A projected gradient descent method for crf inference allowing end-to-end training of arbitrary pairwise potentials
Paper i proceeding, 2018

Are we using the right potential functions in the Conditional Random Field models that are popular in the Vision community? Semantic segmentation and other pixel-level labelling tasks have made significant progress recently due to the deep learning paradigm. However, most state-of-the-art structured prediction methods also include a random field model with a hand-crafted Gaussian potential to model spatial priors, label consistencies and feature-based image conditioning. In this paper, we challenge this view by developing a new inference and learning framework which can learn pairwise CRF potentials restricted only by their dependence on the image pixel values and the size of the support. Both standard spatial and high-dimensional bilateral kernels are considered. Our framework is based on the observation that CRF inference can be achieved via projected gradient descent and consequently, can easily be integrated in deep neural networks to allow for end-to-end training. It is empirically demonstrated that such learned potentials can improve segmentation accuracy and that certain label class interactions are indeed better modelled by a non-Gaussian potential. In addition, we compare our inference method to the commonly used mean-field algorithm. Our framework is evaluated on several public benchmarks for semantic segmentation with improved performance compared to previous state-of-the-art CNN+CRF models.

Convolutional neural networks

Segmentation

Conditional random fields

Författare

Måns Larsson

Chalmers, Elektroteknik, Signalbehandling och medicinsk teknik, Digitala bildsystem och bildanalys

Anurag Arnab

University of Oxford

Fredrik Kahl

Lunds universitet

Chalmers, Elektroteknik, Signalbehandling och medicinsk teknik, Digitala bildsystem och bildanalys

Shuai Zheng

University of Oxford

Philip H.S. Torr

University of Oxford

Lecture Notes in Computer Science

0302-9743 (ISSN)

Vol. 10746 564-579

11th International Conference on Energy Minimization Methods in Computer Vision and Pattern Recognition, EMMVCPR 2017
Venice, Italy,

Ämneskategorier

Annan data- och informationsvetenskap

Bioinformatik (beräkningsbiologi)

Datorseende och robotik (autonoma system)

DOI

10.1007/978-3-319-78199-0_37

Mer information

Senast uppdaterat

2018-04-19