN Sustainable Development Goal Categories
16. Peace, Justice and Strong Institutions
(OpenAlex)
Sources: Web Of Science, ORCID
Added: August 23, 2021
2021 article
High Speed Receiver Modeling Using Generative Adversarial Networks
IEEE 30TH CONFERENCE ON ELECTRICAL PERFORMANCE OF ELECTRONIC PACKAGING AND SYSTEMS (EPEPS 2021).
author keywords: eye diagram; IBIS-AMI; generative model; generative adversarial network; GAN; receiver
TL;DR:
The model is not built with domain knowledge but learned from a wide range of channel conditions and input bitstreams to generate an eye diagram, and a neural network model is developed to evaluate the generated eye diagram's relevant characteristics, such as eye height and width.
(via Semantic Scholar)
UN Sustainable Development Goal Categories
11. Sustainable Cities and Communities
(Web of Science)
Sources: Web Of Science, ORCID
Added: March 21, 2022
2021 journal article
Learning Regional Attraction for Line Segment Detection
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 43(6), 1998–2013.
author keywords: Image segmentation; Image edge detection; Detectors; Lattices; Machine learning; Electronic mail; Training; Line segment detection; low-level vision; deep learning
TL;DR:
The problem of line segment detection (LSD) as a problem of region coloring is posed, and an end-to-end framework to learn attraction field maps for raw input images, followed by a squeeze module to detect line segments is developed.
(via Semantic Scholar)
UN Sustainable Development Goal Categories
11. Sustainable Cities and Communities
(Web of Science)
Sources: Web Of Science, ORCID
Added: June 10, 2021
2021 article
Local Clustering with Mean Teacher for Semi-supervised learning
2020 25TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR), pp. 6243–6250.
TL;DR:
This work proposes a simple yet effective method called Local Clustering (LC) to mitigate the effect of confirmation bias in the Mean Teacher model and demonstrates on semi-supervised benchmark datasets SVHN and CIFAR-10 that adding the LC loss to MT yields significant improvements compared to MT and performance comparable to the state of the art in semi- supervised learning.
(via Semantic Scholar)
UN Sustainable Development Goal Categories
3. Good Health and Well-being
(Web of Science)
4. Quality Education
(OpenAlex)
Sources: Web Of Science, ORCID
Added: August 30, 2021
2021 article
PlaneTR: Structure-Guided Transformers for 3D Plane Recovery
2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2021), pp. 4166–4175.
TL;DR:
A neural network built upon Transformers, namely PlaneTR, to simultaneously detect and reconstruct planes from a single image and achieves state-of-the-art performance on the ScanNet and NYUv2 datasets.
(via Semantic Scholar)
UN Sustainable Development Goal Categories
11. Sustainable Cities and Communities
(Web of Science; OpenAlex)
Sources: Web Of Science, ORCID
Added: July 11, 2022
2021 article
Quantitative Study on Error Sensitivity in Ultrasound Probe Calibration with Hybrid Tracking
INTERNATIONAL ULTRASONICS SYMPOSIUM (IEEE IUS 2021).
author keywords: 3D freehand ultrasound; spatial calibration; error sensitivity
TL;DR:
A linear relationship was found between the perturbation in imaging plane motion estimation and the error caused in the calibration solution, which showed a good error tolerance for the hybrid tracking enabled US probe calibration.
(via Semantic Scholar)
Sources: Web Of Science, ORCID
Added: August 15, 2022
2020 article
Holistically-Attracted Wireframe Parsing
2020 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), pp. 2785–2794.
TL;DR:
This paper presents a fast and parsimonious parsing method to accurately and robustly detect a vectorized wireframe in an input image with a single forward pass, and is thus called Holistically-Attracted Wireframe Parser (HAWP).
(via Semantic Scholar)
UN Sustainable Development Goal Categories
11. Sustainable Cities and Communities
(Web of Science; OpenAlex)
Sources: Web Of Science, ORCID
Added: March 15, 2021
2019 article
AOGNets: Compositional Grammatical Architectures for Deep Learning
2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2019), pp. 6213–6223.
TL;DR:
Deep compositional grammatical architectures which harness the best of two worlds: grammar models and DNNs are presented which integrate compositionality and reconfigurability of the former and the capability of learning rich features of the latter in a principled way.
(via Semantic Scholar)
UN Sustainable Development Goal Categories
3. Good Health and Well-being
(Web of Science)
4. Quality Education
(OpenAlex)
Sources: Web Of Science, ORCID
Added: July 13, 2020
2019 article
Image Synthesis From Reconfigurable Layout and Style
2019 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2019), pp. 10530–10539.
TL;DR:
This paper presents a layout- and style-based architecture for generative adversarial networks (termed LostGANs) that can be trained end-to-end to generate images from reconfigurable layout and style.
(via Semantic Scholar)
UN Sustainable Development Goal Categories
3. Good Health and Well-being
(Web of Science)
11. Sustainable Cities and Communities
(OpenAlex)
Sources: Web Of Science, ORCID
Added: August 10, 2020
2019 journal article
Jointly social grouping and identification in visual dynamics with causality-induced hierarchical Bayesian model
JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION, 59, 62–75.
author keywords: Group activity detection and recognition; Casual context; Granger casual topic model
TL;DR:
A causality-induced hierarchical Bayesian model to tackle the interaction activity video, referring to the “what” interaction activities happen, “where’ interaction atomic occurs in spatial, and “when” group interaction happens in temporal is proposed.
(via Semantic Scholar)
UN Sustainable Development Goal Categories
3. Good Health and Well-being
(Web of Science)
10. Reduced Inequalities
(OpenAlex)
Sources: Web Of Science, ORCID
Added: April 22, 2019
2019 article
Learning Attraction Field Representation for Robust Line Segment Detection
2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2019), pp. 1595–1603.
TL;DR:
A region-partition based attraction field dual representation for line segment maps, which poses the problem of line segment detection (LSD) as the region coloring problem and harnesses the best practices developed in ConvNets based semantic segmentation methods such as the encoder-decoder architecture and the a-trous convolution.
(via Semantic Scholar)
UN Sustainable Development Goal Categories
11. Sustainable Cities and Communities
(Web of Science)
Sources: Web Of Science, ORCID
Added: July 13, 2020
2019 article
Neural Abstract Style Transfer for Chinese Traditional Painting
COMPUTER VISION - ACCV 2018, PT II, Vol. 11362, pp. 212–227.
author keywords: Neural style transfer; Chinese traditional painting
TL;DR:
A Neural Abstract Style Transfer method for Chinese traditional painting that learns to preserve abstraction and other style jointly end-to-end via a novel MXDoG-guided filter and three fully differentiable loss terms.
(via Semantic Scholar)
Sources: Web Of Science, ORCID
Added: November 11, 2019
2019 article
Towards Interpretable Object Detection by Unfolding Latent Structures
2019 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2019), pp. 6032–6042.
TL;DR:
The proposed method focuses on weakly-supervised extractive rationale generation, that is learning to unfold latent discriminative part configurations of object instances automatically and simultaneously in detection without using any supervision for part configurations.
(via Semantic Scholar)
UN Sustainable Development Goal Categories
3. Good Health and Well-being
(Web of Science)
10. Reduced Inequalities
(OpenAlex)
Sources: Web Of Science, ORCID
Added: August 10, 2020
2018 journal article
Vision-based integrated mobile robotic system for real-time applications in construction
Automation in Construction, 96, 470–482.
author keywords: Real-time integrated robotics; Machine vision; SLAM; Context awareness
TL;DR:
This paper proposes a mobile robotic platform that incorporates a stack of embedded platforms with integrated Graphical Processing Units (GPUs) to demonstrate the robustness and feasibility of developing and deploying an autonomous system in the near future.
(via Semantic Scholar)
UN Sustainable Development Goal Categories
9. Industry, Innovation and Infrastructure
(Web of Science; OpenAlex)
Sources: Crossref, ORCID
Added: December 16, 2018
2017 journal article
Online Object Tracking, Learning and Parsing with And-Or Graphs
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 39(12), 2465–2480.
author keywords: Visual tracking; and-or graph; latent SVM; dynamic programming; intrackability
TL;DR:
Three key issues in online inference and learning are addressed: (i) maintaining purity of positive and negative examples collected online, (ii) controling model complexity in latent structure learning, and (iii) identifying critical moments to re-learn the structure of AOG based on its intrackability.
(via Semantic Scholar)
UN Sustainable Development Goal Categories
10. Reduced Inequalities
(OpenAlex)
11. Sustainable Cities and Communities
(Web of Science)
Sources: Web Of Science, ORCID
Added: August 6, 2018
2017 article
Zero-Shot Learning posed as a Missing Data Problem
2017 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION WORKSHOPS (ICCVW 2017), pp. 2616–2622.
TL;DR:
This paper presents a method of zero-shot learning (ZSL) which poses ZSL as the missing data problem, rather than the missing label problem, and explores a simple yet effective transductive framework in the reverse way.
(via Semantic Scholar)
UN Sustainable Development Goal Categories
3. Good Health and Well-being
(Web of Science)
Sources: Web Of Science, ORCID
Added: August 6, 2018
2016 journal article
A Reconfigurable Tangram Model for Scene Representation and Categorization
IEEE Transactions on Image Processing, 25(1), 150–166.
author keywords: Tangram model; scene layout; and-or graph; dynamic programming; scene categorization
TL;DR:
This paper presents a hierarchical and compositional scene layout (i.e., spatial configuration) representation and a method of learning reconfigurable model for scene categorization, which outperform the spatial pyramid model consistently.
(via Semantic Scholar)
Sources: Crossref, ORCID
Added: July 20, 2019
2016 article
Face Detection with End-to-End Integration of a ConvNet and a 3D Model
COMPUTER VISION - ECCV 2016, PT III, Vol. 9907, pp. 420–436.
author keywords: Face detection; Face 3D model; ConvNet; Deep learning; Multi-task learning
TL;DR:
This paper presents a method for face detection in the wild, which integrates a ConvNet and a 3D mean face model in an end-to-end multi-task discriminative learning framework and addresses two issues in adapting state-of-the-art generic object detection ConvNets.
(via Semantic Scholar)
UN Sustainable Development Goal Categories
3. Good Health and Well-being
(Web of Science)
10. Reduced Inequalities
(OpenAlex)
Sources: Web Of Science, ORCID
Added: August 6, 2018
2016 journal article
Learning And-Or Model to Represent Context and Occlusion for Car Detection and Viewpoint Estimation
IEEE Transactions on Pattern Analysis and Machine Intelligence, 38(9), 1829–1843.
author keywords: Car detection; car viewpoint estimation; and-or graph; hierarchical model; context; occlusion modeling
TL;DR:
This paper presents a method for learning an And-Or model to represent context and occlusion for car detection and viewpoint estimation and achieves significant improvement consistently on the four detection datasets, and comparable performance on car viewpoint estimation.
(via Semantic Scholar)
UN Sustainable Development Goal Categories
11. Sustainable Cities and Communities
(OpenAlex)
Sources: Crossref, ORCID
Added: July 20, 2019
2015 journal article
Learning Near-Optimal Cost-Sensitive Decision Policy for Object Detection
IEEE Transactions on Pattern Analysis and Machine Intelligence, 37(5), 1013–1027.
author keywords: Decision policy; cost-sensitive computing; risk minimization; dynamic programming; object detection
TL;DR:
This paper formulate an empirical risk function as the weighted sum of the cost of computation and the loss of false alarm and missing detection and finds that the upper bound is very tight empirically and thus the resulting policy is said to be near-optimal.
(via Semantic Scholar)
UN Sustainable Development Goal Categories
16. Peace, Justice and Strong Institutions
(OpenAlex)
Sources: Crossref, ORCID
Added: July 20, 2019
2014 journal article
A global energy optimization framework for 2.1D sketch extraction from monocular images
Graphical Models, 76(5), 507–521.
author keywords: 2.1D sketch; Global optimization; Local features; Hybrid differential evolution
TL;DR:
A global optimization framework for inferring the 2.1D sketch from a monocular image using a global energy optimization framework and a hybrid evolution algorithm is utilized to minimize the global energy function efficiently.
(via Semantic Scholar)
UN Sustainable Development Goal Categories
7. Affordable and Clean Energy
(OpenAlex)
Sources: Crossref, ORCID
Added: July 20, 2019
2014 journal article
Coupling-and-decoupling: A hierarchical model for occlusion-free object detection
Pattern Recognition, 47(10), 3254–3264.
author keywords: Occlusion modeling; Object detection; And-Or graph; Deformable part-based model; Latent structural SVM
TL;DR:
The proposed method of learning a hierarchical model for X-to-X occlusion-free object detection is compared with state-of-the-art deformable part-based methods, and obtains comparable or better detection performance.
(via Semantic Scholar)
Sources: Crossref, ORCID
Added: July 20, 2019
2014 chapter
Integrating Context and Occlusion for Car Detection by Hierarchical And-Or Model
In D. Fleet, T. Pajdla, B. Schiele, & T. Tuytelaars (Eds.), Computer Vision – ECCV 2014 (pp. 652–667).
Ed(s): D. Fleet, T. Pajdla, B. Schiele & T. Tuytelaars
TL;DR:
A method of learning reconfigurable hierarchical And-Or models to integrate context and occlusion for car detection using Weak-Label Structural SVM and compares with state-of-the-art variants of deformable part-based models and other methods.
(via Semantic Scholar)
UN Sustainable Development Goal Categories
11. Sustainable Cities and Communities
(OpenAlex)
Sources: Crossref, ORCID
Added: August 13, 2021
2014 journal article
Learning mixtures of Bernoulli templates by two-round EM with performance guarantee
Electronic Journal of Statistics, 8(2), 3004–3030.
author keywords: Clustering; performance bounds; unsupervised learning
TL;DR:
It is shown that the two-round EM algorithm can learn mixture of Bernoulli templates with near optimal precision with high probability, if the Bernouelli templates are sufficiently different and if the number of features is sufficiently high.
(via Semantic Scholar)
Sources: Crossref, ORCID
Added: July 20, 2019
2013 chapter
Coupling-and-Decoupling: A Hierarchical Model for Occlusion-Free Car Detection
In K. M. Lee, Y. Matsushita, J. M. Rehg, & Z. Hu (Eds.), Computer Vision – ACCV 2012 (pp. 164–175).
Ed(s): K. Lee, Y. Matsushita, J. Rehg & Z. Hu
TL;DR:
This paper addresses the problem of X-to-X-occlusion-free object detection by utilizing an intuitive coupling-and-decoupling strategy, and compares the model with the state-of-the-art deformable part-based model (DPM) and obtain better detection performance.
(via Semantic Scholar)
Sources: Crossref, ORCID
Added: August 13, 2021
2013 chapter
Tracking Pedestrian with Multi-component Online Deformable Part-Based Model
In K. M. Lee, Y. Matsushita, J. M. Rehg, & Z. Hu (Eds.), Computer Vision – ACCV 2012 (pp. 664–676).
Ed(s): K. Lee, Y. Matsushita, J. Rehg & Z. Hu
TL;DR:
A novel online algorithm to track pedestrian by integrating both the bottom-up and the top-down models of pedestrian, which has three advantages compared with other models: it can efficiently generate match penalty maps of parts preserving the 2bit binary pattern, and can search over all possible configurations in an image in linear-time by distance transforms algorithm.
(via Semantic Scholar)
UN Sustainable Development Goal Categories
11. Sustainable Cities and Communities
(OpenAlex)
Sources: Crossref, ORCID
Added: August 13, 2021
2010 journal article
A Numerical Study of the Bottom-Up and Top-Down Inference Processes in And-Or Graphs
International Journal of Computer Vision, 93(2), 226–252.
author keywords: Bottom-up/Top-down inference; a-beta-gamma process; Information contribution; Hierarchical model; And-Or graph; Object parsing
UN Sustainable Development Goal Categories
10. Reduced Inequalities
(OpenAlex)
Sources: Crossref, ORCID
Added: July 20, 2019
2009 journal article
A stochastic graph grammar for compositional object representation and recognition
Pattern Recognition, 42(7), 1297–1307.
TL;DR:
A hierarchical generative model for representing and recognizing compositional object categories with large intra-category variance that combines the power of a stochastic context free grammar (SCFG) to express the variability of part configurations, and a Markov random field (MRF) to represent the pictorial spatial relationships between these parts.
(via Semantic Scholar)
UN Sustainable Development Goal Categories
4. Quality Education
(OpenAlex)
Sources: Crossref, ORCID
Added: July 20, 2019
Employment
Updated: October 10th, 2023 20:06
2016 - present
NC State University Raleigh, NC, US
Assistant/Associate Professor
electrical and computer engineering
Education
Updated: September 8th, 2016 22:42
2008 - 2011
University of California Los Angeles Los Angeles, CA, US
Ph.D.
Statistics
Citation Index includes data from a number of different sources. If you have questions about the sources of data in the Citation Index or need a set of data which is free to re-distribute, please contact us.
Certain data included herein are derived from the Web of Science© and InCites© (2024) of Clarivate Analytics. All rights reserved. You may not copy or re-distribute this material in whole or in part without the prior written consent of Clarivate Analytics.
