-
Chapter and Conference Paper
A First-Person Vision Dataset of Office Activities
We present a multi-subject first-person vision dataset of office activities. The dataset contains the highest number of subjects and activities compared to existing office activity datasets. Office activities ...
-
Chapter and Conference Paper
Description and Recognition of Activity Patterns Using Sparse Vector Fields
Far-field activities represented as time series or trajectories can be summarized in compact representations of frequent patterns. Popular representations such as clustering or probabilistic modeling of trajec...
-
Chapter and Conference Paper
Pose Guided Human Image Synthesis by View Disentanglement and Enhanced Weighting Loss
View synthesis aims at generating a novel, unseen view of an object. This is a challenging task in the presence of occlusions and asymmetries. In this paper, we present View-Disentangled Generator (VDG), a two...
-
Chapter and Conference Paper
Residual Stacked RNNs for Action Recognition
Action recognition pipelines that use Recurrent Neural Networks (RNN) are currently 5–10% less accurate than Convolutional Neural Networks (CNN). While most works that use RNNs employ a 2D CNN on each frame to...
-
Chapter and Conference Paper
Online Multi-target Tracking with Strong and Weak Detections
We propose an online multi-target tracker that exploits both high- and low-confidence target detections in a Probability Hypothesis Density Particle Filter framework. High-confidence (strong) detections are us...
-
Chapter and Conference Paper
Pop-up Modelling of Hazy Scenes
This paper describes the construction of a layered scene-model, based on a single hazy image of an outdoor scene. A depth map and radiance image are estimated by standard dehazing methods. The radiance image i...