On-the-fly Curve-skeleton Computation for 3D Shapes

The curve-skeleton of a 3D object is an abstract geometrical and topological representation of its 3D shape. It maps the spatial relation of geometrically meaningful parts to a graph structure. Each arc of this graph represents a part of the object with roughly constant diameter or thickness, and approximates its centerline. This makes the curve-skeleton suitable to describe and handle articulated objects such as characters for animation. We present an algorithm to extract such a skeleton on-the-fly, both from point clouds and polygonal meshes. The algorithm is based on a deformable model evolution that captures the object's volumetric shape. The deformable model involves multiple competing fronts which evolve inside the object in a coarse-to-fine manner. We first track these fronts' centers, and then merge and filter the resulting arcs to obtain a curve-skeleton of the object. The process inherits the robustness of the reconstruction technique, being able to cope with noisy input, intricate geometry and complex topology. It creates a natural segmentation of the object and computes a center curve for each segment while maintaining a full correspondence between the skeleton and the boundary of the object.     

全卷积语义分割与物体检测网络

目的 目前主流物体检测算法需要预先划定默认框,通过对默认框的筛选剔除得到物体框。为了保证足够的召回率,就必须要预设足够密集和多尺度的默认框,这就导致了图像中各个区域被重复检测,造成了极大的计算浪费。提出一种不需要划定默认框,实现完全端到端深度学习语义分割及物体检测的多任务深度学习模型（FCDN）,使得检测模型能够在保证精度的同时提高检测速度。方法 首先分析了被检测物体数量不可预知是目前主流物体检测算法需要预先划定默认框的原因,由于目前深度学习物体检测算法都是由图像分类模型拓展而来,被检测数量的无法预知导致无法设置检测模型的输出,为了保证召回率,必须要对足够密集和多尺度的默认框进行分类识别;物体检测任务需要物体的类别信息以实现对不同类物体的识别,也需要物体的边界信息以实现对各个物体的区分、定位;语义分割提取了丰富的物体类别信息,可以根据语义分割图识别物体的种类,同时采用语义分割的思想,设计模块提取图像中物体的边界关键点,结合语义分割图和边界关键点分布图,从而完成物体的识别和定位。结果 为了验证基于语义分割思想的物体检测方法的可行性,训练模型并在VOC（visual object classes）2007 test数据集上进行测试,与目前主流物体检测算法进行性能对比,结果表明,利用新模型可以同时实现语义分割和物体检测任务,在训练样本相同的条件下训练后,其物体检测精度优于经典的物体检测模型;在算法的运行速度上,相比于FCN,减少了8 ms,比较接近于YOLO（you only look once）等快速检测算法。结论 本文提出了一种新的物体检测思路,不再以图像分类为检测基础,不需要对预设的密集且多尺度的默认框进行分类识别;实验结果表明充分利用语义分割提取的丰富信息,根据语义分割图和边界关键点完成物体检测的方法是可行的,该方法避免了对图像的重复检测和计算浪费;同时通过减少语义分割预测的像素点数量来提高检测效率,并通过实验验证简化后的语义分割结果仍足够进行物体检测任务。     

Minimal surfaces based object segmentation

A geometric approach for 3D object segmentation and representation is presented. The segmentation is obtained by deformable surfaces moving towards the objects to be detected in the 3D image. The model is based on curvature motion and the computation of surfaces with minimal areas, better known as minimal surfaces. The space where the surfaces are computed is induced from the 3D image (volumetric data) in which the objects are to be detected. The model links between classical deformable surfaces obtained via energy minimization, and intrinsic ones derived from curvature based flows. The new approach is stable, robust, and automatically handles changes in the surface topology during the deformation     

IFTrace: Video segmentation of deformable objects using the Image Foresting Transform

We introduce IFTrace, a method for video segmentation of deformable objects. The algorithm makes minimal assumptions about the nature of the tracked object: basically, that it consists of a few connected regions, and has a well-defined border. The objects to be tracked are interactively segmented in the first frame of the video, and a set of markers is then automatically selected in the interior and immediate surroundings of the object. These markers are then located in the next frame by a combination of KLT feature finding and motion extrapolation. Object boundaries are then identified from these markers by the Image Foresting Transform (IFT). These steps are repeated for all subsequent frames until the end of the movie. Thanks to the IFT and a special boundary detection operator, IFTrace can reliably track deformable objects in the presence of partial and total occlusions, camera motion, lighting and color changes, and other complications. Tests on real videos show that the IFT is better suited to this task than Graph-Cut methods, and that IFTrace is more robust than other state-of-the art algorithms – namely, the OpenCV Snake and CamShift algorithms, Hess’s Particle-Filter, and Zhong and Chang’s method based on spatio-temporal consistency.     

A multiple object geometric deformable model for image segmentation

Deformable models are widely used for image segmentation, most commonly to find single objects within an image. Although several methods have been proposed to segment multiple objects using deformable models, substantial limitations in their utility remain. This paper presents a multiple object segmentation method using a novel and efficient object representation for both two and three dimensions. The new framework guarantees object relationships and topology, prevents overlaps and gaps, enables boundary-specific speeds, and has a computationally efficient evolution scheme that is largely independent of the number of objects. Maintaining object relationships and straightforward use of object-specific and boundary-specific smoothing and advection forces enables the segmentation of objects with multiple compartments, a critical capability in the parcellation of organs in medical imaging. Comparing the new framework with previous approaches shows its superior performance and scalability.     

The role of model-based segmentation in the recovery of volumetricparts from range data

We present a method for segmenting and estimating the shape of 3D objects from range data. The technique uses model views, or aspects, to constrain the fitting of deformable models to range data. Based on an initial region segmentation of a range image, regions are grouped into aspects corresponding to the volumetric parts that make up an object. The qualitative segmentation of the range image into a set of volumetric parts not only captures the coarse shape of the parts, but qualitatively encodes the orientation of each part through its aspect. Knowledge of a part's coarse shape, its orientation, as well as the mapping between the faces in its aspect and the surfaces on the part provides strong constraints on the fitting of a deformable model (supporting both global and local deformations) to the data. Unlike previous work in physics-based deformable model recovery from range data, the technique does not require presegmented data. Furthermore, occlusion is handled at segmentation time and does not complicate the fitting process, as only 3D points known to belong to a part participate in the fitting of a model to the part. We present the approach in detail and apply it to the recovery of objects from range data     

Silhouette-based occluded object recognition through curvature scale space

A complete and practical system for occluded object recognition has been developed which is very robust with respect to noise and local deformations of shape (due to weak perspective distortion, segmentation errors and non-rigid material) as well as scale, position and orientation changes of the objects. The system has been tested on a wide variety of free-form 3D objects. An industrial application is envisaged where a fixed camera and a light-box are utilized to obtain images. Within the constraints of the system, every rigid 3D object can be modeled by a limited number of classes of 2D contours corresponding to the object's resting positions on the light-box. The contours in each class are related to each other by a 2D similarity transformation. The Curvature Scale Space technique [26, 28] is then used to obtain a novel multi-scale segmentation of the image and the model contours. Object indexing [16, 32, 36] is used to narrow down the search space. An efficient local matching algorithm is utilized to select the best matching models. Received: 5 August 1996 / Accepted: 19 March 1997     

Shape-aware Volume Illustration

We introduce a novel volume illustration technique for regularly sampled volume datasets. The fundamental difference between previous volume illustration algorithms and ours is that our results are shape-aware, as they depend not only on the rendering styles, but also the shape styles. We propose a new data structure that is derived from the input volume and consists of a distance volume and a segmentation volume. The distance volume is used to reconstruct a continuous field around the object boundary, facilitating smooth illustrations of boundaries and silhouettes. The segmentation volume allows us to abstract or remove distracting details and noise, and apply different rendering styles to different objects and components. We also demonstrate how to modify the shape of illustrated objects using a new 2D curve analogy technique. This provides an interactive method for learning shape variations from 2D hand-painted illustrations by drawing several lines. Our experiments on several volume datasets demonstrate that the proposed approach can achieve visually appealing and shape-aware illustrations. The feedback from medical illustrators is quite encouraging.     

Dynamic learning from multiple examples for semantic object segmentation and search

We present a novel “dynamic learning” approach for an intelligent image database system to automatically improve object segmentation and labeling without user intervention, as new examples become available, for object-based indexing. The proposed approach is an extension of our earlier work on “learning by example,” which addressed labeling of similar objects in a set of database images based on a single example. The proposed dynamic learning procedure utilizes multiple example object templates to improve the accuracy of existing object segmentations and labels. Multiple example templates may be images of the same object from different viewing angles, or images of related objects. This paper also introduces a new shape similarity metric called normalized area of symmetric differences (NASD), which has desired properties for use in the proposed “dynamic learning” scheme, and is more robust against boundary noise that results from automatic image segmentation. Performance of the dynamic learning procedures has been demonstrated by experimental results.     

From 3D magnetic resonance images to structural representations of the cortex topography using topology preserving deformations

We propose an algorithm allowing the construction of a structural representation of the cortical topography from a T1-weighted 3D MR image. This representation is an attributed relational graph (ARG) inferred from the 3D skeleton of the object made up of the union of gray matter and cerebro-spinal fluid enclosed in the brain hull. In order to increase the robustness of the skeletonization, topological and regularization constraints are included in the segmentation process using an original method: the homotopically deformable regions. This method is halfway between deformable contour and Markovian segmentation approaches. The 3D skeleton is segmented in simple surfaces (SSs) constituting the ARG nodes (mainly cortical folds). The ARG relations are of two types: first, theSS pairs connected in the skeleton; second, theSS pairs delimiting a gyrus. The described algorithm has been developed in the frame of a project aiming at the automatic detection and recognition of the main cortical sulci. Indeed, the ARG is a synthetic representation of all the information required by the sulcus identification. This project will contribute to the development of new methodologies for human brain functional mapping and neurosurgery operation planning.     

融合边界信息的高分辨率遥感影像分割优化算法

目的 针对目前区域分割算法获取的区域边界与真实地物边界不一致问题,利用高分辨率遥感影像地物内具有均质性和地物间边缘信息突出的特点,提出一种融合边界信息的高分辨率遥感影像分割优化算法。方法 首先采用Canny算法对遥感影像进行边缘提取并进行边缘连接处理,产生闭合边界;然后将边界与初始分割结果进行融合处理,获得新的分割结果;最后在闭合边界约束下,基于灰度相似性准则对新的分割结果进行区域合并,获得优化后的最终分割结果。结果 采用本文提出的分割优化算法对Mean Shift算法和eCognition软件获得的分割结果进行优化处理,优化后的分割结果与初始分割结果相比正确分割率（RR）平均提高了4%,验证了本文算法的有效性。结论 该优化算法适用性广,可优化基于区域、基于边界和基于聚类等多种分割方法,同时该算法既能保持高分辨率遥感影像分割的区域完整性,又能保持地物边缘细节特征,提高了分割精度。     

FORMS: A flexible object recognition and modelling system

We describe a flexible object recognition and modelling system (FORMS) which represents and recognizes animate objects from their silhouettes. This consists of a model for generating the shapes of animate objects which gives a formalism for solving the inverse problem of object recognition. We model all objects at three levels of complexity: (i) the primitives, (ii) the mid-grained shapes, which are deformations of the primitives, and (iii) objects constructed by using a grammar to join mid-grained shapes together. The deformations of the primitives can be characterized by principal component analysis or modal analysis. When doing recognition the representations of these objects are obtained in a bottom-up manner from their silhouettes by a novel method for skeleton extraction and part segmentation based on deformable circles. These representations are then matched to a database of prototypical objects to obtain a set of candidate interpretations. These interpretations are verified in a top-down process. The system is demonstrated to be stable in the presence of noise, the absence of parts, the presence of additional parts, and considerable variations in articulation and viewpoint. Finally, we describe how such a representation scheme can be automatically learnt from examples.     

Strings: variational deformable models of multivariate continuous boundary features

We propose a new image segmentation technique called strings. A string is a variational deformable model that is learned from a collection of example objects rather than built from a priori analytical or geometrical knowledge. As opposed to existing approaches, an object boundary is represented by a one-dimensional multivariate curve in functional space, a feature function, rather than by a point in vector space. In the learning phase, feature functions are defined by extraction of multiple shape and image features along continuous object boundaries in a given learning set. The feature functions are aligned, then subjected to functional principal components analysis and functional principal regression to summarize the feature space and to model its content, respectively. Also, a Mahalanobis distance model is constructed for evaluation of boundaries in terms of their feature functions, taking into account the natural variations seen in the learning set. In the segmentation phase, an object boundary in a new image is searched for with help of a curve. The curve gives rise to a feature function, a string, that is weighted by the regression model and evaluated by the Mahalanobis model. The curve is deformed in an iterative procedure to produce feature functions with minimal Mahalanobis distance. Strings have been compared with active shape models on 145 vertebra images, showing that strings produce better results when initialized close to the target boundary, and comparable results otherwise.     

基于YOLOv5和FCN-DenseNet水下图像多目标语义分割算法

带视觉系统的水下机器人作业离不开对水下目标准确的分割, 但水下环境复杂, 场景感知精度和识别精度不高等问题会严重影响目标分割算法的性能. 针对此问题本文提出了一种综合YOLOv5和FCN-DenseNet的多目标分割算法. 本算法以FCN-DenseNet算法为主要分割框架, YOLOv5算法为目标检测框架. 采用YOLOv5算法检测出每个种类目标所在位置; 然后输入针对不同类别的FCN-DenseNet语义分割网络, 实现多分支单目标语义分割, 最后融合分割结果实现多目标语义分割. 此外, 本文在Kaggle竞赛平台上的海底图片数据集上将所提算法与PSPNet算法和FCN-DenseNet算法两种经典的语义分割算法进行了实验对比. 结果表明本文所提的多目标图像语义分割算法与PSPNet算法相比, 在MIoU和IoU指标上分别提高了14.9%和11.6%; 与FCN-DenseNet算法在MIoU和IoU指标上分别提高了8%和7.7%, 更适合于水下图像分割.     

Automatic Image Segmentation by Tree Pruning

The Image Foresting Transform (IFT) is a tool for the design of image processing operators based on connectivity, which reduces image processing problems into an optimum-path forest problem in a graph derived from the image. A new image operator is presented, which solves segmentation by pruning trees of the forest. An IFT is applied to create an optimum-path forest whose roots are seed pixels, selected inside a desired object. In this forest, object and background are connected by optimum paths (leaking paths), which cross the object’s boundary through its “most weakly connected” parts (leaking pixels). These leaking pixels are automatically identified and their subtrees are eliminated, such that the remaining forest defines the object. Tree pruning runs in linear time, is extensible to multidimensional images, is free of ad hoc parameters, and requires only internal seeds, with little interference from the heterogeneity of the background. These aspects favor solutions for automatic segmentation. We present a formal definition of the obtained objects, algorithms, sufficient conditions for tree pruning, and two applications involving automatic segmentation: 3D MR-image segmentation of the human brain and image segmentation of license plates. Given that its most competitive approach is the watershed transform by markers, we also include a comparative analysis between them.     

Motion Competition: A Variational Approach to Piecewise Parametric Motion Segmentation

We present a novel variational approach for segmenting the image plane into a set of regions of parametric motion on the basis of two consecutive frames from an image sequence. Our model is based on a conditional probability for the spatio-temporal image gradient, given a particular velocity model, and on a geometric prior on the estimated motion field favoring motion boundaries of minimal length.Exploiting the Bayesian framework, we derive a cost functional which depends on parametric motion models for each of a set of regions and on the boundary separating these regions. The resulting functional can be interpreted as an extension of the Mumford-Shah functional from intensity segmentation to motion segmentation. In contrast to most alternative approaches, the problems of segmentation and motion estimation are jointly solved by continuous minimization of a single functional. Minimizing this functional with respect to its dynamic variables results in an eigenvalue problem for the motion parameters and in a gradient descent evolution for the motion discontinuity set.We propose two different representations of this motion boundary: an explicit spline-based implementation which can be applied to the motion-based tracking of a single moving object, and an implicit multiphase level set implementation which allows for the segmentation of an arbitrary number of multiply connected moving objects.Numerical results both for simulated ground truth experiments and for real-world sequences demonstrate the capacity of our approach to segment objects based exclusively on their relative motion.