Zobrazeno 1 - 10
of 262
pro vyhledávání: '"Sun Ming-Ting"'
Autor:
Li, Dianqi, Zhang, Yizhe, Peng, Hao, Chen, Liqun, Brockett, Chris, Sun, Ming-Ting, Dolan, Bill
Adversarial examples expose the vulnerabilities of natural language processing (NLP) models, and can be used to evaluate and improve their robustness. Existing techniques of generating such examples are typically driven by local heuristic rules that
Externí odkaz:
http://arxiv.org/abs/2009.07502
Nonparametric approaches have shown promising results on reconstructing 3D human mesh from a single monocular image. Unlike previous approaches that use a parametric human model like skinned multi-person linear model (SMPL), and attempt to regress th
Externí odkaz:
http://arxiv.org/abs/2003.00052
Text style transfer refers to the task of rephrasing a given text in a different style. While various methods have been proposed to advance the state of the art, they often assume the transfer output follows a delta distribution, and thus their model
Externí odkaz:
http://arxiv.org/abs/2002.06525
Publikováno v:
EURASIP Journal on Image and Video Processing, Vol 2010, Iss 1, p 626324 (2010)
We address the problem of unsupervised discovery of action classes in video data. Different from all existing methods thus far proposed for this task, we present a space-time link analysis approach which consistently matches or exceeds the performanc
Externí odkaz:
https://doaj.org/article/4c489500409b4c3c808e97aa8d5a5257
Publikováno v:
EURASIP Journal on Advances in Signal Processing, Vol 2007, Iss 1, p 018578 (2007)
Externí odkaz:
https://doaj.org/article/6fdff0c399da4a2d9089cecd1ee1d7b6
Publikováno v:
EURASIP Journal on Advances in Signal Processing, Vol 2006, Iss 1, p 027579 (2006)
Externí odkaz:
https://doaj.org/article/d9c0d4c38f8d44ea84676ad28371b10f
Text style transfer without parallel data has achieved some practical success. However, in the scenario where less data is available, these methods may yield poor performance. In this paper, we examine domain adaptation for text style transfer to lev
Externí odkaz:
http://arxiv.org/abs/1908.09395
Supervised deep learning with pixel-wise training labels has great successes on multi-person part segmentation. However, data labeling at pixel-level is very expensive. To solve the problem, people have been exploring to use synthetic data to avoid t
Externí odkaz:
http://arxiv.org/abs/1907.05193
Autor:
Kuo C-C Jay, van der Schaar Mihaela, Steinbach Eckehard, Chen Jie, Sun Ming-Ting, Xiong Zixiang
Publikováno v:
EURASIP Journal on Advances in Signal Processing, Vol 2004, Iss 2, p 107917 (2004)
Externí odkaz:
https://doaj.org/article/4140c628c5b54c6f82e884271ca36fa2