Efficient Pipeline
Efficient pipelines are automated workflows designed to streamline complex processes, primarily aiming to improve speed, accuracy, and resource utilization. Current research focuses on optimizing various pipeline stages, employing diverse techniques like graph neural networks, reinforcement learning for resource allocation, and multi-task learning architectures to handle varied data types and tasks. These advancements are impacting diverse fields, from accelerating 3D model reconstruction and robotic manipulation to enhancing medical image analysis and large language model training, ultimately improving efficiency and enabling new applications.
Papers
PointVoxel: A Simple and Effective Pipeline for Multi-View Multi-Modal 3D Human Pose Estimation
Zhiyu Pan, Zhicheng Zhong, Wenxuan Guo, Yifan Chen, Jianjiang Feng, Jie Zhou
Uncovering communities of pipelines in the task-fMRI analytical space
Elodie Germani (EMPENN), Elisa Fromont (LACODAM), Camille Maumet (EMPENN)