• Publications
  • Influence
Unsupervised Video Summarization with Adversarial LSTM Networks
TLDR
This paper addresses the problem of unsupervised video summarization, formulated as selecting a sparse subset of video frames that optimally represent the input video. Expand
  • 187
  • 42
  • PDF
Local-Learning-Based Feature Selection for High-Dimensional Data Analysis
TLDR
We propose a new feature-selection algorithm that addresses several major issues with prior work, including problems with algorithm implementation, computational complexity, and solution accuracy. Expand
  • 287
  • 25
  • PDF
Adaptive boosting for SAR automatic target recognition
TLDR
The paper proposed a novel automatic target recognition (ATR) system for classification of three types of ground vehicles in the moving and stationary target acquisition and recognition database. Expand
  • 224
  • 23
Learning spatiotemporal graphs of human activities
TLDR
We advance prior work by learning what activity parts and their spatiotemporal relations should be captured to represent the activity, and how relevant they are for enabling efficient inference in realistic videos. Expand
  • 233
  • 14
  • PDF
Multiobject tracking as maximum weight independent set
TLDR
We present a new, polynomial-time MWIS algorithm, and prove that it converges to an optimum. Expand
  • 260
  • 13
  • PDF
CERN: Confidence-Energy Recurrent Network for Group Activity Recognition
TLDR
This work is about recognizing human activities occurring in videos at distinct semantic levels, including individual actions, interactions, and group activities. Expand
  • 59
  • 12
  • PDF
A chains model for localizing participants of group activities in videos
  • M. Amer, S. Todorovic
  • Computer Science
  • International Conference on Computer Vision
  • 6 November 2011
TLDR
We specify a new, mid-level, video feature aimed at summarizing local visual cues into bags of the right detections (BORDs). Expand
  • 62
  • 12
  • PDF
Monocular Depth Estimation Using Neural Regression Forest
  • A. Roy, S. Todorovic
  • Computer Science
  • IEEE Conference on Computer Vision and Pattern…
  • 27 June 2016
TLDR
This paper presents a novel deep architecture, called neural regression forest (NRF), for depth estimation from a single image. Expand
  • 194
  • 11
  • PDF
Video object segmentation by tracking regions
TLDR
We use a low-level segmentation to extract regions in all frames, and then we transitively match and cluster them across the video. Expand
  • 189
  • 11
  • PDF
AFEW-VA database for valence and arousal estimation in-the-wild
TLDR
We propose a new dataset of highly accurate per-frame annotations of valence and arousal for 600 challenging video clips extracted from feature films (also used in part for the AFEW dataset). Expand
  • 65
  • 11
  • PDF