On August 27, XPeng announced the first major upgrade to its second-generation VLA. The theme is time. As the designer and productization lead for the Infini-VLA architecture behind this upgrade, I want to use this post to explain what we changed, why we changed it, and how it differs from what the rest of the industry is doing.
 |
Multi-style Generative Network for Real-time Transfer [arXiv] [project] Hang Zhang, Kristin Dana
@article{zhang2017multistyle,
title={Multi-style Generative Network for Real-time Transfer},
author={Zhang, Hang and Dana, Kristin},
journal={arXiv preprint arXiv:1703.06953},
year={2017}
}
|
 |
This is a CVPR 2017 submission Deep TEN: Texture Encoding Network. Here we share the idea of this paper briefly. We integrate the entire dictionary learning and encoding framework into a single CNN model. This model is compatible with existing deep learning framework and enables end-to-end material/texture recognition. We have achieved state-of-the-art results on several gold-standard benchmarks. As a side product, we obtain 49.8% relative improvement on STL-10 dataset comparing to previous state-of-the-art by joint training with CIFAR-10 dataset. |
This is a tutorial of converting Torch pre-trained model to Caffe model for inference. This toolbox is originally provided by facebook AI, but it is kind of out of date and hard to install. We added support for some components (such as batch normalizatin and spatial dropout). This tutorial is trying to save your life from debugging and installation.
