Detecting Temporally Consistent Objects in Videos through Object Class Label Propagation

TL;DR

This paper introduces OVERLAP, a novel method for efficient, temporally consistent object detection in videos using label propagation and clustering of object proposals, achieving state-of-the-art results.

Contribution

It presents a new VOP generation method and a streaming clustering approach that reduces computation by classifying only a small subset of proposals.

Findings

Achieves state-of-the-art detection performance on Youtube-Objects dataset.

Reduces computational load by classifying fewer proposals per frame.

Demonstrates effective temporal consistency through label propagation.

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. Source code will be made available soon.