Motion-based segmentation by principal singular vector (PSV) clustering method
نویسندگان
چکیده
MOTION-BASED SEGMENTATION BY PRINCIPAL SINGULAR VECTOR (PSV) CLUSTERING METHOD S.Y. Kung Yun-Ting Lin Yen-Kuang Chen Princeton University ABSTRACT Motion-based segmentation has recently drawn a lot of attentions. The task of identifying independent objects is called segmentation, which can use cues such as edge (boundary), color, texture, etc. Motion-based segmentation has a broad video application domain, for instance, video mosaic, object-based video coding, synthesized video, scene analysis, and object recognition. An approach based on Principal Singular Vectors (PSVs) of the image measurement matrix was proposed for separating independent moving objects[1]. After applying SVD (Singular Value Decomposition), feature blocks with di erent object-based motions tend to form separate clusters on the PSV space. Therefore, a frame can be divided into regions each with consistent motion. Our approach o ers several additional features: (1) A multi-candidate feature tracker is adopted. (2) Multiple frames are utilized to facilitate motion-based separation. (3) We would like to achieve not only accurate motion estimation, the object regions should also retain some neighborhood property (to save the bits for coding boundary). For this, a neighborhood sensitivity parameter is introduced. One application of motion-based segmentation is low-bit-rate video compression. In very low-bit-rate video coding, only motion vectors of nite regions and the region boundary (coded in prediction error) need to be transmitted. Yet simulations yield quite respectable compensated frames.
منابع مشابه
Spectral Clustering for Robust Motion Segmentation
In this paper, we propose a robust motion segmentation method based on the matrix factorization and subspace separation. We, first, mathematically prove that the shape interaction matrix can be derived using QR decomposition rather than Singular Value Decomposition(SVD). Using shape interaction matrix, we solve the motion segmentation problem using spectral graph clustering technique. We exploi...
متن کاملMotion Segmentation by Velocity Clustering with Estimation of Subspace Dimension
The performance of clustering based motion segmentation methods depends on the dimension of the subspace where the point trajectories are projected. This paper presents a strategy for estimating the best subspace dimension using a novel clustering error measure. For each obtained segmentation, the proposed measure estimates the average least square error between the point trajectories and synth...
متن کاملMulti-body Segmentation and Motion Number Estimation via Over-Segmentation Detection
This paper studies the problem of multi-body segmentation and motion number estimation. It is well known that motion number plays a critical role in the success of multi-body segmentation. Most of the existing methods exploit only motion affinity to segment and determine the number of motions. Motion number estimated in this way is often seriously affected by noise. In this paper, we recast the...
متن کاملA Principal Component Clustering Approach to Object-Oriented Motion Segmentation and Estimation
This paper presents a framework for object-oriented scene segmentation in video, which uses motion as the major characteristic to distinguish different moving objects and then to segment the scene into object regions. From the feature block (FB) correspondences through at least two frames obtained via a tracking algorithm, the reference feature measurement matrix and feature displacement matrix...
متن کاملLow Level Segmentation of Motion Capture Data based on Hierarchical Clustering with Cosine Distance
3D motion capture is to track and record human movements. In recent years, it has been applied into many fields, such as human computer interaction, animation, etc. Lowlevel segmentation of motion capture data is of significance to the various applications of 3D motion capture; however, due to the high dimensionality of motion capture data, traditional low-level segmentation methods can hardly ...
متن کامل