Filters








32 Hits in 9.8 sec

Invariant Teacher and Equivariant Student for Unsupervised 3D Human Pose Estimation [article]

Chenxin Xu, Siheng Chen, Maosen Li, Ya Zhang
2020 arXiv   pre-print
We propose a novel method based on teacher-student learning framework for 3D human pose estimation without any 3D annotation or side information.  ...  To solve this unsupervised-learning problem, the teacher network adopts pose-dictionary-based modeling for regularization to estimate a physically plausible 3D pose.  ...  The main contributions of this paper are as follow: • We propose an unsupervised teacher-student learning framework, called invariant teacher and equivariant student (ITES), to estimate the 3D human pose  ... 
arXiv:2012.09398v1 fatcat:vxyf6bjsards5c45mj74tumc3i

Unsupervised Cross-Modal Alignment for Multi-Person 3D Pose Estimation [article]

Jogendra Nath Kundu, Ambareesh Revanur, Govind Vitthal Waghmare, Rahul Mysore Venkatesh, R. Venkatesh Babu
2020 arXiv   pre-print
We present a deployment friendly, fast bottom-up framework for multi-person 3D human pose estimation.  ...  In the absence of any paired supervision, we leverage a frozen network, as a teacher model, which is trained on an auxiliary task of multi-person 2D pose estimation.  ...  (See Table 3 and Fig. 2) Conclusion In this paper we have introduced an unsupervised approach for multi-person 3D pose estimation by infusing structural constraints of human pose.  ... 
arXiv:2008.01388v1 fatcat:rlfpgoy6vjayhmwyhhvihdizlm

Unsupervised domain adaptation for clinician pose estimation and instance segmentation in the operating room [article]

Vinkle Srivastav, Afshin Gangi, Nicolas Padoy
2022 arXiv   pre-print
To address these concerns, we first study how joint person pose estimation and instance segmentation can be performed on low resolutions images with downsampling factors from 1x to 12x.  ...  Computer vision models for person pixel-based segmentation and body-keypoints detection are needed to better understand the clinical activities and the spatial layout of the OR.  ...  KM-ORPose KM-ORPose (Srivastav et al., 2020) uses the teacherstudent learning paradigm for the domain adaptation in the OR for joint person detection and 2D/3D human pose estimation.  ... 
arXiv:2108.11801v4 fatcat:4fxsoswfpvfejbcqmzvkdorpky

Small Data Challenges in Big Data Era: A Survey of Recent Progress on Unsupervised and Semi-Supervised Methods [article]

Guo-Jun Qi, Jiebo Luo
2021 arXiv   pre-print
between transformation equivariance for unsupervised learning and supervised invariance for supervised learning, and unify unsupervised pretraining and supervised finetuning.  ...  We will also provide a broader outlook of future directions to unify transformation and instance equivariances for representation learning, connect unsupervised and semi-supervised augmentations, and explore  ...  Teacher-Student Models The idea behind teacher-student models for semisupervised learning is to obtain a single or an ensemble of teachers, and use the predictions on unlabeled examples as targets to supervise  ... 
arXiv:1903.11260v2 fatcat:hjya3ojzmfh7nnldhqkdx6o37a

Non-Local Latent Relation Distillation for Self-Adaptive 3D Human Pose Estimation [article]

Jogendra Nath Kundu, Siddharth Seth, Anirudh Jamkhandi, Pradyumna YM, Varun Jampani, Anirban Chakraborty, R. Venkatesh Babu
2022 arXiv   pre-print
We evaluate different self-adaptation settings and demonstrate state-of-the-art 3D human pose estimation performance on standard benchmarks.  ...  Available 3D human pose estimation approaches leverage different forms of strong (2D/3D pose) or weak (multi-view or depth) paired supervision.  ...  Acknowledgments and Disclosure of Funding  ... 
arXiv:2204.01971v2 fatcat:fgdpgc3t4jfc3gh47idlxkhnqy

A Unified Framework for Domain Adaptive Pose Estimation [article]

Donghyun Kim, Kaihong Wang, Kate Saenko, Margrit Betke, Stan Sclaroff
2022 arXiv   pre-print
Our method outperforms existing baselines on human pose estimation by up to 4.5 percent points (pp), hand pose estimation by up to 7.4 pp, and animal pose estimation by up to 4.8 pp for dogs and 3.3 pp  ...  While several domain adaptive pose estimation models have been proposed recently, they are not generic but only focus on either human pose or animal pose estimation, and thus their effectiveness is somewhat  ...  Acknowledgments This work has been partially supported by NSF Award, DARPA, ONR MURI grant N00014-19-1-2571 associated with AUSMURIB000001 (to M.B.) and by NSF grant 1551572 (to. M.B.).  ... 
arXiv:2204.00172v2 fatcat:bz7ev3xfdvdb3nmdk4bfzuvrpm

Deep Learning for Omnidirectional Vision: A Survey and New Perspectives [article]

Hao Ai, Zidong Cao, Jinjing Zhu, Haotian Bai, Yucheng Chen, Lin Wang
2022 arXiv   pre-print
This paper presents a systematic and comprehensive review and analysis of the recent progress in DL methods for omnidirectional vision.  ...  with the 2D planar image data; (ii) A structural and hierarchical taxonomy of the DL methods for omnidirectional vision; (iii) A summarization of the latest novel learning strategies and applications;  ...  For semi-supervised learning, Tran et al. [150] exploited the 'Mean-Teacher' model [203] for 3D room layout reconstruction by learning from the labeled and unlabeled data in the same scenario.  ... 
arXiv:2205.10468v2 fatcat:73fks33oafa6zgxliccydvdbeq

Learning Neural Textual Representations for Citation Recommendation

Binh Thanh Kieu, Inigo Jauregi Unanue, Son Bao Pham, Hieu Xuan Phan, Massimo Piccardi
2021 2020 25th International Conference on Pattern Recognition (ICPR)  
Single Image 3D Human Pose Estimation DAY 2 -Jan 13, 2021 Rahmon, Gani; Bunyak, Filiz; Palaniappan, Kannappan 2237 Multi-cue and Multi-stream Encoder-Decoder Network for Robust Moving Object  ...  Incremental Learning: The Handwriting Recognition Use Case DAY 2 -Jan 13, 2021 Xiao, Ruixin; Liu, Zhilei; Wu, Baoyuan 2272 Teacher-Student Competition for Unsupervised Domain Adaption DAY 2 -  ... 
doi:10.1109/icpr48806.2021.9412725 fatcat:3vge2tpd2zf7jcv5btcixnaikm

A Survey on Vision Transformer [article]

Kai Han, Yunhe Wang, Hanting Chen, Xinghao Chen, Jianyuan Guo, Zhenhua Liu, Yehui Tang, An Xiao, Chunjing Xu, Yixing Xu, Zhaohui Yang, Yiman Zhang (+1 others)
2021 arXiv   pre-print
Given its high performance and less need for vision-specific inductive bias, transformer is receiving more and more attention from the computer vision community.  ...  Toward the end of this paper, we discuss the challenges and provide several further research directions for vision transformers.  ...  Transformer for Human Pose Estimation. Lin et al. [138] proposed a mesh transformer (METRO) for predicting 3D human pose and mesh from a single RGB image.  ... 
arXiv:2012.12556v4 fatcat:ldtbdgy6tbdttfqzhzml7n577m

Table of Contents

2020 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
of Advanced Technology, Chinese Academy of Science) Multiview-Consistent Semi-Supervised Learning for 3D Human Pose Estimation 6906 Rahul Mitra (IIT Bombay), Nitesh B.  ...  Vision, Graz University of Technology, Austria) Cross-View Tracking for Multi-Human 3D Pose Estimation at Over 100 FPS 3276 Long Chen (Department of Computer Science and Technology, Tsinghua University  ... 
doi:10.1109/cvpr42600.2020.00004 fatcat:c7els2kee5cq7lh6cemeqhdcoa

A Comprehensive Study on Deep Learning-Based 3D Hand Pose Estimation Methods

Theocharis Chatzis, Andreas Stergioulas, Dimitrios Konstantinidis, Kosmas Dimitropoulos, Petros Daras
2020 Applied Sciences  
The field of 3D hand pose estimation has been gaining a lot of attention recently, due to its significance in several applications that require human-computer interaction (HCI).  ...  hand pose estimation methods.  ...  [132] proposed a framework to transfer knowledge from a dataset containing more than one source, e.g., RGB and depth maps, to a single source target dataset, using a teacher-student scheme.  ... 
doi:10.3390/app10196850 fatcat:hgyqkoyetbbilncksguarqz3bq

Embracing Imperfect Datasets: A Review of Deep Learning Solutions for Medical Image Segmentation [article]

Nima Tajbakhsh, Laura Jeyaseelan, Qian Li, Jeffrey Chiang, Zhihao Wu, Xiaowei Ding
2020 arXiv   pre-print
body of research has studied the problem of medical image segmentation with imperfect datasets, tackling two major dataset limitations: scarce annotations where only limited annotated data is available for  ...  We further compare the benefits and requirements of the surveyed methodologies and provide our recommended solutions.  ...  The unsupervised loss minimizes the dissimilarity between the segmentation masks of the original and transformed images, which are generated by the teacher and student networks, respectively.  ... 
arXiv:1908.10454v2 fatcat:mjvfbhx75bdkbheysq3r7wmhdi

Unsupervised domain adaptation for clinician pose estimation and instance segmentation in the OR [article]

Vinkle Srivastav, Afshin Gangi, Nicolas Padoy
2021
To address these concerns, we first study how joint person pose estimation and instance segmentation can be performed on low resolutions images from 1x to 12x.  ...  Computer vision models for person pixel-based segmentation and body-keypoints detection are needed to better understand the clinical activities and the spatial layout of the OR.  ...  KM-ORPose KM-ORPose (Srivastav et al., 2020) uses the teacherstudent learning paradigm for the domain adaptation in the OR for joint person detection and 2D/3D human pose estimation.  ... 
doi:10.48550/arxiv.2108.11801 fatcat:sfmrlp46qvbxxbi4guktv7g6pm

Transformers in Vision: A Survey [article]

Salman Khan, Muzammal Naseer, Munawar Hayat, Syed Waqas Zamir, Fahad Shahbaz Khan, Mubarak Shah
2021 arXiv   pre-print
Different from convolutional networks, Transformers require minimal inductive biases for their design and are naturally suited as set-functions.  ...  processing (e.g., activity recognition, video forecasting), low-level vision (e.g., image super-resolution, image enhancement, and colorization) and 3D analysis (e.g., point cloud classification and segmentation  ...  We would also like to thank Mohamed Afham for his help with a figure.  ... 
arXiv:2101.01169v4 fatcat:ynsnfuuaize37jlvhsdki54cy4

Transforming medical imaging with Transformers? A comparative review of key properties, current progresses, and future perspectives [article]

Jun Li, Junyu Chen, Yucheng Tang, Ce Wang, Bennett A. Landman, S. Kevin Zhou
2022 arXiv   pre-print
, we offer a comprehensive review of the state-of-the-art Transformer-based approaches for medical imaging and exhibit current research progresses made in the areas of medical image segmentation, recognition  ...  In particular, what distinguishes our review lies in its organization based on the Transformer's key defining properties, which are mostly derived from comparing the Transformer and CNN, and its type of  ...  The USST pre-training framework is developed based on the student-teacher paradigm, in which both the student and teacher paths share the identical architecture, but the teacher path is updated using an  ... 
arXiv:2206.01136v2 fatcat:dgrx5ftlhrdjni74tzyld4mriu
« Previous Showing results 1 — 15 out of 32 results