109. DETECTING TEMPORALLY CONSISTENT OBJECTS IN VIDEOS THROUGH OBJECT CLASS LABEL PROPAGATION

Department: Electrical & Computer Engineering
Faculty Advisor(s): Truong Nguyen | Serge J. Belongie

Primary Student
Name: Subarna Tripathi
Email: stripath@ucsd.edu
Phone: 858-999-5306
Grad Year: 2018

Abstract
Object proposals for detecting moving or static video objects need to address issues such as speed, memory complexity and temporal consistency. We propose an efficient Video Object Proposal (VOP) generation method and show its efficacy in learning a better video object detector. A deep-learning based video object detector learned using the proposed VOP achieves state-of-the-art detection performance on the Youtube-Objects dataset. We further propose a clustering of VOPs which can efficiently be used for detecting objects in video in a streaming fashion. As opposed to applying per-frame convolutional neural network (CNN) based object detection, our proposed method called Objects in Video Enabler thRough LAbel Propagation (OVERLAP) needs to classify only a small fraction of all candidate proposals in every video frame through streaming clustering of object proposals and class-label propagation. Project details are available at http://acsweb.ucsd.edu/~stripath/research/VOP.html

Industry Application Area(s)
Software, Analytics

Related Links:

  1. https://github.com/subtri/streaming_VOP_clustering
  2. http://vision.cornell.edu/se3/wp-content/uploads/2016/01/OVERLAP_WACV_275.pdf

« Back to Posters or Search Results


Contact:   researchexpo@soe.ucsd.edu   (858) 534-6068