Abstract:This paper investigates a convex-relaxed kernel mapping formulation of image segmentation. We optimize, under some partition constraints, a functional containing two characteristic terms: 1) a data term, which maps the observation space to a higher (possibly infinite) dimensional feature space via a kernel function, thereby evaluating nonlinear distances between the observations and segments parameters and 2) a total-variation term, which favors smooth segment surfaces (or boundaries). The algorithm iterates two steps: 1) a convex-relaxation optimization with respect to the segments by solving an equivalent constrained problem via the augmented Lagrange multiplier method and 2) a convergent fixed-point optimization with respect to the segments parameters. The proposed algorithm can bear with a variety of image types without the need for complex and application-specific statistical modeling, while having the computational benefits of convex relaxation. Our solution is amenable to parallelized implementations on graphics processing units (GPUs) and extends easily to high dimensions. We evaluated the proposed algorithm with several sets of comprehensive experiments and comparisons, including: 1) computational evaluations over 3D medical-imaging examples and high-resolution large-size color photographs, which demonstrate that a parallelized implementation of the proposed method run on a GPU can bring a significant speed-up and 2) accuracy evaluations against five state-of-the-art methods over the Berkeley color-image database and a multimodel synthetic data set, which demonstrates competitive performances of the algorithm.

Convex Relaxation for a 3D Spatiotemporal Segmentation Model Using the Primal-Dual Method

Primal-dual method for spatiotemporal tracking model with moving background

Fast Global Segmentation Based on the Dual Formulation of TV-norm

Spatio-Temporal Video Segmentation of Static Scenes and Its Applications

Motion Segmentation Model Based on Total Variation and Split Bregman Algorithm

Motion Objects Segmentation Using a New Level Set Based Method

Spatio-Temporal Segmentation with Depth-Inferred Videos of Static Scenes

3D Video Based Segmentation and Motion Estimation with Active Surface Evolution

A Spatiotemporal Multiple Moving Objects Segmentation and Tracking with Level Set

PWP3D: Real-Time Segmentation and Tracking of 3D Objects

RGB-D Video Segmentation via Geodesic Spatio-Temporal Propagation

3D Object Segmentation Using Cross-Window Point Transformer with Latent Semantic Boundary Guidance

Convex-relaxed kernel mapping for image segmentation

Simultaneous multi-body stereo and segmentation

Spatiotemporal Strategies for Joint Segmentation and Motion Tracking from Cardiac Image Sequences

THREE-DIMENSIONAL ANIMATION COMPRESSION BASED ON SPACE-TIME SEGMENTATION

A variational approach for ego-motion estimation and segmentation based on 3D TOF camera

A Unified Spatiotemporal Prior based on Geodesic Distance for Video Object Segmentation

Video Object Segmentation with 3D Convolution Network

New Real-Time Algorithm for Multiview Disparity-Based 3D-Object Segmentation.

Video-object segmentation and 3D-trajectory estimation for monocular video sequences