Multi View
Multi-view analysis integrates data from multiple perspectives to improve accuracy and robustness in various applications, primarily aiming to overcome limitations of single-view approaches. Current research focuses on developing efficient algorithms and model architectures, such as transformers and graph neural networks, to handle high-dimensional data and address challenges like data incompleteness, view misalignment, and computational constraints. This field is significant for advancing computer vision, medical image analysis, robotics, and other domains by enabling more accurate and reliable inferences from complex, multi-faceted data.
Papers
MV-MR: multi-views and multi-representations for self-supervised learning and knowledge distillation
Vitaliy Kinakh, Mariia Drozdova, Slava Voloshynovskiy
Exploring Object-Centric Temporal Modeling for Efficient Multi-View 3D Object Detection
Shihao Wang, Yingfei Liu, Tiancai Wang, Ying Li, Xiangyu Zhang
Sample4Geo: Hard Negative Sampling For Cross-View Geo-Localisation
Fabian Deuser, Konrad Habel, Norbert Oswald
Learning to Select Camera Views: Efficient Multiview Understanding at Few Glances
Yunzhong Hou, Stephen Gould, Liang Zheng
Exploring Recurrent Long-term Temporal Fusion for Multi-view 3D Perception
Chunrui Han, Jinrong Yang, Jianjian Sun, Zheng Ge, Runpei Dong, Hongyu Zhou, Weixin Mao, Yuang Peng, Xiangyu Zhang