Výsledky vyhledávání

Report

Pose Guided Human Image Synthesis with Partially Decoupled GAN

Autor: Wu, Jianhan, Wang, Jianzong, Si, Shijing, Qu, Xiaoyang, Xiao, Jing

Pose Guided Human Image Synthesis (PGHIS) is a challenging task of transforming a human image from the reference pose to a target pose while preserving its style. Most existing methods encode the texture of the whole reference human image into a late

Externí odkaz: http://arxiv.org/abs/2210.03627

Zobrazit plný text záznamu

Report

Improving Human Image Synthesis with Residual Fast Fourier Transformation and Wasserstein Distance

Autor: Wu, Jianhan, Si, Shijing, Wang, Jianzong, Xiao, Jing

With the rapid development of the Metaverse, virtual humans have emerged, and human image synthesis and editing techniques, such as pose transfer, have recently become popular. Most of the existing techniques rely on GANs, which can generate good hum

Externí odkaz: http://arxiv.org/abs/2205.12022

Zobrazit plný text záznamu

Report

Augmentation-induced Consistency Regularization for Classification

Autor: Wu, Jianhan, Si, Shijing, Wang, Jianzong, Xiao, Jing

Deep neural networks have become popular in many supervised learning tasks, but they may suffer from overfitting when the training dataset is limited. To mitigate this, many researchers use data augmentation, which is a widely used and effective meth

Externí odkaz: http://arxiv.org/abs/2205.12461

Zobrazit plný text záznamu

Report

Variational Information Bottleneck for Effective Low-resource Audio Classification

Autor: Si, Shijing, Wang, Jianzong, Sun, Huiming, Wu, Jianhan, Zhang, Chuanyao, Qu, Xiaoyang, Cheng, Ning, Chen, Lei, Xiao, Jing

Large-scale deep neural networks (DNNs) such as convolutional neural networks (CNNs) have achieved impressive performance in audio classification for their powerful capacity and strong generalization ability. However, when training a DNN model on low

Externí odkaz: http://arxiv.org/abs/2107.04803

Zobrazit plný text záznamu