Self-tuned Evolution-COnstructed features for general object recognition

Object recognition is a well studied but extremely challenging field. We present a novel approach to feature construction for object detection called Evolution-COnstructed Features (ECO features). Most current approaches rely on human experts to construct features for object recognition. ECO features are automatically constructed by uniquely employing a standard genetic algorithm to discover multiple series of transforms that are highly discriminative. Using ECO features provides several advantages over other object detection algorithms including: no need for a human expert to build feature sets or tune their parameters, ability to generate specialized feature sets for different objects, no limitations to certain types of image sources, and ability to find both global and local feature types. We show in our experiments that the ECO features compete well against state-of-the-art object recognition algorithms.     

Soft-assigned bag of features for object tracking

Multimedia Systems - Hard assignment-based bag of features (BoF) representation inevitably brings in quantization errors, which may lead to inaccuracy, even failure in object tracking. In this...     

Local and global Gabor features for object recognition

Invariant object recognition is one of the most central problems in computer vision. To be successful when occlusion and distortions are present, object recognition has to be based on local features. The features should express the significant information while being robust in the presence of noise and distortions, and stable in terms of feature parameters. In this study, Gabor filtering based features is analyzed in terms of the above requirements. Two classes of Gabor features are introduced: global Gabor features and fundamental frequency Gabor features. The Gabor filter response and stability issues are analyzed in terms of the filtering parameters. The robustness of the proposed features is examined through experiments. Both analytical and experimental results indicate that when certain conditions on the filter parameters are met, Gabor filtering can be reliably used in low-level feature extraction in image processing, and the filter responses can be used to construct robust invariant recognition systems.     

Off-the-shelf CNN features for 3D object retrieval

Multimedia Tools and Applications - Effective feature representation is crucial to view-based 3D object retrieval (V3OR). Most previous works employed hand-crafted features to represent the views...     

多尺度特征深度复用的显著性目标检测算法

针对传统显著性目标检测方法在检测不同尺度的多个显著性目标方面的不足,提出了一种多尺度特征深度复用的显著性目标检测算法,网络模型由垂直堆叠的双向密集特征聚合模块和水平堆叠的多分辨率语义互补模块组成。首先,双向密集特征聚合模块基于ResNet骨干网络提取不同分辨率语义特征;然后,依次在top-down和bottom-up两条通路上进行自适应融合,以获取不同层次多尺度表征特征;最后,通过多分辨率语义互补模块对两个相邻层次的多尺度特征进行融合,以消除不同层次上特征之间的相互串扰来增强预测结果的一致性。在五个基准数据集上进行的实验结果表明,该方法在Fmax、Sm、MAE最高能达到0.939、0.921、0.028,且检测速率可达74.6 fps,与其他对比算法相比有着更好的检测性能。     

Texture features for browsing and retrieval of image data

Image content based retrieval is emerging as an important research area with application to digital libraries and multimedia databases. The focus of this paper is on the image processing aspects and in particular using texture information for browsing and retrieval of large image data. We propose the use of Gabor wavelet features for texture analysis and provide a comprehensive experimental evaluation. Comparisons with other multiresolution texture features using the Brodatz texture database indicate that the Gabor features provide the best pattern retrieval accuracy. An application to browsing large air photos is illustrated     

Robust object matching for persistent tracking with heterogeneous features

This paper addresses the problem of matching vehicles across multiple sightings under variations in illumination and camera poses. Since multiple observations of a vehicle are separated in large temporal and/or spatial gaps, thus prohibiting the use of standard frame-to-frame data association, we employ features extracted over a sequence during one time interval as a vehicle fingerprint that is used to compute the likelihood that two or more sequence observations are from the same or different vehicles. Furthermore, since our domain is aerial video tracking, in order to deal with poor image quality and large resolution and quality variations, our approach employs robust alignment and match measures for different stages of vehicle matching. Most notably, we employ a heterogeneous collection of features such as lines, points, and regions in an integrated matching framework. Heterogeneous features are shown to be important. Line and point features provide accurate localization and are employed for robust alignment across disparate views. The challenges of change in pose, aspect, and appearances across two disparate observations are handled by combining a novel feature-based quasi-rigid alignment with flexible matching between two or more sequences. However, since lines and points are relatively sparse, they are not adequate to delineate the object and provide a comprehensive matching set that covers the complete object. Region features provide a high degree of coverage and are employed for continuous frames to provide a delineation of the vehicle region for subsequent generation of a match measure. Our approach reliably delineates objects by representing regions as robust blob features and matching multiple regions to multiple regions using Earth Mover's Distance (EMD). Extensive experimentation under a variety of real-world scenarios and over hundreds of thousands of Confirmatory Identification (CID) trails has demonstrated about 95 percent accuracy in vehicle reacquisition with both visible and Infrared (IR) imaging cameras.     

Sharing visual features for multiclass and multiview object detection

We consider the problem of detecting a large number of different classes of objects in cluttered scenes. Traditional approaches require applying a battery of different classifiers to the image, at multiple locations and scales. This can be slow and can require a lot of training data since each classifier requires the computation of many different image features. In particular, for independently trained detectors, the (runtime) computational complexity and the (training-time) sample complexity scale linearly with the number of classes to be detected. We present a multitask learning procedure, based on boosted decision stumps, that reduces the computational and sample complexity by finding common features that can be shared across the classes (and/or views). The detectors for each class are trained jointly, rather than independently. For a given performance level, the total number of features required and, therefore, the runtime cost of the classifier, is observed to scale approximately logarithmically with the number of classes. The features selected by joint training are generic edge-like features, whereas the features chosen by training each class separately tend to be more object-specific. The generic features generalize better and considerably reduce the computational cost of multiclass object detection     

Boosted cascade of scattered rectangle features for object detection

This paper presents a variant of Haar-like feature used in Viola and Jones detection framework,called scattered rectangle feature,based on the common-component analysis of local region feature. Three common components,feature filter,feature structure and feature form,are extracted without concern-ing the details of the studied region features,which cast a new light on region feature design for spe-cific applications and requirements: modifying some component(s) of a feature for an improved one or combining different components of existing features for a new favorable one. Scattered rectangle feature follows the former way,extending the feature structure component of Haar-like feature out of the restriction of the geometry adjacency rule,which results in a richer representation that explores much more orientations other than horizontal,vertical and diagonal,as well as misaligned,detached and non-rectangle shape information that is unreachable to Haar-like feature. The training result of the two face detectors in the experiments illustrates the benefits of scattered rectangle feature empirically; the comparison of the ROC curves under a rigid and objective detection criterion on MIT CMU upright face test set shows that the cascade based on scattered rectangle features outperforms that based on Haar-like features.     

Evaluating clustering quality using features salience: a promising approach

Neural Computing and Applications - This paper focuses on using feature salience to evaluate the quality of a partition when dealing with hard clustering. It is based on the hypothesis that a good...     

Three-dimensional object registration using wavelet features

Recent developments in shape-based modeling and data acquisition have brought three-dimensional models to the forefront of computer graphics and visualization research. New data acquisition methods are producing large numbers of models in a variety of fields. Three-dimensional registration (alignment) is key to the useful application of such models in areas from automated surface inspection to cancer detection and surgery. The algorithms developed in this research accomplish automatic registration of three-dimensional voxelized models. We employ features in a wavelet transform domain to accomplish registration. The features are extracted in a multi-resolutional format, thus delineating features at various scales for robust and rapid matching. Registration is achieved by using a voting scheme to select peaks in sets of rotation quaternions, then separately identifying translation. The method is robust to occlusion, clutter, and noise. The efficacy of the algorithm is demonstrated through examples from solid modeling and medical imaging applications.     

Enhanced SSD with interactive multi-scale attention features for object detection

Single Shot MultiBox Detector (SSD) method using multi-scale feature maps for object detection, showing outstanding performance in object detection task. However, as a one-stage detection method, it’s difficult for SSD methods to quickly notice significant areas of objects in the image. In the SSD network structure, feature maps of different scales are used to independently predict object, and there is a lack of interaction between low-level feature maps and high-level feature maps. In this paper we propose an enhanced SSD method using interactive multi-scale attention features (MA-SSD). Our method uses the attention mechanism to generate attention features of multiple scales and adds it to the original detection branch of the SSD method, which effectively enhances the feature representation ability and improves the detection accuracy. At the same time, the feature of different detection scales interacts with each other, and all the detection branches in our method have a parallel structure, which ensures the detection efficiency. Our proposed method achieves competitive performance on the public dataset PascalVOC.

     

Haar-like features with optimally weighted rectangles for rapid object detection

This article proposes an extension of Haar-like features for their use in rapid object detection systems. These features differ from the traditional ones in that their rectangles are assigned optimal weights so as to maximize their ability to discriminate objects from clutter (non-objects). These features maintain the simplicity of evaluation of the traditional formulation while being more discriminative. The proposed features were trained to detect two types of objects: human frontal faces and human heart regions. Our experimental results suggest that the object detectors based on the proposed features are more accurate and faster than the object detectors built with traditional Haar-like features.     

用于目标快速识别的特征选取方法

现有的基于图像局部特征的目标识别算法,在保证较高识别率的情况下无法满足实时性要求。针对这个问题,考虑到多数局部特征是不稳定、不可靠或与目标无关的,可通过正确匹配的训练图像,对图像局部特征选取一个子集用于目标识别。提出一种在特征包方法基础上,通过无监督地选取鲁棒性强及足够特殊、稳定的局部特征用于目标识别的新方法并应用于目标识别实验。实验结果证实该方法在仅仅使用原图像约4%的局部特征的情况下获得了与使用全部局部特征几乎相近的目标识别率,目标识别时间由秒缩短至几十毫秒,满足了目标识别实时性要求。     

Learning optimized features for hierarchical models of invariant object recognition

There is an ongoing debate over the capabilities of hierarchical neural feedforward architectures for performing real-world invariant object recognition. Although a variety of hierarchical models exists, appropriate supervised and unsupervised learning methods are still an issue of intense research. We propose a feedforward model for recognition that shares components like weight sharing, pooling stages, and competitive nonlinearities with earlier approaches but focuses on new methods for learning optimal feature-detecting cells in intermediate stages of the hierarchical network. We show that principles of sparse coding, which were previously mostly applied to the initial feature detection stages, can also be employed to obtain optimized intermediate complex features. We suggest a new approach to optimize the learning of sparse features under the constraints of a weight-sharing or convolutional architecture that uses pooling operations to achieve gradual invariance in the feature hierarchy. The approach explicitly enforces symmetry constraints like translation invariance on the feature set. This leads to a dimension reduction in the search space of optimal features and allows determining more efficiently the basis representatives, which achieve a sparse decomposition of the input. We analyze the quality of the learned feature representation by investigating the recognition performance of the resulting hierarchical network on object and face databases. We show that a hierarchy with features learned on a single object data set can also be applied to face recognition without parameter changes and is competitive with other recent machine learning recognition approaches. To investigate the effect of the interplay between sparse coding and processing nonlinearities, we also consider alternative feedforward pooling nonlinearities such as presynaptic maximum selection and sum-of-squares integration. The comparison shows that a combination of strong competitive nonlinearities with sparse coding offers the best recognition performance in the difficult scenario of segmentation-free recognition in cluttered surround. We demonstrate that for both learning and recognition, a precise segmentation of the objects is not necessary.     

Spatial and semantic convolutional features for robust visual object tracking

Multimedia Tools and Applications - Robust and accurate visual tracking is a challenging problem in computer vision. In this paper, we exploit spatial and semantic convolutional features extracted...     

Moving object detection for unseen videos via truncated weighted robust principal component analysis and salience convolution neural network

Multimedia Tools and Applications - Moving object detection is a basic and important work in intelligent video analysis. Recently, a lot of methods have sprung up. Among them, the methods based on...     

基于互补特征的纹理图像检索

针对互补特征可以有效地改善图像检索系统性能的特点,提出一种在改进Contourlet变换域采用L1能量与广义高斯分布参数特征的纹理图像检索方法。首先,应用改进的方法对方向子带系数进行广义高斯统计建模。然后,分别单独利用各个特征和相应的相似性测度进行检索。最后,基于直接的相似性测度和,采用这两种互补的特征进行检索。实验结果表明,和采用单一特征相比较,互补特征由于充分地反映了图像的结构信息和随机分布信息,从而有效地提高了纹理图像数据库的平均检索率。