Unsupervised Moving Object Detection via Contextual Information Separation
TL
TL;DR
An adversarial contextual model for detecting moving objects in images that 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.
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.
References42 items
1
Pyramid Dilated Deeper ConvLSTM for Video Salient Object Detection
2
Extending Layered Models to 3D Motion
3
Conditional Prior Networks for Optical Flow
4
CNN in MRF: Video Object Segmentation via Inference in a CNN-Based Higher-Order Spatio-Temporal MRF
5