• Publications
  • Influence
VTR 7.0: Next Generation Architecture and CAD System for FPGAs
TLDR
VTR can now generate a netlist of the final post-routed circuit which enables detailed simulation of a design for a variety of purposes. Expand
  • 289
  • 71
  • PDF
DeepVO: Towards end-to-end visual odometry with deep Recurrent Convolutional Neural Networks
TLDR
This paper presents a novel end-to-end monocular VO algorithm based on Deep Learning based on deep Recurrent Convolutional Neural Networks. Expand
  • 310
  • 52
  • PDF
Safety Verification of Deep Neural Networks
TLDR
We develop a novel automated verification framework for feed-forward multi-layer neural networks based on Satisfiability Modulo Theory (SMT). Expand
  • 486
  • 28
  • PDF
Optimal multipath congestion control and request forwarding in Information-Centric Networks
TLDR
The evolution of the Internet into a distributed Information access system calls for a paradigm shift to enable an evolvable future network architecture. Expand
  • 104
  • 21
  • PDF
VidLoc: A Deep Spatio-Temporal Model for 6-DoF Video-Clip Relocalization
TLDR
We use temporal smoothness of image-sequences to improve the accuracy of 6-DoF camera re-localization. Expand
  • 110
  • 16
  • PDF
End-to-end, sequence-to-sequence probabilistic visual odometry through deep neural networks
TLDR
An end-to-end, sequence- to-sequence probabilistic visual odometry framework is proposed for the monocular VO based on deep recurrent convolutional neural networks. Expand
  • 91
  • 13
  • PDF
VINet: Visual-Inertial Odometry as a Sequence-to-Sequence Learning Problem
TLDR
In this paper we present an end-to-end trainable method for visual-inertial odometry which performs fusion of the data at an intermediate feature-representation level. Expand
  • 154
  • 11
  • PDF
Reinforced Self-Attention Network: a Hybrid of Hard and Soft Attention for Sequence Modeling
TLDR
In this paper, we integrate both soft and hard attention into one context fusion model, "reinforced self-attention (ReSA)", for the mutual benefit of each other. Expand
  • 73
  • 11
  • PDF
Multimodal sentiment analysis with word-level fusion and reinforcement learning
TLDR
We propose the Gated Multimodal Embedding LSTM with Temporal Attention (GME-LSTM(A)) model that is composed of 2 modules. Expand
  • 101
  • 9
  • PDF
Fingerprint enhancement in the singular point area
  • S. Wang, Y. Wang
  • Computer Science
  • IEEE Signal Processing Letters
  • 7 January 2004
TLDR
We present a new method and design a new filter to enhance a fingerprint in the singular point area, and the time required for our algorithm is reduced. Expand
  • 93
  • 8