Joel Vidal's research while affiliated with Universitat de Girona and other places
What is this page?
This page lists the scientific contributions of an author, who either does not have a ResearchGate profile, or has not yet added these contributions to their profile.
It was automatically created by ResearchGate to create a record of this author's body of work. We create such pages to advance our goal of creating and maintaining the most comprehensive scientific repository possible. In doing so, we process publicly available (personal) data relating to the author as a member of the scientific community.
If you're a ResearchGate member, you can follow this page to keep up with this author's work.
If you are this author, and you don't want us to display this page anymore, please let us know.
It was automatically created by ResearchGate to create a record of this author's body of work. We create such pages to advance our goal of creating and maintaining the most comprehensive scientific repository possible. In doing so, we process publicly available (personal) data relating to the author as a member of the scientific community.
If you're a ResearchGate member, you can follow this page to keep up with this author's work.
If you are this author, and you don't want us to display this page anymore, please let us know.
Publications (14)
During the last few years, supervised deep convolutional neural networks have become the state-of-the-art for image recognition tasks. Nevertheless, their performance is severely linked to the amount and quality of the training data. Acquiring and labeling data is a major challenge that limits their expansion to new applications, especially with li...
Recently, 6D pose estimation methods have shown robust performance on highly cluttered scenes and different illumination conditions. However, occlusions are still challenging, with recognition rates decreasing to less than 10% for half-visible objects in some datasets. In this paper, we propose to use top-down visual attention and color cues to boo...
Real-time robotic grasping, supporting a subsequent precise object-in-hand operation task, is a priority target towards highly advanced autonomous systems. However, such an algorithm which can perform sufficiently-accurate grasping with time efficiency is yet to be found. This paper proposes a novel method with a 2-stage approach that combines a fa...
Dynamic Contrast Enhanced Magnetic Resonance Imaging (DCE-MRI) has been recognized as an effective tool for Breast Cancer (BC) diagnosis. Automatic BC analysis from DCE-MRI depends on features extracted particularly from lesions, hence, lesions need to be accurately segmented as a prior step. Due to the time and experience required to manually segm...
Real-time robotic grasping, supporting a subsequent precise object-in-hand operation task, is a priority target towards highly advanced autonomous systems. However, such an algorithm which can perform sufficiently-accurate grasping with time efficiency is yet to be found. This paper proposes a novel method with a 2-stage approach that combines a fa...
Dynamic Contrast Enhanced Magnetic Resonance Imaging (DCE-MRI) is a popular tool for the diagnosis of breast lesions due to its effectiveness, especially in a high risk population. Accurate lesion segmentation is an important step for subsequent analysis, especially for computer aided diagnosis systems. However, manual breast lesion segmentation of...
In manufacturing industries, offline programming (OLP) platforms provide an independent methodology for robot integration using 3D model simulation away from the actual robot cell and production process, reducing integration time and costs. However, traditional OLP platforms still require prior knowledge of the workpiece position in a predefined en...
We propose a benchmark for 6D pose estimation of a rigid object from a single RGB-D input image. The training data consists of a texture-mapped 3D object model or images of the object in known 6D poses. The benchmark comprises of: (i) eight datasets in a unified format that cover different practical scenarios, including two new datasets focusing on...
We propose a benchmark for 6D pose estimation of a rigid object from a single RGB-D input image. The training data consists of a texture-mapped 3D object model or images of the object in known 6D poses. The benchmark comprises of: i) eight datasets in a unified format that cover different practical scenarios, including two new datasets focusing on...
Pose estimation of free-form objects is a crucial task towards flexible and reliable highly complex autonomous systems. Recently, methods based on range and RGB-D data have shown promising results with relatively high recognition rates and fast running times. On this line, this paper presents a feature-based method for 6D pose estimation of rigid o...
The Point Pair Feature (Drost et al. 2010) has been one of the most successful 6D pose estimation method among model-based approaches as an efficient, integrated and compromise alternative to the traditional local and global pipelines. During the last years, several variations of the algorithm have been proposed. Among these extensions, the solutio...
Citations
... In addition, it struggles with simulating global illumination effects like indirect light, caustics, or interreflections [25]. Physically Based Rendering (PBR) strives for realism by closely modeling light physics, but this commitment to realism can also limit its flexibility [26,27] making the assumptions and simplifications in PBR not accurate to represent all real-world scenarios and achieving the desired level of photorealism often requires a great deal of computational resources and time. ...
... 9 Several studies have been conducted to assess the performance of using 2D and 3D U-Nets for lesion segmentation from breast DCE-MRI. [10][11][12][13][14] These methods have been developed using different DCE timepoints, datasets sizes, or unique ensembles of modified U-Nets. The evaluation criteria for these studies have been reported across a wide range, demonstrating the complexity of our task. ...
... In real grasping case, these shape-based methods are inaccurate to classify the objects into complete 3D shapes, even less to define the grasping primitives. In recent years, grasp detection methods based on 6D pose estimation have been proposed for grasping regression [8][9][10][11][12][13][14]. Other methods are proposed in order to obtain grasp data directly from the sensors of the robot arm without estimating the object's pose [15][16][17]. ...
![[object Object]](https://i1.rgstatic.net/ii/profile.image/702891513880585-1544593804953_Q64/Tang-Le-3.jpg)
![[object Object]](https://i1.rgstatic.net/ii/profile.image/273768396357653-1442282880752_Q64/Shawn-Le-2.jpg)
![[object Object]](https://i1.rgstatic.net/ii/profile.image/277551164477441-1443184762272_Q64/Chyi-Yeu-Lin.jpg)
... In this work we propose an automated segmentation method for breast lesions in DCE-MRI. This method is based on our previous work in which a ROI guided, 3D patch based U-Net framework was proposed [19]. In this paper we improve the method by using a modified U-Net architecture that incorporates residual basic blocks. ...
... Zheng et al. [8] used vision and CAD interactive activities to establish a mapping relationship between workpiece and model, and realized robot welding path planning. Using the OPEN CASCADE (OCC) opensource library, Bedaka et al. [9,10] built a path-planning platform in which CAD models were input to generate the surface-gluing path. Andulkar et al. [11] proposed an incremental trajectory-generation method to achieve free-form surface painting during the manufacturing of car rears and hoods. ...
![[object Object]](https://i1.rgstatic.net/ii/profile.image/11431281142204318-1681134562640_Q64/Amit-Kumar-Bedaka.jpg)
... The experiments are implemented on a desktop with an Intel Core i7 at 3.20 GHz CPU and a Nvidia Geforce RTX 2080 (8G) GPU. As stated in Benchmark for 6D Object Pose Estimation (BOP) [71], we utilize the maximum symmetry-aware projection distance (MSPD) metric to evaluate the augmented reality tracking effectiveness as follows. ...
![[object Object]](https://i1.rgstatic.net/ii/profile.image/954830499622912-1604660739210_Q64/Tomas-Hodan.jpg)
![[object Object]](https://i1.rgstatic.net/ii/profile.image/579463432097792-1515166256045_Q64/Frank-Michel.jpg)
![[object Object]](https://i1.rgstatic.net/ii/profile.image/472110716985348-1489571374633_Q64/Eric-Brachmann.jpg)
![[object Object]](https://i1.rgstatic.net/ii/profile.image/279693673680899-1443695576951_Q64/Wadim-Kehl-2.jpg)
![[object Object]](https://i1.rgstatic.net/ii/profile.image/279469509103632-1443642131631_Q64/Carsten-Rother.jpg)
... Unlike Drost's method [20], we use a clustering downsampling method that takes into account normal, like Refs. [26,27]. However, we also focus on edge points in the point cloud. ...
![[object Object]](https://i1.rgstatic.net/ii/profile.image/278764568236034-1443474060300_Q64/Xavier-Llado.jpg)
... In terms of methodology, approaches include template-matching methods such as [16,15] that rely on a pre-created set of templates for each object that is associated with ground truth poses and matched to scene objects. Feature-based methods, on the other hand, rely on the extraction and matching of special features such as point-pair features [20] or 3D local features [3]. Other methods try to learn the pose of scene objects directly from monocular input images [24] or from RGB-D data of the scene [21] using end-to-end deep learning architectures. ...
... As such encoded representations are more general and do improve through multi-task learning [101]. 5) Refinement: Classical multi-stage depth-based approaches [60], [34], [143], [67], [2] refine estimates using ICP [121]. Also methods that estimate poses in a monocular fashion [80], [113], [82], [151] exploit additional depth information for ICP-based refinement. ...
... https://www.mdpi.com/journal/electronics Electronics 2023, 12, 3093 2 of 20 calculated [13,14]. This method performs poorly in dynamic environments, and the processing could be more time-consuming. ...
