Affine Motion Model Research Articles

Simultaneous motion estimation and filtering of image sequences

In this paper, we present an algorithm to simultaneously estimate multiple-frame motions and filter image sequences. The relative motion dk(x) between the reference frame s(x) and the kth frame s(x-dk(x)) in the presence of additive white Gaussian noise (AWGN) is estimated using the maximum-likelihood (ML) principle. The reference frame is also filtered in the linear minimum mean square error (LMMSE) sense during the process of motion estimation. Simulation experiments are performed using the affine motion model to illustrate the performance of this method.

A bee-inspired visual homing using color images

This paper presents a visual homing algorithm for autonomous robots inspired by the behaviour of bees and other social insects. The homing method presented is based on an affine motion model whose parameters are estimated by a best matching criterion. No attempts are made to recognize the objects or to extract 3D models from the scene. Improvements in the algorithm and in the use of colour information are introduced in order to enhance the efficiency of the navigation vector estimate. Tests and results are presented.

On the computation of warping-based motion compensation in video sequence coding

The paper illustrates a method for affine warping-based motion compensation using the computing architecture of H263 advanced prediction (AP) mode, namely, changing the (constant) weighting matrices in the H263 AP algorithm. In particular, we analytically derive a realization of affine image prediction based on a regular mesh in presence of linear resampling. The work shows, by means of analytical derivation and simulation results, an intrinsic similarity relationship between the affine motion model and the H.263 advanced prediction technique. The analysis offers a rational basis to explain the performance of H.263 AP as a coarse approximation of affine prediction.

Tube-based video coding

Abstract This paper presents a novel video compression strategy, based on structured video representation, to generate a content-addressable bit-stream supporting retrieval and composition. The structured video representation, extracted by the spatiotemporal segmentation algorithm, comprises a hierarchy of sequences, episodes, shots and motion events that constitute the building blocks of a scripted video stream. The main characteristics of this approach are that (i) it relies on the assumption that temporal redundancy can be efficiently exploited through temporal coherence on a tubewise basis, (ii) it controls the bit assignment according to motion complexity and temporal relations among video elements, (iii) it parameterizes the motion of the entire tube with an affine motion model, and (iv) it synthesizes images from representative texture patterns or anatomical models. The coding-decoding system is based on the construction of a specific video transformations — a tube code —, which, when interpolated and composed according to the temporal relations, produces a sequence of images that approximate the original. Coding efficiency is enhanced because structured video representation allows optimal reduction of temporal redundancy. We show how to design such a system for the coding of CIF-format color digital video ‘Miss America’ (30 frames/s) at rates of 10 and 37 kb/s, using 3-D face wireframe model and MC-DCT to encode textural changes, respectively. The structured video representation, once marked semantically, can facilitate interactive and content-based operations on image sequences, such as editing, browsing and content-based access and filtering.

A layered video object coding system using sprite and affine motion model

A layered video object coding system is presented in this paper. The goal is to improve video coding efficiency by exploiting the layering of video and to support content-based functionality. These two objectives are accomplished using a sprite technique and an affine motion model on a per-object basis. Several novel algorithms have been developed for mask processing and coding, trajectory coding, sprite accretion and coding, locally affine motion compensation, error signal suppression, and image padding. Compared with conventional frame-based coding methods, better experimental results on both hybrid and natural scenes have been obtained using our coding scheme. We also demonstrate content-based functionality which can be easily achieved in our system.

Two-stage motion compensation using adaptive global MC and local affine MC

This paper describes a high-efficiency video coding method based on ITU-T H.263. To improve the coding efficiency of H.263, a two-stage motion compensation (MC) method is proposed, consisting of global MC (GMC) for predicting camera motion and local MC (LMC) for macroblock prediction. First, global motion such as panning, tilting, and zooming is estimated, and the global-motion-compensated image is produced for use as a reference in LMC. Next, LMC is performed both for the global-motion-compensated reference image and for the image without GMC. LMC employs an affine motion model in the context of H.263's overlapped block motion compensation. Using the overlapped block affine MC, rotation and scaling of small objects can be predicted, in addition to translational motion. In the proposed method, GMC is adaptively turned on/off for each macroblock since GMC cannot be used for prediction in all regions in a frame. In addition, either an affine or a translational motion model is adaptively selected in LMC for each macroblock. Simulation results show that the proposed video coding technique using the two-stage MC significantly outperforms H.263 under identical conditions, especially for sequences with fast camera motion. The performance improvements in peak-to-peak SNR (PSNR) are about 3 dB over the original H.263, which does not use the two-stage MC.

Communication

The visual servoing formalism provides a framework for achieving tasks in an active closed-loop fashion based on visual information. We consider a typical visual servoing ap proach that uses geometric information about image features for controlling the position and attitude of a camera and extend the applicability of this approach by using image motion infor mation. We present two different approaches to visual tasks that use motion information. The first uses the focus of expansion, which is related to the translational velocity of the camera. The second incorporates the parameters of the 2D affine motion model in control equations. Both of these approaches are il lustrated by means of a task of aligning the optical axis of the camera with the unknown direction of translational motion of the system on which it is mounted. Results of simulation experi ments and real experiments on a six degrees of freedom (DOF) robot with a camera on its end effector are presented.

Closing the loop: detection and pursuit of a moving object by a moving observer

We present an integrated system which is able to pursue a moving object and maintain the object centered in the image by controlling a robot-head. The system contains three parts: independent motion detection, tracking, and control of a robot-head. This paper focuses on the detection mechanism, and briefly discusses the tracking and control issues. The system runs continuously in time and updates the object localization at a frame-rate of 25 Hz. The moving object can be tracked, although the observer performs an unknown independent motion, involving both translation and rotation. We focus on a simple motion detection algorithm, since computational cost is of major importance for real-time systems with feedback. The algorithm is noniterative and computationally inexpensive. The running time complexity is O ( n) for an image containing n pixels. The image-processing takes place on a MaxVideo 200 pipeline processor, and the head control algorithm is running on a Transputer T800 network. Some offline experimental results are presented, where comparisons are made between affine and translational image motion models.

特集 マルチメディア時代の画像処理 １画素マッチングによる誤差平面を利用した動き推定

We propose a novel framework for motion estimation based on pixel-wise matching. The proposed system consists of two steps. First, a distortion plane, which is a function of the displaced-frame difference of a single pixel, is determined and stored for each pixel. Second, motion is estimated based on a criterion which employs an operator working on the obtained distortion planes. The motion vector that gives the minimum (or maximum) value for the criterion is then estimated. The new framework is flexible in allowing selection of various motion models and operators. In this paper, we focus on a nonlinear weighted-summation operator and a majority-decision operator, in order to deal with discontinuities in the motion field. We also discuss a Hough-transform operator for an affine motion model. Experimental results show that the proposed methods allow us to estimate motion with high accuracy.

3-D motion estimation in model-based facial image coding

An approach to estimating the motion of the head and facial expressions in model-based facial image coding is presented. An affine nonrigid motion model is set up. The specific knowledge about facial shape and facial expression is formulated in this model in the form of parameters. A direct method of estimating the two-view motion parameters that is based on the affine method is discussed. Based on the reasonable assumption that the 3-D motion of the face is almost smooth in the time domain, several approaches to predicting the motion of the next frame are proposed. Using a 3-D model, the approach is characterized by a feedback loop connecting computer vision and computer graphics. Embedding the synthesis techniques into the analysis phase greatly improves the performance of motion estimation. Simulations with long image sequences of real-world scenes indicate that the method not only greatly reduces computational complexity but also substantially improves estimation accuracy. >

Motion segmentation and qualitative dynamic scene analysis from an image sequence

This article deals with analysis of the dynamic content of a scene from an image sequence irrespective of the static or dynamic nature of the camera. The tasks involved can be the detection of moving objects in a scene observed by a mobile camera, or the identification of the movements of some relevant components of the scene relatively to the camera. This problem basically requires a motion-based segmentation step. We present a motion-based segmentation method relying on 2-D affine motion models and a statistical regularization approach which ensures stable motion-based partitions. This can be done without the explicit estimation of optic flow fields. Besides these partitions are linked in time. Therefore, the motion interpretation process can be performed on more than two successive frames. The ability to follow a given coherently moving region within an interval of several images of the sequence makes the interpretation process more robust and more comprehensive. Identification of the kinematic components of the scene is induced from an intermediate layer accomplishing a generic qualitative motion labeling. No 3-D measurements are required. Results obtained on several real-image sequences corresponding to complex outdoor situations are reported.