no code implementations • ECCV 2020 • Xuejian Rong, Denis Demandolx, Kevin Matzen, Priyam Chatterjee, YingLi Tian
As a result, imaging pipelines often rely on computational photography to improve SNR by fusing multiple short exposures.
no code implementations • 27 Nov 2023 • Elahe Vahdani, YingLi Tian
This paper addresses the challenge of point-supervised temporal action detection, in which only one frame per action instance is annotated in the training set.
no code implementations • 20 Oct 2023 • Elahe Vahdani, YingLi Tian
This paper tackles the challenge of point-supervised temporal action detection, wherein only a single frame is annotated for each action instance in the training set.
no code implementations • SignLang (LREC) 2022 • Saad Hassan, Matthew Seita, Larwan Berke, YingLi Tian, Elaine Gale, Sooyeon Lee, Matt Huenerfauth
We are releasing a dataset containing videos of both fluent and non-fluent signers using American Sign Language (ASL), which were collected using a Kinect v2 sensor.
no code implementations • 20 Apr 2022 • HaiYan Wang, YingLi Tian
Point cloud has drawn more and more research attention as well as real-world applications.
no code implementations • CVPR 2022 • HaiYan Wang, Will Hutchcroft, Yuguang Li, Zhiqiang Wan, Ivaylo Boyadzhiev, YingLi Tian, Sing Bing Kang
In this paper, we propose a new deep learning-based method for estimating room layout given a pair of 360 panoramas.
1 code implementation • 29 Mar 2022 • Ziyue Feng, Liang Yang, Longlong Jing, HaiYan Wang, YingLi Tian, Bing Li
Conventional self-supervised monocular depth prediction methods are based on a static environment assumption, which leads to accuracy degradation in dynamic scenes due to the mismatch and occlusion problems introduced by object motions.
no code implementations • 9 Jan 2022 • Kien Nguyen, Clinton Fookes, Sridha Sridharan, YingLi Tian, Feng Liu, Xiaoming Liu, Arun Ross
The rapid emergence of airborne platforms and imaging sensors are enabling new forms of aerial surveillance due to their unprecedented advantages in scale, mobility, deployment and covert observation capabilities.
1 code implementation • 22 Oct 2021 • Zhimin Chen, Longlong Jing, Yang Liang, YingLi Tian, Bing Li
This paper explores how the coherence of different modelities of 3D data (e. g. point cloud, image, and mesh) can be used to improve data efficiency for both 3D classification and retrieval tasks.
no code implementations • 30 Sep 2021 • Elahe Vahdani, YingLi Tian
The task of temporal activity detection in untrimmed videos aims to localize the temporal boundary of actions and classify the action categories.
no code implementations • 29 Sep 2021 • Longlong Jing, Zhimin Chen, Bing Li, YingLi Tian
Our proposed novel self-supervised model learns two types of distinct features: modality-invariant features and modality-specific features.
2 code implementations • 20 Sep 2021 • Ziyue Feng, Longlong Jing, Peng Yin, YingLi Tian, Bing Li
Unlike the existing methods that use sparse LiDAR mainly in a manner of time-consuming iterative post-processing, our model fuses monocular image features and sparse LiDAR features to predict initial depth maps.
Ranked #1 on Depth Completion on KITTI
no code implementations • CVPR 2021 • Longlong Jing, Elahe Vahdani, Jiaxing Tan, YingLi Tian
Cross-modal retrieval aims to learn discriminative and modal-invariant features for data from different modalities.
1 code implementation • CVPR 2021 • HaiYan Wang, Jiahao Pang, Muhammad A. Lodhi, YingLi Tian, Dong Tian
Scene flow depicts the dynamics of a 3D scene, which is critical for various applications such as autonomous driving, robot navigation, AR/VR, etc.
no code implementations • 8 Aug 2020 • Longlong Jing, Elahe Vahdani, Jiaxing Tan, YingLi Tian
Cross-modal retrieval aims to learn discriminative and modal-invariant features for data from different modalities.
no code implementations • 2 Jun 2020 • Yu-cheng Chen, YingLi Tian, Mingyi He
Vision-based monocular human pose estimation, as one of the most fundamental and challenging problems in computer vision, aims to obtain posture of the human body from input images or video sequences.
no code implementations • 28 May 2020 • Longlong Jing, Yu-cheng Chen, Ling Zhang, Mingyi He, YingLi Tian
By exploring the inherent multi-modality attributes of 3D objects, in this paper, we propose to jointly learn modal-invariant and view-invariant features from different modalities including image, point cloud, and mesh with heterogeneous networks for 3D data.
no code implementations • 1 May 2020 • Elahe Vahdani, Longlong Jing, YingLi Tian, Matt Huenerfauth
Our system is able to recognize grammatical elements on ASL-HW-RGBD from manual gestures, facial expressions, and head movements and successfully detect 8 ASL grammatical mistakes.
no code implementations • LREC 2020 • Saad Hassan, Larwan Berke, Elahe Vahdani, Longlong Jing, YingLi Tian, Matt Huenerfauth
We have collected a new dataset consisting of color and depth videos of fluent American Sign Language (ASL) signers performing sequences of 100 ASL signs from a Kinect v2 sensor.
no code implementations • 26 Apr 2020 • Hai-Yan Wang, Xuejian Rong, Liang Yang, Jinglun Feng, Jizhong Xiao, YingLi Tian
The deficiency of 3D segmentation labels is one of the main obstacles to effective point cloud segmentation, especially for scenes in the wild with varieties of different objects.
no code implementations • 13 Apr 2020 • Longlong Jing, Yu-cheng Chen, Ling Zhang, Mingyi He, YingLi Tian
Specifically, 2D image features of rendered images from different views are extracted by a 2D convolutional neural network, and 3D point cloud features are extracted by a graph convolution neural network.
no code implementations • 29 Feb 2020 • Longlong Jing, Toufiq Parag, Zhe Wu, YingLi Tian, Hongcheng Wang
To minimize the dependence on a large annotated dataset, our proposed semi-supervised method trains from a small number of labeled examples and exploits two regulatory signals from unlabeled data.
no code implementations • 25 Jul 2019 • Jingya Liu, Liangliang Cao, Oguz Akin, YingLi Tian
Accurate detection of pulmonary nodules with high sensitivity and specificity is essential for automatic lung cancer diagnosis from CT scans.
no code implementations • 8 Jun 2019 • Jingya Liu, Liangliang Cao, Oguz Akin, YingLi Tian
Accurate detection of pulmonary nodules with high sensitivity and specificity is essential for automatic lung cancer diagnosis from CT scans.
no code implementations • 7 Jun 2019 • Longlong Jing, Elahe Vahdani, Matt Huenerfauth, YingLi Tian
In this paper, we propose a 3D Convolutional Neural Network (3DCNN) based multi-stream framework to recognize American Sign Language (ASL) manual signs (consisting of movements of the hands, as well as non-manual face movements in some cases) in real-time from RGB-D videos, by fusing multimodality features including hand gestures, facial expressions, and body poses from multi-channels (RGB, depth, motion, and skeleton joints).
no code implementations • 16 Feb 2019 • Longlong Jing, YingLi Tian
This paper provides an extensive review of deep learning-based self-supervised general visual feature learning methods from images or videos.
Self-Supervised Image Classification Self-Supervised Learning
1 code implementation • 11 Jan 2019 • Jiaxing Tan, Longlong Jing, Yumei Huo, YingLi Tian, Oguz Akin
Lung segmentation in computerized tomography (CT) images is an important procedure in various lung disease diagnosis.
no code implementations • 28 Dec 2018 • Longlong Jing, Yu-cheng Chen, YingLi Tian
The enhanced coarse mask is fed to a fully convolutional neural network to be recursively refined.
no code implementations • NeurIPS 2019 • Benjamin Planche, Xuejian Rong, Ziyan Wu, Srikrishna Karanam, Harald Kosch, YingLi Tian, Jan Ernst, Andreas Hutter
We present a method to incrementally generate complete 2D or 3D scenes with the following properties: (a) it is globally consistent at each step according to a learned scene prior, (b) real observations of a scene can be incorporated while observing global consistency, (c) unobserved regions can be hallucinated locally in consistence with previous observations, hallucinations and global priors, and (d) hallucinations are statistical in nature, i. e., different scenes can be generated from the same observations.
no code implementations • 29 Nov 2018 • Yuancheng Ye, Xiaodong Yang, YingLi Tian
In this paper, we address the challenging problem of spatial and temporal action detection in videos.
no code implementations • 28 Nov 2018 • Longlong Jing, Xiaodong Yang, Jingen Liu, YingLi Tian
The success of deep neural networks generally requires a vast amount of training data to be labeled, which is expensive and unfeasible in scale, especially for video collections.
Ranked #42 on Self-Supervised Action Recognition on HMDB51
Self-Supervised Action Recognition Temporal Action Localization +1
no code implementations • 15 Sep 2017 • Yang Xian, YingLi Tian
Afterwards, during the training of sg-LSTM on the rest training data, this guiding information serves as additional input to the network along with the image representations and the ground-truth descriptions.
no code implementations • CVPR 2017 • Xuejian Rong, Chucai Yi, YingLi Tian
Text instance as one category of self-described objects provides valuable information for understanding and describing cluttered scenes.
no code implementations • CVPR 2014 • Xiaodong Yang, YingLi Tian
This paper presents a new framework for human activity recognition from video sequences captured by a depth camera.