Research Projects

Factorized Bilinear Models for Image Recognition

ICCV 2017
Y. Li, N. Wang, J. Liu, and X. Hou

In this paper, we propose a novel Factorized Bilinear (FB) layer to model the pairwise feature inter- actions by considering the quadratic terms in the transformations. Compared with existing methods that tried to in- corporate complex non-linearity structures into CNNs, the factorized parameterization makes our FB layer only re- quire a linear increase of parameters and affordable computational cost. To further reduce the risk of overfitting of the FB layer, a specific remedy called DropFactor is devised during the training process.
arXiv Code Project Page

Demystifying Neural Style Transfer

IJCAI 2017
Y. Li, N. Wang, J. Liu, and X. Hou

In this paper, we propose a novel interpretation of neural style transfer by treating it as a domain adaptation problem. Specifically, we theoretically show that matching the Gram matrices of feature maps is equivalent to minimize the Maximum Mean Discrepancy (MMD) with the second order polynomial kernel. Thus, we argue that the essence of neural style transfer is to match the feature distributions between the style images and the generated images.
arXiv Code Project Page

Revisiting Batch Normalization for Practical Domain Adaptation

ICLRW 2017
Y. Li, N. Wang, J. Shi, J. Liu, and X. Hou

In this paper, we propose a simple yet powerful remedy, called Adaptive Batch Normalization (AdaBN) to increase the generalization ability of a DNN. By modulating the statistics from the source domain to the target domain in all Batch Normalization layers across the network, our approach achieves deep adaptation effect for domain adaptation tasks. In contrary to other deep learning domain adaptation methods, our method does not require additional components, and is parameter-free.
ICLR Workshop Project Page

Online Human Action Detection using Joint Classification-Regression Recurrent Neural Networks

ECCV 2016
Y. Li, C. Lan, J. Xing, W. Zeng, C. Yuan and J. Liu

In this paper, we propose a Joint Classification-Regression Recurrent Neural Network to accurately detect the actions and localize the start and end positions of the actions on the fly. At the same time, we can forecast their occurrences in advance based on the regressed curves.
arXiv Project Page

Multi-Pose Face Hallucination via Neighbor Embedding for Facial Components

ICIP 2015
Y. Li, J. Liu, W. Yang and Z. Guo

In this paper, we propose a novel multi-pose face hallucination method based on Neighbor Embedding for Facial Components (NEFC) to magnify face images with various poses and expressions.
PDF Code Project Page

Neighborhood Regression for Edge-Preserving Image Super-Resolution

Y. Li, J. Liu, W. Yang and Z. Guo

In this project, we propose a novel edge-preserving super-resolution algorithm, which reconstructs low- and high-frequency components separately, rather than reconstructing the whole components in the image directly.
PDF Code Project Page

Image Transformation using Limited Reference with Application to Photo-Sketch Synthesis

VCIP 2014
W. Bai, Y. Li, J. Liu and Z. Guo

In this project, we propose a sparse representation based framework of transforming images with limited reference, which can be used for the typical image transformation application, photo-sketch synthesis.
PDF Project Page

Course Projects

Weibo Visuliazation

Introduction to Visualization, 2014 - 2015 Fall

Implemented a Weibo visualization and interaction system for users to explore top events among thousands of weibos in a specific time.