Latest Research
Capture, Learning, and Synthesis of 3D Speaking Styles
CVPR 2019
• TimoBolkart/voca
• 
We then train a neural network on our dataset that factors identity from facial motion.
Reweighted Expectation Maximization
• adjidieng/REM
• 
The typical workaround is to use variational inference (VI) and maximize a lower bound to the log marginal likelihood of the data.
Training Neural Networks for and by Interpolation
• oval-group/ali-g
•
We empirically show that ALI-G outperforms adaptive gradient methods such as Adam, and provides comparable performance with SGD, although SGD benefits from manual learning rate schedules.
Grid R-CNN Plus: Faster and Better
• STVIR/Grid-R-CNN
•
Grid R-CNN is a well-performed objection detection framework.
Cognitive Knowledge Graph Reasoning for One-shot Relational Learning
Inferring new facts from existing knowledge graphs (KG) with explainable reasoning processes is a significant problem and has received much attention recently.
The Replica Dataset: A Digital Replica of Indoor Spaces
• facebookresearch/Replica-Dataset
We introduce Replica, a dataset of 18 highly photo-realistic 3D indoor scene reconstructions at room and building scale.
Learning Spatio-Temporal Representation with Local and Global Diffusion
CVPR 2019 • ZhaofanQiu/local-and-global-diffusion-networks
Diffusions effectively interact two aspects of information, i. e., localized and holistic, for more powerful way of representation learning.
Unsupervised Neural Single-Document Summarization of Reviews via Learning Latent Discourse Structure and its Ranking
This paper focuses on the end-to-end abstractive summarization of a single product review without supervision.
Contrastive Multiview Coding
• HobbitLong/CMC
•
Humans view the world through many sensory channels, e. g., the long-wavelength light channel, viewed by the left eye, or the high-frequency vibrations channel, viewed by the right ear.
A JIT Compiler for Neural Network Inference
This paper describes a C++ library that compiles neural network models at runtime into machine code that performs inference.