Filters








326 Hits in 2.5 sec

Regularizing Generative Adversarial Networks under Limited Data [article]

Hung-Yu Tseng, Lu Jiang, Ce Liu, Ming-Hsuan Yang, Weilong Yang
2021 arXiv   pre-print
Recent years have witnessed the rapid progress of generative adversarial networks (GANs). However, the success of the GAN models hinges on a large amount of training data. This work proposes a regularization approach for training robust GAN models on limited data. We theoretically show a connection between the regularized loss and an f-divergence called LeCam-divergence, which we find is more robust under limited training data. Extensive experiments on several benchmark datasets demonstrate
more » ... the proposed regularization scheme 1) improves the generalization performance and stabilizes the learning dynamics of GAN models under limited training data, and 2) complements the recent data augmentation methods. These properties facilitate training GAN models to achieve state-of-the-art performance when only limited training data of the ImageNet benchmark is available.
arXiv:2104.03310v1 fatcat:b5th6vdcafgc5ostnr64xdzpte

RetrieveGAN: Image Synthesis via Differentiable Patch Retrieval [article]

Hung-Yu Tseng, Hsin-Ying Lee, Lu Jiang, Ming-Hsuan Yang, Weilong Yang
2020 arXiv   pre-print
., Yang, X., Liu, M.Y., Wang, T.C., Lu, Y.D., Yang, M.H., Kautz, J.: Dancing to music. In: NeurIPS (2019) 2 22.  ...  ., Yang, W., Jiang, L., Le, M., Essa, I., Gong, H., Yang, M.H.: Neural design network: Graphic layout generation with constraints. In: ECCV (2020) 2 21.  ... 
arXiv:2007.08513v1 fatcat:6mqqe7diczcrteoor4sk5o3ruy

Automatic Non-Linear Video Editing Transfer [article]

Nathan Frey, Peggy Chi, Weilong Yang, Irfan Essa
2021 arXiv   pre-print
We propose an automatic approach that extracts editing styles in a source video and applies the edits to matched footage for video creation. Our Computer Vision based techniques considers framing, content type, playback speed, and lighting of each input video segment. By applying a combination of these features, we demonstrate an effective method that automatically transfers the visual and temporal styles from professionally edited videos to unseen raw footage. We evaluated our approach with
more » ... l-world videos that contained a total of 3872 video shots of a variety of editing styles, including different subjects, camera motions, and lighting. We reported feedback from survey participants who reviewed a set of our results.
arXiv:2105.06988v1 fatcat:iuwfhdur4jeydheuukp5manehq

Latent Boosting for Action Recognition

Zhi Feng Huang, Weilong Yang, Yang Wang, Greg Mori
2011 Procedings of the British Machine Vision Conference 2011  
In this paper we present LatentBoost, a novel learning algorithm for training models with latent variables in a boosting framework. This algorithm allows for training of structured latent variable models with boosting. The popular latent SVM framework allows for training of models with structured latent variables in a max-margin framework. LatentBoost provides an analogous capability for boosting algorithms. The effectiveness of this framework is highlighted by an application to human action
more » ... ognition. We show that LatentBoost can be used to train an action recognition model in which the trajectory of a person is a latent variable. This model outperforms baselines on a variety of datasets.
doi:10.5244/c.25.132 dblp:conf/bmvc/HuangY0M11 fatcat:akbmzefu5ncihkcsqc25p5zs6u

Efficient Human Action Detection Using a Transferable Distance Function [chapter]

Weilong Yang, Yang Wang, Greg Mori
2010 Lecture Notes in Computer Science  
In this paper, we address the problem of efficient human action detection with only one template. We choose the standard slidingwindow approach to scan the template video against test videos, and the template video is represented by patch-based motion features. Using generic knowledge learnt from previous training sets, we weight the patches on the template video, by a transferable distance function. Based on the patch weighting, we propose a cascade structure which can efficiently scan the
more » ... late video over test videos. Our method is evaluated on a human action dataset with cluttered background, and a ballet video with complex human actions. The experimental results show that our cascade structure not only achieves very reliable detection, but also can significantly improve the efficiency of patch-based human action detection, with an order of magnitude improvement in efficiency.
doi:10.1007/978-3-642-12304-7_39 fatcat:ehrnlx3zlnfqvawojl2jzuxqsi

Model-based Statistical Depth with Applications to Functional Data [article]

Weilong Zhao, Zishen Xu, Yun Yang, Wei Wu
2019 arXiv   pre-print
Statistical depth, a commonly used analytic tool in non-parametric statistics, has been extensively studied for multivariate and functional observations over the past few decades. Although various forms of depth were introduced, they are mainly procedure-based whose definitions are independent of the generative model for observations. To address this problem, we introduce a generative model-based approach to define statistical depth for both multivariate and functional data. The proposed
more » ... ased depth framework permits simple computation via Monte Carlo sampling and improves the depth estimation accuracy. When applied to functional data, the proposed depth can capture important features such as continuity, smoothness, or phase variability, depending on the defining criteria. Specifically, we view functional data as realizations from a second-order stochastic process, and define their depths through the eigensystem of the covariance operator. These new definitions are given through a proper metric related to the reproducing kernel Hilbert space of the covariance operator. We propose efficient algorithms to compute the proposed depths and establish estimation consistency. Through simulations and real data, we demonstrate that the proposed functional depths reveal important statistical information such as those captured by the median and quantiles, and detect outliers.
arXiv:1909.12412v1 fatcat:h6cshkcifvehli5umpg6hjb42i

Human action recognition from a single clip per action

Weilong Yang, Yang Wang, Greg Mori
2009 2009 IEEE 12th International Conference on Computer Vision Workshops, ICCV Workshops  
Learning-based approaches for human action recognition often rely on large training sets. Most of these approaches do not perform well when only a few training samples are available. In this paper, we consider the problem of human action recognition from a single clip per action. Each clip contains at most 25 frames. Using a patch based motion descriptor and matching scheme, we can achieve promising results on three different action datasets with a single clip as the template. Our results are
more » ... mparable to previously published results using much larger training sets. We also present a method for learning a transferable distance function for these patches. The transferable distance function learning extracts generic knowledge of patch weighting from previous training sets, and can be applied to videos of new actions without further learning. Our experimental results show that the transferable distance function learning not only improves the recognition accuracy of the single clip action recognition, but also significantly enhances the efficiency of the matching scheme.
doi:10.1109/iccvw.2009.5457663 dblp:conf/iccvw/Yang0M09 fatcat:6ib727kbu5f5ddwsgozikmudhy

Image Retrieval with Structured Object Queries Using Latent Ranking SVM [chapter]

Tian Lan, Weilong Yang, Yang Wang, Greg Mori
2012 Lecture Notes in Computer Science  
We consider image retrieval with structured object queriesqueries that specify the objects that should be present in the scene, and their spatial relations. An example of such queries is "car on the road". Existing image retrieval systems typically consider queries consisting of object classes (i.e. keywords). They train a separate classifier for each object class and combine the output heuristically. In contrast, we develop a learning framework to jointly consider object classes and their
more » ... ions. Our method considers not only the objects in the query ("car" and "road" in the above example), but also related object categories can be useful for retrieval. Since we do not have ground-truth labeling of object bounding boxes on the test image, we represent them as latent variables in our model. Our learning method is an extension of the ranking SVM with latent variables, which we call latent ranking SVM. We demonstrate image retrieval and ranking results on a dataset with more than a hundred of object classes.
doi:10.1007/978-3-642-33783-3_10 fatcat:f4u2aee765b6jma5ukrnqz2ed4

Neural Design Network: Graphic Layout Generation with Constraints [article]

Hsin-Ying Lee, Lu Jiang, Irfan Essa, Phuong B Le, Haifeng Gong, Ming-Hsuan Yang, Weilong Yang
2020 arXiv   pre-print
Graphic design is essential for visual communication with layouts being fundamental to composing attractive designs. Layout generation differs from pixel-level image synthesis and is unique in terms of the requirement of mutual relations among the desired components. We propose a method for design layout generation that can satisfy user-specified constraints. The proposed neural design network (NDN) consists of three modules. The first module predicts a graph with complete relations from a
more » ... with user-specified relations. The second module generates a layout from the predicted graph. Finally, the third module fine-tunes the predicted layout. Quantitative and qualitative experiments demonstrate that the generated layouts are visually similar to real design layouts. We also construct real designs based on predicted layouts for a better understanding of the visual quality. Finally, we demonstrate a practical application on layout recommendation.
arXiv:1912.09421v2 fatcat:t77s4gw43rfadces7hkkms7pcu

Recent Development of pH-Responsive Polymers for Cancer Nanomedicine

Houliang Tang, Weilong Zhao, Jinming Yu, Yang Li, Chao Zhao
2018 Molecules  
Reproduced with permission from Xinpeng Ma, Yiguang Wang, Tian Zhao, Yang Li, Lee-Chun Su, Zhaohui Wang, Gang Huang, Baran D. Sumer, and Jinming Gao.  ...  Reproduced with permission from Xinpeng Ma, Yiguang Wang, Tian Zhao, Yang Li, Lee-Chun Su, Zhaohui Wang, Gang Huang, Baran D. Sumer, and Jinming Gao.  ... 
doi:10.3390/molecules24010004 fatcat:rkggml2z4bfdlbbpjpyyxjzgma

A new furostanol saponin from Dendrobium chrysanthum Lindl. with cytotoxic activity

Chao Yang, Weilong Lin, Lin Zhao, Jinyan Cai
2018 Figshare  
A new furostanol saponin, (25R)-26-O-(α-d-glucopyranosyl)-(1→2)-α-l-rhamnopyranosyl-furost-5-ene-3β, 22α, 26-triol-3-O-α-d-glucopyranoside (1), together with four known compounds 2–5 were isolated from the ethanolic extract of the stems of Dendrobium chrysanthum Lindl. The structures of these new compounds were identified by extensive spectroscopic analysis including 1D and 2D NMR and HR-ESI-MS, as well as chemical methods. Compounds 1–3 were isolated from D. chrysanthum for the first time.
more » ... hermore, the inhibitory effects of the compounds on tumor cells were evaluated, and compounds 1–2 exhibited significant cytotoxic activities potentially against SPC-A1, MCF-7 and HeLa human cancer cell lines. Compounds 3–5 showed inhibitory activity against the SPC-A1 and MCF-7.
doi:10.6084/m9.figshare.6026804.v1 fatcat:mag5h4zevrdrvd6grpan5gqhny

Recognizing human actions from still images with latent poses

Weilong Yang, Yang Wang, Greg Mori
2010 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition  
We consider the problem of recognizing human actions from still images. We propose a novel approach that treats the pose of the person in the image as latent variables that will help with recognition. Different from other work that learns separate systems for pose estimation and action recognition, then combines them in an ad-hoc fashion, our system is trained in an integrated fashion that jointly considers poses and actions. Our learning objective is designed to directly exploit the pose
more » ... ation for action recognition. Our experimental results demonstrate that by inferring the latent poses, we can improve the final action recognition results.
doi:10.1109/cvpr.2010.5539879 dblp:conf/cvpr/YangWM10 fatcat:jysuanuvrvcjrb36rwvn7klg6a

DearFSAC: An Approach to Optimizing Unreliable Federated Learning via Deep Reinforcement Learning [article]

Chenghao Huang, Weilong Chen, Yuxi Chen, Shunji Yang, Yanru Zhang
2022 arXiv   pre-print
Then, the objective of clients is converted into an empirical risk minimization [Yang et al., 2019] as follows: min w g ∈R d F (w g ) = 1 N N i=1 f i (w g ) , (1) f i (w g ) = E ξi∼Di [f (w g , ξ i )  ... 
arXiv:2201.12701v1 fatcat:ybjjj43b55df3nhi3exkge3mgu

Discriminative tag learning on YouTube videos with latent sub-tags

Weilong Yang, George Toderici
2011 CVPR 2011  
Another application of the tag learning could be verifying that *This work was done while Weilong Yang was an intern at Google, Inc. Figure 1 .  ... 
doi:10.1109/cvpr.2011.5995402 dblp:conf/cvpr/YangT11 fatcat:s7uzw2735fe2pfeiee4ddawhg4

Liposuction-Like Sclerotherapy Technique for Microcystic Lymphatic Malformation

Huaijie Wang, Chong Xie, Weilong Lin, Jinbang Zhou, Weijia Yang, Zhengtuan Guo
2022 Cureus  
The treatment for microcystic lymphatic malformation (LM) remains challenging. We describe the liposuction-like sclerotherapy technique, a new treatment for extensive microcystic LM. Methods LM data was retrospectively reviewed. This study included patients with a microcystic LM component treated by liposuction-like technique with bleomycin sclerotherapy. Results Between June 2016 and October 2019, 39 consecutive patients (male/female ratio: 21:18; mean age, 33.6 months; range: 5 months to 15
more » ... ars) with microcystic LM were treated by liposuction-like sclerotherapy (LS-LS) technique. Fifty-six sessions of LS-LS were performed (mean of 1.44 sessions per patient; range: one to four sessions). Follow-up ranged 6-30 months (mean of 21 months). We observed no major complications. Transient minor complications included: postoperative noninfectious fever, vomiting, temporary skin edema, pigmentation, mild local depressions, and/or irregularities, and a small hyperpigmented scar at the incision. No postoperative infection, skin ulcer, or necrosis occurred. The patients' symptoms were successfully resolved or stable. A sub-complete response and partial response were observed for 26 (76%) and 13 patients (33%), respectively. Conclusion The LS-LS technique for microcystic LMs is safe, feasible, and effective. This technique is an effective intervention with which it is possible to manage and potentially cure microcystic LM clinically.
doi:10.7759/cureus.22795 pmid:35291728 pmcid:PMC8896249 fatcat:b3l5kwgytffcjirpxd75lxzy6e
« Previous Showing results 1 — 15 out of 326 results