Jia Liu · 2023
<p>In this paper, we define the freestyle object localization problem where a model is expected to autonomically learn and locate an arbitrarily specified object given only the name. However, learning the semantic features of an object requires corresponding training data which are not always available in existing datasets. Obviously, Internet provides massive off-the-rack data for learning given only the keyword while few existing methods are able to directly use them. To directly use them, a localization model should firstly overcome two difficulties: (1) learning from single-class annotated images and (2) from images with unpredictable sizes. For single-class annotation, we establish a novel probabilistic model to capture the distribution of the object in the background data space. Based on a deep hierarchical architecture with alterable depth that is designed for unpredictable sizes, we propose an association module and design an energy function to drive the probabilistic model. By maximizing the log-likelihood, the proposed model is able to highlight the semantic features of object that has the highest probability in training images, i.e., the specified object. In experiments, the proposed model achieves state-of-the-art performance among weakly-supervised localization methods on public datasets and freestyle localization of objects that are rare in public datasets.</p>
Cited by 0Open access
Through Sensor Fusion, Pedro Marcelino, Pedro Nunes et al. · 2007
This paper introduces a method for representing, communicating and fusing distributed, noisy and uncertain observations of objects by multiple robots.
Cited by 0
Kate Sullivan, Kate Sullivan, Organisation Name · 1999
Cited by 0
Jia Liu · 2023
<p>In this paper, we define the freestyle object localization problem where a model is expected to autonomically learn and locate an arbitrarily specified object given only the name. However, learning the semantic features of an object requires corresponding training data which are not always available in existing datasets. Obviously, Internet provides massive off-the-rack data for learning given only the keyword while few existing methods are able to directly use them. To directly use them, a localization model should firstly overcome two difficulties: (1) learning from single-class annotated images and (2) from images with unpredictable sizes. For single-class annotation, we establish a novel probabilistic model to capture the distribution of the object in the background data space. Based on a deep hierarchical architecture with alterable depth that is designed for unpredictable sizes, we propose an association module and design an energy function to drive the probabilistic model. By maximizing the log-likelihood, the proposed model is able to highlight the semantic features of object that has the highest probability in training images, i.e., the specified object. In experiments, the proposed model achieves state-of-the-art performance among weakly-supervised localization methods on public datasets and freestyle localization of objects that are rare in public datasets.</p>
Cited by 0Open access
Using Fuzzy Logic, J. P. Canovas, K. Leblanc et al. · 2007
Cited by 0
Rafał Staszak, Dominik Belter · 2022 17th International Conference on Control, Automation, Robotics and Vision (ICARCV) · 2022
In this research, we deal with the problem of estimating 3D object positions based on 2D localization data and 2D object bounding boxes determined on the RGB images by a CNN object detector. We use a mobile-manipulating robot equipped with an RGB-D camera and a 2D laser scanner. In contrast to other methods, which are based either on end-to-end neural networks or machine learning solutions, we propose an approach that allows estimating 3D object positions from a sequence of images and 2D robot poses obtained from an onboard 2D localization system. We determine a set of lines from the robot localization data and the object detections, which define the observation directions. The closest point to a set of lines determines an approximate object location. We define the 3D object position estimation as an optimization problem and solve using efficient GPU implementation. Finally, we present results based on data collected on the real robot in an unstructured environment.
Cited by 1
Rydlo, Štěpán · Brno University of Technology Digital Library (Brno University of Technology) · 2017
Tato práce se zaobírá zpracování radiového signálu a určení polohy přijímače na počítačích s využitím sofrwarově definovaného radia. Cílem této práce je provést paralelizaci programu tak, aby zpracování informací probíhalo v reálném čase. Pro dosažení zpracování v reálném čase bude paralelizace programu probíhat i na grafických procesorech. Obsahem práce je popis dvou na sobě nezávislých možnosti přenosu dat a jeho zpracování.
Cited by 0
Ehud Ahissar, Per Magne Knutsen · Biological Cybernetics · 2008
Cited by 98
Yunhan Ren, Bo Li, Chengyang Zhang et al. · arXiv (Cornell University) · 2024
Existing object localization methods are tailored to locate specific classes of objects, relying heavily on abundant labeled data for model optimization. However, acquiring large amounts of labeled data is challenging in many real-world scenarios, significantly limiting the broader application of localization models. To bridge this research gap, this paper defines a novel task named Few-Shot Object Localization (FSOL), which aims to achieve precise localization with limited samples. This task achieves generalized object localization by leveraging a small number of labeled support samples to query the positional information of objects within corresponding images. To advance this field, we design an innovative high-performance baseline model. This model integrates a dual-path feature augmentation module to enhance shape association and gradient differences between supports and query images, alongside a self query module to explore the association between feature maps and query images. Experimental results demonstrate a significant performance improvement of our approach in the FSOL task, establishing an efficient benchmark for further research. All codes and data are available at https://github.com/Ryh1218/FSOL.
Cited by 0Open access
Alexei Borissov, Jakob Janecek, Alexei Borissov et al. · 2015
Autonomous object localization and identification is a challenging task. At present time, a number of solutions exist to deal with this problem. Although many solutions are feasible, they tend to focus on the idea of a ‘super-robot’ that performs the entire task by itself using a myriad of approaches for object identification, utilising knowledge from the entire spectrum of image process-ing. This thesis presents a novel approach to object localization and identifica-tion. Using a distributed system within an intelligent environment, we propose a two-step process combining fixed vision, provided by environment cameras, and mobile vision, provided by a camera fixed on a mobile robot platform. The first step is responsible for providing a candidate list with position estimates for each object in the environment. Using this list, the second step, employing the robot and its camera, will then attempt to find the object at that location. Our novel, two-step approach not only reduces the time required by the robot
Cited by 0
Gerald Schweighofer, Andreas Opelt, Axel Pinz · 2004
Cited by 3
Rui Xu, Yong Luo, Han Hu et al. · arXiv (Cornell University) · 2023
Weakly supervised object localization (WSOL) is one of the most popular and challenging tasks in computer vision. This task is to localize the objects in the images given only the image-level supervision. Recently, dividing WSOL into two parts (class-agnostic object localization and object classification) has become the state-of-the-art pipeline for this task. However, existing solutions under this pipeline usually suffer from the following drawbacks: 1) they are not flexible since they can only localize one object for each image due to the adopted single-class regression (SCR) for localization; 2) the generated pseudo bounding boxes may be noisy, but the negative impact of such noise is not well addressed. To remedy these drawbacks, we first propose to replace SCR with a binary-class detector (BCD) for localizing multiple objects, where the detector is trained by discriminating the foreground and background. Then we design a weighted entropy (WE) loss using the unlabeled data to reduce the negative impact of noisy bounding boxes. Extensive experiments on the popular CUB-200-2011 and ImageNet-1K datasets demonstrate the effectiveness of our method.
Cited by 0Open access
Oh, Seong Joon · Universitätsbibliothek Tübingen · 2025
Cited by 0Open access
Dongjun Hwang, Seong Joon Oh, Junsuk Choe · Neurocomputing · 2025
Cited by 0
Dongjun Hwang, Seong Joon Oh, Junsuk Choe · SSRN Electronic Journal · 2024
Cited by 0Open access
Dongjun Hwang, Seong Joon Oh, Junsuk Choe · SSRN Electronic Journal · 2024
Cited by 0Open access
Zhaohui Zheng, Rongguang Ye, Qibin Hou et al. · IEEE Transactions on Pattern Analysis and Machine Intelligence · 2023
Previous knowledge distillation (KD) methods for object detection mostly focus on feature imitation instead of mimicking the prediction logits due to its inefficiency in distilling the localization information. In this paper, we investigate whether logit mimicking always lags behind feature imitation. Towards this goal, we first present a novel localization distillation (LD) method which can efficiently transfer the localization knowledge from the teacher to the student. Second, we introduce the concept of valuable localization region that can aid to selectively distill the classification and localization knowledge for a certain region. Combining these two new components, for the first time, we show that logit mimicking can outperform feature imitation and the absence of localization distillation is a critical reason for why logit mimicking under-performs for years. The thorough studies exhibit the great potential of logit mimicking that can significantly alleviate the localization ambiguity, learn robust feature representation, and ease the training difficulty in the early stage. We also provide the theoretical connection between the proposed LD and the classification KD, that they share the equivalent optimization effect. Our distillation scheme is simple as well as effective and can be easily applied to both dense horizontal object detectors and rotated object detectors. Extensive experiments on the MS COCO, PASCAL VOC, and DOTA benchmarks demonstrate that our method can achieve considerable AP improvement without any sacrifice on the inference speed. Our source code and pretrained models are publicly available at https://github.com/HikariTJU/LD.
Cited by 71
A.S. Wallack, John Canny, Dinesh Manocha · 2002
A localization technique using reliable, highly accurate, robust, inexpensive light beam sensors designed for mechanical assembly and manufacturing is presented. The technique involves passing polyhedral objects through a set of crossed beams and storing the positions when any beam is broken or reconnects. The geometrical constraints simplify the correspondence problem, so that it is solvable in linear time or can be solved using a precomputed hash table. The global minimum least-squared error pose is computed exactly using methods from algebra and numerical analysis. The limitations of this technique result from the beam sensor's limitations.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">></ETX>
Cited by 27
Aditay Tripathi, Rajath R Dani, Anand Mishra et al. · arXiv (Cornell University) · 2022
Consider a scenario in one-shot query-guided object localization where neither an image of the object nor the object category name is available as a query. In such a scenario, a hand-drawn sketch of the object could be a choice for a query. However, hand-drawn crude sketches alone, when used as queries, might be ambiguous for object localization, e.g., a sketch of a laptop could be confused for a sofa. On the other hand, a linguistic definition of the category, e.g., a small portable computer small enough to use in your lap" along with the sketch query, gives better visual and semantic cues for object localization. In this work, we present a multimodal query-guided object localization approach under the challenging open-set setting. In particular, we use queries from two modalities, namely, hand-drawn sketch and description of the object (also known as gloss), to perform object localization. Multimodal query-guided object localization is a challenging task, especially when a large domain gap exists between the queries and the natural images, as well as due to the challenge of combining the complementary and minimal information present across the queries. For example, hand-drawn crude sketches contain abstract shape information of an object, while the text descriptions often capture partial semantic information about a given object category. To address the aforementioned challenges, we present a novel cross-modal attention scheme that guides the region proposal network to generate object proposals relevant to the input queries and a novel orthogonal projection-based proposal scoring technique that scores each proposal with respect to the queries, thereby yielding the final localization results. ...
Cited by 0Open access
Ehud Ahissar, Per Magne Knutsen · 2007
Abstract Rats use their large facial hairs (whiskers) to detect, localize and identify objects in their proximal threedimensional (3D) space. Here, we focus on recent evidence of how object location is encoded in the neural sensory pathways of the rat whisker system. Behavioral and neuronal observations have recently converged to the point where object location in 3D appears to be encoded by an efficient orthogonal scheme supported by primary sensory-afferents: each primary-afferent can signal object location by a spatial (labeled-line) code for the vertical axis (along whisker arcs), a temporal code for the horizontal axis (along whisker rows), and an intensity code for the radial axis (from the face out). Neuronal evidence shows that (i) the identities of activated sensory neurons convey information about the vertical coordinate of an object, (ii) the timing of their firing, in relation to other reference signals, conveys information about the horizontal object coordinate, and (iii) the intensity of firing conveys information about the radial object coordinate. Such a triple-coding scheme allows for efficient multiplexing of 3D object location information in the activity of single neurons. Also, this scheme provides redundancy since the same information may be represented in the activity of many neurons. These features of orthogonal coding increase accuracy and reliability. We propose that the multiplexed information is conveyed in parallel to different readout circuits, each decoding a specific spatial variable. Such decoding reduces ambiguity, and simplifies the required decoding algorithms, since different readout circuits can be optimized for a particular variable.
Cited by 0