Learning to segment moving objects in videos
Open Access
- 1 June 2015
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 10636919,p. 4083-4090
Abstract
We segment moving objects in videos by ranking spatio-temporal segment proposals according to “moving objectness”; how likely they are to contain a moving object. In each video frame, we compute segment proposals using multiple figure-ground segmentations on per frame motion boundaries. We rank them with a Moving Objectness Detector trained on image and motion fields to detect moving objects and discard over/under segmentations or background parts of the scene. We extend the top ranked segments into spatio-temporal tubes using random walkers on motion affinities of dense point trajectories. Our final tube ranking consistently outperforms previous segmentation methods in the two largest video segmentation benchmarks currently available, for any number of proposals. Further, our per frame moving object proposals increase the detection rate up to 7% over previous state-of-the-art static proposal methods.Keywords
All Related Versions
This publication has 25 references indexed in Scilit:
- Fast Object Segmentation in Unconstrained Video2013
- A Unified Video Segmentation Benchmark: Annotation, Metrics and Analysis2013
- Hierarchical Video Representation with Trajectory Binary Partition Tree2013
- Higher order motion models and spectral clustering2012
- Detection free tracking: Exploiting motion and topology for segmenting and tracking under entanglement2011
- Occlusion boundary detection and figure/ground assignment from optical flow2011
- Hough Forests for Object Detection, Tracking, and Action RecognitionIEEE Transactions on Pattern Analysis and Machine Intelligence, 2011
- Learning to Find Object Boundaries Using Motion Cues2007
- Multiclass spectral clustering2003
- Exploiting Discontinuities in Optical FlowInternational Journal of Computer Vision, 1998