no code implementations • 21 Mar 2024 • Shrishail Baligar, Mikolaj Kegler, Bryce Irvin, Marko Stamenovic, Shawn Newsam
First, we explore the utility of context by providing the TSE model with oracle information about what sound classes make up the input mixture, where the objective of the model is to extract one or more sources of interest indicated by the user.
no code implementations • 20 Oct 2022 • Dalton Lunga, Yingjie Hu, Shawn Newsam, Song Gao, Bruno Martins, Lexie Yang, Xueqing Deng
Geospatial Artificial Intelligence (GeoAI) is an interdisciplinary field enjoying tremendous adoption.
no code implementations • 12 Apr 2022 • Xueqing Deng, Dawei Sun, Shawn Newsam, Peng Wang
Specifically, given a pair of student and teacher networks, DistPro first sets up a rich set of KD connection from the transmitting layers of the teacher to the receiving layers of the student, and in the meanwhile, various transforms are also proposed for comparing feature maps along its pathway for the distillation.
1 code implementation • CVPR 2022 • Xueqing Deng, Peng Wang, Xiaochen Lian, Shawn Newsam
Notably, NightLab contains models at two levels of granularity, i. e. image and regional, and each level is composed of light adaptation and segmentation modules.
no code implementations • 8 Mar 2022 • Yuxin Tian, Shawn Newsam, Kofi Boakye
Effective image retrieval with text feedback stands to impact a range of real-world applications, such as e-commerce.
no code implementations • 24 Jun 2021 • Xueqing Deng, Yi Zhu, Yuxin Tian, Shawn Newsam
Neural network-based semantic segmentation has achieved remarkable results when large amounts of annotated data are available, that is, in the supervised case.
1 code implementation • 8 Dec 2020 • Xueqing Deng, Yi Zhu, Yuxin Tian, Shawn Newsam
Land-cover classification using remote sensing imagery is an important Earth observation task.
no code implementations • 23 Dec 2019 • Xueqing Deng, Yi Zhu, Yuxin Tian, Shawn Newsam
Inspired by this, we investigate methods to inform or guide deep learning models for geospatial image analysis to increase their performance when a limited amount of training data is available or when they are applied to scenarios other than which they were trained on.
no code implementations • 24 Jul 2019 • Yi Zhu, Shawn Newsam
Motivated by our observation that motion information is the key to good anomaly detection performance in video, we propose a temporal augmented network to learn a motion-aware feature.
no code implementations • 19 Feb 2019 • Xueqing Deng, Yi Zhu, Shawn Newsam
This paper develops a deep-learning framework to synthesize a ground-level view of a location given an overhead image.
5 code implementations • CVPR 2019 • Yi Zhu, Karan Sapra, Fitsum A. Reda, Kevin J. Shih, Shawn Newsam, Andrew Tao, Bryan Catanzaro
In this paper, we present a video prediction-based methodology to scale up training sets by synthesizing new training samples in order to improve the accuracy of semantic segmentation networks.
Ranked #2 on Semantic Segmentation on KITTI Semantic Segmentation (using extra training data)
no code implementations • 30 Oct 2018 • Yi Zhu, Jia Xue, Shawn Newsam
Deep neural networks have led to a series of breakthroughs in computer vision given sufficient annotated training datasets.
no code implementations • 30 Oct 2018 • Yi Zhu, Shawn Newsam
However, this does not work well for multirate videos, in which actions or subactions occur at different speeds.
no code implementations • 13 Jun 2018 • Xueqing Deng, Yi Zhu, Shawn Newsam
More significantly, we show the generated images are representative of the locations and that the representations learned by the cGANs are informative.
no code implementations • 7 May 2018 • Yi Zhu, Shawn Newsam
Despite the significant progress that has been made on estimating optical flow recently, most estimation methods, including classical and deep learning approaches, still have difficulty with multi-scale estimation, real-time computation, and/or occlusion reasoning.
no code implementations • CVPR 2018 • Yi Zhu, Yang Long, Yu Guan, Shawn Newsam, Ling Shao
Unseen Action Recognition (UAR) aims to recognise novel action categories without training examples.
Ranked #14 on Action Recognition on ActivityNet
no code implementations • 21 Feb 2018 • Xueqing Deng, Yi Zhu, Shawn Newsam
We also show that the spatial morphing kernel improves the results.
no code implementations • 7 Feb 2018 • Yi Zhu, Xueqing Deng, Shawn Newsam
We perform fine-grained land use mapping at the city scale using ground-level images.
1 code implementation • 19 Jul 2017 • Yi Zhu, Shawn Newsam
Classical approaches for estimating optical flow have achieved rapid progress in the last decade.
no code implementations • 24 Jun 2017 • Yi Zhu, Sen Liu, Shawn Newsam
This paper is the first work to perform spatio-temporal mapping of human activity using the visual content of geo-tagged videos.
no code implementations • 11 Jun 2017 • Weixun Zhou, Shawn Newsam, Congmin Li, Zhenfeng Shao
Current benchmark datasets are deficient in that 1) they were originally collected for land use/land cover classification and not image retrieval, 2) they are relatively small in terms of the number of classes as well the number of sample images per class, and 3) the retrieval performance has saturated.
no code implementations • 11 Apr 2017 • Yi Zhu, Shawn Newsam, Zaikun Xu
This notebook paper describes our system for the untrimmed classification task in the ActivityNet challenge 2016.
3 code implementations • 2 Apr 2017 • Yi Zhu, Zhenzhong Lan, Shawn Newsam, Alexander G. Hauptmann
State-of-the-art action recognition approaches rely on traditional optical flow estimation methods to pre-compute motion information for CNNs.
Ranked #20 on Action Recognition on UCF101
no code implementations • 8 Feb 2017 • Yi Zhu, Zhenzhong Lan, Shawn Newsam, Alexander G. Hauptmann
We study the unsupervised learning of CNNs for optical flow estimation using proxy ground truth data.
no code implementations • 22 Dec 2016 • Yi Zhu, Shawn Newsam
We employ a multi-task learning framework that performs the three highly related steps of action proposal, action recognition, and action localization refinement in parallel instead of the standard sequential pipeline that performs the steps in order.
no code implementations • 10 Oct 2016 • Weixun Zhou, Shawn Newsam, Congmin Li, Zhenfeng Shao
In this paper, we investigate how to extract deep feature representations based on convolutional neural networks (CNN) for high-resolution remote sensing image retrieval (HRRSIR).
no code implementations • 21 Sep 2016 • Yi Zhu, Shawn Newsam
Land use mapping is a fundamental yet challenging task in geographic science.
no code implementations • 21 Sep 2016 • Yi Zhu, Shawn Newsam
We perform spatio-temporal analysis of public sentiment using geotagged photo collections.
no code implementations • 15 Aug 2016 • Yi Zhu, Shawn Newsam
This paper performs the first investigation into depth for large-scale human action recognition in video where the depth cues are estimated from the videos themselves.