Multi-Label Segmentation via Residual-Driven Adaptive Regularization

نویسندگان

  • Byung-Woo Hong
  • Ja-Keoung Koo
  • Stefano Soatto
چکیده

We present a variational multi-label segmentation algorithm based on a robust Huber loss for both the data and the regularizer, minimized within a convex optimization framework. We introduce a novel constraint on the common areas, to bias the solution towards mutually exclusive regions. We also propose a regularization scheme that is adapted to the spatial statistics of the residual at each iteration, resulting in a varying degree of regularization being applied as the algorithm proceeds: the effect of the regularizer is strongest at initialization, and wanes as the solution increasingly fits the data. This minimizes the bias induced by the regularizer at convergence. We design an efficient convex optimization algorithm based on the alternating direction method of multipliers using the equivalent relation between the Huber function and the proximal operator of the one-norm. We empirically validate our proposed algorithm on synthetic and real images and offer an information-theoretic derivation of the cost-function that highlights the modeling choices made.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Joint Classification-Regression Forests for Spatially Structured Multi-object Segmentation

In many segmentation scenarios, labeled images contain rich structural information about spatial arrangement and shapes of the objects. Integrating this rich information into supervised learning techniques is promising as it generates models which go beyond learning class association, only. This paper proposes a new supervised forest model for joint classification-regression which exploits both...

متن کامل

Combining Topological Maps, Multi-Label Simple Points, and Minimum-Length Polygons for Efficient Digital Partition Model

Deformable models have shown great potential for image segmentation. They include discrete models whose combinatorial formulation leads to efficient and sometimes optimal minimization algorithms. In this paper, we propose a new discrete framework to deform any partition while preserving its topology. We show how to combine the use of multilabel simple points, topological maps and minimum-length...

متن کامل

Graph Based Microscopic Images Semi and Unsupervised Classification and Segmentation

In this paper, we propose a general formulation of discrete functional regularization on weighted graphs. This framework can be used to on any multi-dimensional data living on graphs of arbitrary topologies. But, in this work, we focus on the microscopic image segmentation and classification with a semi and unsupervised schemes. Moreover, to provide a fast image segmentation we propose a graph ...

متن کامل

Modality Propagation: Coherent Synthesis of Subject-Specific Scans with Data-Driven Regularization

We propose a general database-driven framework for coherent synthesis of subject-specific scans of desired modality, which adopts and generalizes the patch-based label propagation (LP) strategy. While modality synthesis has received increased attention lately, current methods are mainly tailored to specific applications. On the other hand, the LP framework has been extremely successful for cert...

متن کامل

Multi-resolution graph-based analysis of histopathological whole slide images: Application to mitotic cell extraction and visualization

In this paper, we present a graph-based multi-resolution approach for mitosis extraction in breast cancer histological whole slide images. The proposed segmentation uses a multi-resolution approach which reproduces the slide examination done by a pathologist. Each resolution level is analyzed with a focus of attention resulting from a coarser resolution level analysis. At each resolution level,...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:
  • CoRR

دوره abs/1702.08336  شماره 

صفحات  -

تاریخ انتشار 2017