Interactive object registration and recognition for augmented desk interface

In this paper, an object recognition method and a pose estimation approach using stereo vision is presented. The proposed approach was used for position based visual servoing of a 6 DoF manipulator. The object detection and recognition method was designed with the purpose of increasing robustness. A RGB color-based object descriptor and an online correction method is proposed for object detection and recognition. Pose was estimated by using the depth information derived from stereo vision camera and an SVD based method. Transformation between the desired pose and object pose was calculated and later used for position based visual servoing. Experiments were carried out to verify the proposed approach for object recognition. The stereo camera was also tested to see whether the depth accuracy is adequate. The proposed object recognition method is invariant to scale, orientation and lighting condition which increases the level of robustness. The accuracy of stereo vision camera can reach 1 mm. The accuracy is adequate for tasks such as grasping and manipulation.

A robot vision system can be called humanoid if it possesses an oculomotor system similar to human eyes and if it is capable to simultaneously acquire and process images of varying resolution. Designers of a number of humanoid robots attempted to mimic the foveated structure of the human eye. Foveation is useful because, firstly, it enables the robot to monitor and explore its surroundings in images of low resolution, thereby increasing the efficiency of the search process, and secondly, it makes it possible to simultaneously extract additional information – once the area of interest is determined – from higher resolution foveal images that contain more detail. There are several visual tasks that can benefit from foveated vision. One of the most prominent among them is object recognition. General object recognition on a humanoid robot is difficult because it requires the robot to detect objects in dynamic environments and to control the eye gaze to get the objects into the fovea and to keep them there. Once these tasks are accomplished, the robot can determine the identity of the object by processing foveal views. Approaches proposed to mimic the foveated structure of biological vision systems include the use of two cameras per eye (Atkeson et al., 2000; Breazeal et al., 2001; Kozima & Yano, 2001; Scassellati, 1998) (Cog, DB, Infanoid, Kismet, respectively), i. e. a narrow-angle foveal camera and a wide-angle camera for peripheral vision; lenses with space-variant resolution (Rougeaux & Kuniyoshi, 1998) (humanoid head ESCHeR), i. e. a very high definition area in the fovea and a coarse resolution in the periphery; and space-variant log-polar sensors with retina-like distribution of photo-receptors (Sandini &Metta, 2003) (Babybot). It is also possible to implement log-polar sensors by transforming standard images into log-polar ones (Engel et al., 1994), but this approach requires the use of high definition cameras to get the benefit of varying resolution. Systems with zoom lenses have some of the advantages of foveated vision, but cannot simultaneously acquire wide angle and high resolution images. Our work follows the first approach (see Fig. 1) and explores the advantage of foveated vision for object recognition over standard approaches, which use equal resolution across the visual field. While log-polar sensors are a closer match to biology, we note that using two cameras per eye can be advantageous because cameras with standard chips can be utilized. This makes it possible to equip a humanoid robot with miniature cameras (lipstick size and 1

Object recognition is a typical task of aerial reconnaissance and especially in military applications, to determine the class of an unknown object on the battlefield can give valuable information on its capabilities and its threat. RecceMan (Reconnaissance Manual) is a decision support system for object recognition developed by the Fraunhofer IOSB. It supports object recognition by automating the tedious task of matching the object features with the set of possible object classes, while leaving the assessment of features to the trained human interpreter. The quality of the features assessed by the user is influenced by several factors such as the quality of the image of the object. These factors are potential sources of error, which can lead to an incorrect classification and therefore have to be considered by the system. To address this issue, two methods for consideration of uncertainty in human feature assessment - a probabilistic and a heuristic approach - are presented and compared based on an experiment in the exemplary domain of flower recognition.

An effective human-robot interaction is essential for wide penetration of service robots into the market. Such robot needs a vision system to recognize objects. It is, however, difficult to realize vision systems that can work in various conditions. More robust techniques of object recognition and image segmentation are essential. Thus, we have proposed to use the human user's assistance for objects recognition through speech. This paper presents a system that recognizes objects in occlusion and/or multicolor cases using geometric and photometric analysis of images. Based on the analysis results, the system makes a hypothesis of the scene. Then, it asks the user for confirmation by describing the hypothesis. If the hypothesis is not correct, the system generates another hypothesis until it correctly understands the scene. Through experiments on a real mobile robot, we have confirmed the usefulness of the system.

In picture processing, picture segmentation, object classification and recognition is an impressive concept in present days because of increasing computer vision applications in real time applications. Based on required application, divide a picture into particular relevant clusters based on equivalent partitioning. Object identification is a minimum work to identify objects in a given picture or video with pixel values. The main representation of object classification from a given picture is to detect picture instances with respect to categorization of object. Object recognition is a complex task, when the pictures are poor quality, occlusion, and noise and in picture with background cutter sequences, this becomes more and efficient challenging task in picture processing. To appear object categorization and recognition effectively in pictures, there are several methods and approaches were introduced to solve object representation effectively. In this paper we review the different techniques of contextual feature extraction with respect to picture information of object classification, categorization and recognition of scalability and optimizations in real time picture processing applications. Our research gives better review of different approaches in picture categorization.

<p>In this work, the focus was on the YOLO model and the features of its work. Tests and comparisons with other state-of-the-art were given, which show that YOLO is the fastest model for real time objects recognition today. The architecture of the algorithm and the essential components on which the field experiment was conducted are described. Based on the results obtained, it was concluded that the YOLOv4 recognition model shows excellent results in terms of accuracy and speed</p>

Object identification in pictures, videos, and signal processing is not a big frontier and has been around for a few years. Though object recognition in static pictures has proven highly promising in terms of specific items such as facial recognition systems, illness diagnosis, and so on, it has been tricky when it comes to video processing and real-time image processing. Traditional object identification techniques are coupled with machine learning methodologies to improve algorithms' speed and accuracy. The paper focuses on real-time object recognition systems and how recent advances in the realm of object recognition and identification have been made while bearing in mind the real-time scenarios for recognizing varied objects.The paper discusses the current implementations illustrated by varied authors using far more widely used algorithms including YOLO for real-time visualizations. Keywords: Object recognition, YOLO, Machine learning, RCNN, faster RCNN

The field of computer vision has undergone rapid development. Starting from recognition tasks such as classification, detection, and segmentation, the focus of visual analysis has gradually shifted towards learning spatiotemporal information. This thesis presents research on spatiotemporal learning, with a particular emphasis on motion estimation and visual recognition. First, we address the problem of video object tracking. Previous methods have primarily re-lied on learning improved appearance representations, while the spatiotemporal relationships of individual objects have been underexplored. We propose leveraging optical flow features to achieve higher generalization in semi-supervised video object segmentation, directly incorporating these features into both the target representation and the decoder network. Our experiments and analysis show that enriching feature representations with spatiotemporal information improves segmentation quality and generalization capability. Next, we investigate spatiotemporal learning in 3D for motion estimation, specifically scene flow estimation. Scene flow estimation as an important research topic in 3D computer vision is crucial for applications such as robotics, autonomous driving, embodied navigation, and tracking. We investigate the problem in different perspectives: 1. What is the best formulation for solving the problem and how to learn a better spatiotemporal feature representation? 2. Can we introduce uncertainty estimation to the task, which is of crucial importance for safety-critical downstream tasks? 3. How to scale the estimation to large-scale data, e.g., autonomous scenes, and leverage the temporal information without introducing much computation overheads? To answer these questions, we explore the use of transformers for improved feature representation, diffusion models for uncertainty estimation, and efficient feature learning methods for multi-frame, large-scale autonomous driving scenarios. Finally, we extend our research to joint visual segmentation, tracking, and open-vocabulary recognition in LiDAR sequences, particularly for autonomous scenes. In such environments, precise segmentation, tracking, and recognition of objects are essential for downstream analysis and control. Current human-annotated open-source datasets allow for reasonable tracking of traffic participants such as cars and pedestrians. However, we aim to advance beyond this towards segmenting and tracking any object in LiDAR data. To this end, we propose a pseudo-labeling engine that leverages the 2D visual foundation model SAM v2 and the vision-language model CLIP to automatically label LiDAR streams. We further introduce the SAL-4D model, capable of segmenting, tracking, and recognizing any object in a zero-shot manner. In summary, we explore the learning of spatiotemporal information in both 2D image and 3D point cloud domains. In the image domain, we demonstrate that spatiotemporal information improves video object segmentation quality and generalization. In the 3D point cloud domain, we show that spatiotemporal learning enables more accurate motion estimation and facilitates the first method for zero-shot segmentation, tracking, and open-vocabulary recognition of arbitrary objects.

There are many research efforts in object recognition. Most existing methods for object recognition are based on full objects. However, many images contain multiple objects with occluded shapes and regions. Due to the occlusion of objects, image retrieval can provide incomplete, uncertain, and inaccurate results. To resolve this problem, we propose a new method to reconstruct objects using symmetry properties since most objects in a given image database are represented by symmetrical figures. Even though there have been several efforts in object recognition with occlusion, current methods have been highly sensitive to object pose, rotation, scaling, and visible portion of occluded objects. In addition, many appearance-based and model-based object recognition methods assumed that they have known occluded regions of objects or images through extensive training processes with statistical approach. However, our new approach is not limited to recognizing occluded objects by pose and scale changes, and does not need extensive training processes. Unlike existing methods, the proposed method finds shapes and regions to reconstruct occluded shapes and regions within objects. We assume that we only consider the elliptical objects in recognition. The proposed approach can handle object rotation and scaling for dealing with occlusion, and does not require extensive training processes. The main advantage of our proposed approach is that it becomes simple to reconstruct objects from occlusions using symmetry. We present a robust method, which is based on the contours of objects, for recognizing partially occluded objects based on symmetry properties. The contour-based approach finds a symmetry axis using the maximum diameter from the occluded object. In experiments, we demonstrate how a proposed method reconstructs and recognizes occluded shapes and regions using symmetry. Experiments use rotated and scaled objects for dealing with occlusion. We use mirror symmetry to find possible occluded regions in objects. Examples of partially occluded objects are shown in Figure 1.1. We also evaluate the recognition rate of the reconstructed objects using symmetry and the visible portion of the occluded objects for recognition. The method produces average recognition rates for cups and plates above 88% with 30% occlusion. In this case, part of the

Most previous methods for generic object recognition explicitly or implicitly assume that an image contains objects from a single category, although objects from multiple categories often appear together in an image.In this paper, we present a novel method for object recognition that explicitly deals with objects of multiple categories coexisting in an image.Furthermore, our proposed method aims to recognize objects by taking advantage of a scene's context represented by the co-occurrence relationship between object categories.Specifically, our method estimates the mixture ratios of multiple categories in an image via MAP regression, where the likelihood is computed based on the linear combination model of frequency distributions of local features, and the prior probability is computed from the co-occurrence relation.We conducted a number of experiments using the PASCAL dataset, and obtained the results that lend support to the effectiveness of the proposed method.

Detection, classification and identification of underwater objects are both complex and essential tasks for perception and navigation of Autonomous Underwater Vehicles (AUVs) and Remotely Operated Vehicles (ROVs). AUVs need to detect obstacles to avoid them and need to identify certain objects in tasks such as Automatic Target Recognition (ATR) or in manipulation tasks in the case of ROVs. Leveraging a dataset published by the Robotics for Asset Maintenance and Inspection (RAMI) competition for the 2023 Virtual competition, in this work, we detail our machine learning-based solution for the visual detection, classification and identification of several objects including buoys of different colours, markers with numbers, pipes and ring poles among others. The results show excellent performance in training dataset and very good performance for detection and classification in the unknown test dataset showing the goodness of the methods in harsh environments with low visibility. Identification remains a challenging task especially for not previously seen data and requires further work including synthethic data generation.

It is an important task for a robot to bring objects requested by human via voice. In order to achieve the task, object recognition using speech and images is needed. Ozasa et al. has proposed the method for the object recognition by integrating speech and image information. Although this method requires both speech (word) and image models, the speech models are automatically constructed by combining phonemic acoustic models according to the dictionary. However, the image models have to be constructed manually in advance. In this paper, instead of the manual construction of the image models, we propose an automatic image model construction method for object recognition using Web images. The effectiveness of the proposed method is verified in the object recognition by integrating speech and image information.

A Review of Recent Advancements in Appearance-based Object Recognition

In this paper, we address object recognition for a mobile robot which is deployed in a multistory building. To move to another floor, a mobile robot should recognize various objects related to an elevator, e.g., elevator control, call buttons, and LED displays. To this end, we propose a neural network based retrainable framework for object recognition, which consists of four components--preprocessing, binary classification, object identification, and outlier rejection. The binary classifier, a key component of our system, is a neural network that can be retrained, the motivation of which is to adapt to varying environments, especially with illuminations. Without incurring any extra process to prepare new training samples for retraining, they are freely obtained as a result of the outlier rejection component, being extracted on-line. To realize a practical system, we adopt a parallel architecture integrating both recognition and retraining processes for seamless object recognition, and furthermore detect and cope with the deterioration of a retrained neural network to ensure high reliability. We demonstrate the positive effect of retraining on the object recognition performance by conducting experiments over hundreds of images obtained in daytime and nighttime.

Automatization for the picking and placing of a variety of objects stored on shelves is a challenging problem for robotic picking systems in distribution warehouses. Here, object recognition using image processing is especially effective at picking and placing a variety of objects. In this study, we propose an efficient method for object recognition based on object grasping position for picking robots. We use a convolutional neural network (CNN) that can achieve highly accurate object recognition. In typical CNN methods for object recognition, objects are recognized by using an image containing picking targets from which object regions suitable for grasping can be detected. However, these methods increase the computational cost because a large number of weight filters are convoluted with the whole image. The proposed method detects all graspable positions from an image as a first step. In the next step, it classifies an optimal grasping position by feeding an image of the local region at the grasping point to the CNN. By recognizing the grasping positions of the objects first, the computational cost is reduced because of the fewer convolutions of the CNN.