Zobrazeno 1 - 10
of 224
pro vyhledávání: '"ILIĆ, Slobodan A."'
Autor:
Baumann, Alexander, Ayala, Leonardo, Studier-Fischer, Alexander, Sellner, Jan, Özdemir, Berkin, Kowalewski, Karl-Friedrich, Ilic, Slobodan, Seidlitz, Silvia, Maier-Hein, Lena
Hyperspectral imaging (HSI) is emerging as a promising novel imaging modality with various potential surgical applications. Currently available cameras, however, suffer from poor integration into the clinical workflow because they require the lights
Externí odkaz:
http://arxiv.org/abs/2409.07094
Autor:
Vutukur, Shishir Reddy, Brock, Heike, Busam, Benjamin, Birdal, Tolga, Hutter, Andreas, Ilic, Slobodan
Publikováno v:
3DV 2024
Object Pose Estimation is a crucial component in robotic grasping and augmented reality. Learning based approaches typically require training data from a highly accurate CAD model or labeled training data acquired using a complex setup. We address th
Externí odkaz:
http://arxiv.org/abs/2406.13796
Recent learning methods for object pose estimation require resource-intensive training for each individual object instance or category, hampering their scalability in real applications when confronted with previously unseen objects. In this paper, we
Externí odkaz:
http://arxiv.org/abs/2403.01517
Autor:
Qin, Zheng, Yu, Hao, Wang, Changjian, Guo, Yulan, Peng, Yuxing, Ilic, Slobodan, Hu, Dewen, Xu, Kai
We study the problem of extracting accurate correspondences for point cloud registration. Recent keypoint-free methods have shown great potential through bypassing the detection of repeatable keypoints which is difficult to do especially in low-overl
Externí odkaz:
http://arxiv.org/abs/2308.03768
Autor:
Jung, HyunJun, Ruhkamp, Patrick, Zhai, Guangyao, Brasch, Nikolas, Li, Yitong, Verdie, Yannick, Song, Jifei, Zhou, Yiren, Armagan, Anil, Ilic, Slobodan, Leonardis, Ales, Navab, Nassir, Busam, Benjamin
Learning-based methods to solve dense 3D vision problems typically train on 3D sensor data. The respectively used principle of measuring distances provides advantages and drawbacks. These are typically not compared nor discussed in the literature due
Externí odkaz:
http://arxiv.org/abs/2303.14840
The intrinsic rotation invariance lies at the core of matching point clouds with handcrafted descriptors. However, it is widely despised by recent deep matchers that obtain the rotation invariance extrinsically via data augmentation. As the finite nu
Externí odkaz:
http://arxiv.org/abs/2303.08231
Generative adversarial networks (GANs) offer an effective solution to the image-to-image translation problem, thereby allowing for new possibilities in medical imaging. They can translate images from one imaging modality to another at a low cost. For
Externí odkaz:
http://arxiv.org/abs/2210.06257
Autor:
Yu, Hao, Hou, Ji, Qin, Zheng, Saleh, Mahdi, Shugurov, Ivan, Wang, Kai, Busam, Benjamin, Ilic, Slobodan
Successful point cloud registration relies on accurate correspondences established upon powerful descriptors. However, existing neural descriptors either leverage a rotation-variant backbone whose performance declines under large rotations, or encode
Externí odkaz:
http://arxiv.org/abs/2209.13252
Publikováno v:
IEEE Robotics and Automation Letters, 2021
This paper introduces a novel multi-view 6 DoF object pose refinement approach focusing on improving methods trained on synthetic data. It is based on the DPOD detector, which produces dense 2D-3D correspondences between the model vertices and the im
Externí odkaz:
http://arxiv.org/abs/2207.02811
Publikováno v:
IEEE Transactions on Pattern Analysis and Machine Intelligence 2021
We propose a three-stage 6 DoF object detection method called DPODv2 (Dense Pose Object Detector) that relies on dense correspondences. We combine a 2D object detector with a dense correspondence estimation network and a multi-view pose refinement me
Externí odkaz:
http://arxiv.org/abs/2207.02805