Spatio-Temporal Moving Object Proposals
By Katerina Fragkiadaki, Pablo Arbeláez, Panna Felsen, Jitendra Malik. 2014
11 citations recorded by the source corpus.
We present a method that segments moving objects in monocular uncalibrated videos using a combination of moving segment proposal generation and moving objectness ranking. We compute segment proposals in each frame by multiple figure-ground segmentations on optical flow boundaries, we call them Moving Object Proposals (MOPs). MOPs increase the object detection rate by 7\% over static segment proposals. MOPs are ranked by a Moving Objectness Detector (MOD) trained from image and motion fields using a multilayer convolutional network. The MOD discards over/under segmentations and focuses on the moving parts of the scene, but remains agnostic to the object categories contained in the video. Finally, the filtered 2D MOPs are extended into temporally coherent spatio-temporal tubes by label diffusion in a dense point trajectory embedding. Our results on the popular Moseg and VSB100 video benchmarks show the proposed method achieves much higher overlap with ground-truth objects using fewer tube proposals than previous works; further, it reaches levels of ground-truth coverage much beyond the maximum possible with existing supervoxel or trajectory clustering approaches.
Subfields: Computer Vision and Pattern Recognition
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computer Vision and Pattern Recognition
- Luc Van Gool
- Dacheng Tao
- Bernt Schiele
- Jitendra Malik
- Xiaogang Wang
- Andrew Zisserman
- Trevor Darrell
- Cordelia Schmid
Citations in the corpus
- What Makes for Effective Detection Proposals? (2015, background): “An earlier version of this work appeared in [17].”
- Object-Proposal Evaluation Protocol is ‘Gameable’ (2015, background): “Then average recall is measured over all the ground truth instances: where I [ · ] is an indicator function for the logical preposition in the argument.”
- Bilinear CNN Models for Fine-Grained Visual Recognition (2015, methodology): “Finally, a number of recent methods have proposed “two-stream” architectures for video where one networks models the temporal aspect using optical flow, while the other models the spatial aspect [12, 28].”
- SfM-Net: Learning of Structure and Motion from Video (2017, methodology): “See, for example, the last column of Figure 5 from [10], whose proposed methods for moving object proposals achieve IoU around 0.3 with four proposals.”