Recent Posts

Two papers accepted to CVPR2020

Towards Interpretable Image Synthesis by Learning Sparsely Connected AND-OR Networks, Xianglei Xing, Tianfu Wu, Song-Chun Zhu and Ying Nian Wu Holistically-Attracted Wireframe Parsing, Nan Xue,...

New Framework Improves Performance of Deep Neural Networks

A new framework for building deep neural networks outperforms existing state-of-the-art artificial intelligence frameworks, including the widely-used ResNet and DenseNet systems, in visual recognition tasks.