RGB D Data
RGB-D data, combining color (RGB) and depth information, fuels advancements in 3D scene understanding and object manipulation. Current research emphasizes developing robust and efficient algorithms, often employing deep learning architectures like convolutional neural networks and transformers, to perform tasks such as 3D reconstruction, object pose estimation, and semantic segmentation from this rich data source. These improvements are driving progress in robotics, augmented reality, and medical applications, particularly in areas requiring precise 3D perception and interaction in complex environments. The development of new, diverse datasets is also a key focus, enabling the training and evaluation of more generalizable models.
29papers
Papers - Page 2
March 24, 2023
December 11, 2022
March 26, 2022
November 17, 2021
MPF6D: Masked Pyramid Fusion 6D Pose Estimation
Nuno Pereira, Luís A. AlexandreARKitScenes: A Diverse Real-World Dataset For 3D Indoor Scene Understanding Using Mobile RGB-D Data
Gilad Baruch, Zhuoyuan Chen, Afshin Dehghan, Tal Dimry, Yuri Feigin, Peter Fu, Thomas Gebauer, Brandon Joffe, Daniel Kurz, Arik Schwartz+1