Unsupervised Moving Object Detection via Contextual Information Separation
Preprint
Abstract
We propose an adversarial contextual model for detecting moving objects in images. A deep neural network is trained to predict the optical flow in a region using information from everywhere else but that region (context), while another network attempts to make such context as uninformative as possible. The result is a model where hypotheses naturally compete with no need for explicit regularization or hyper-parameter tuning. Although our method requires no supervision whatsoever, it outperforms several methods that are pre-trained on large annotated datasets. Our model can be thought of as a generalization of classical variational generative region-based segmentation, but in a way that avoids explicit regularization or solution of partial differential equations at run-time.Keywords
All Related Versions
This publication has 10 references indexed in Scilit:
- Unsupervised Moving Object Detection via Contextual Information Separation2019
- Video Object Segmentation without Temporal InformationIEEE Transactions on Pattern Analysis and Machine Intelligence, 2018
- DroNet: Learning to Fly by DrivingIEEE Robotics and Automation Letters, 2018
- Towards the quantitative evaluation of visual attention modelsVision Research, 2015
- Shape Tracking with Occlusions via Coarse-to-Fine Region-Based Sobolev DescentIEEE Transactions on Pattern Analysis and Machine Intelligence, 2014
- Motion Competition: A Variational Approach to Piecewise Parametric Motion SegmentationInternational Journal of Computer Vision, 2004
- Active contours without edgesIEEE Transactions on Image Processing, 2001
- A saliency-based search mechanism for overt and covert shifts of visual attentionVision Research, 2000
- Representing moving images with layersIEEE Transactions on Image Processing, 1994
- Optimal approximations by piecewise smooth functions and associated variational problemsCommunications on Pure and Applied Mathematics, 1989