English(EN)Hearing the Room Through the Shape of the Drum: Modal-Guided Sound Recovery from Multi-Point Surface Vibrations
新AI方法推动3D重建、图像分割和声音恢复
作者PulseAugur 编辑部·[8 个来源]·
研究人员开发了用于图像分割和重建的新方法。一篇论文介绍了一种使用可微分方法对连续值图像上的简单点检测进行拓扑保持图像分割的新颖方法。另一项研究回顾了制造业中的3D重建技术，强调了深度学习在提高准确性和速度方面的作用，其中质量检测是一个主要应用。第三篇论文提出了一种用于从物体表面振动中恢复声音的物理引导模型，在挑战性场景中表现优于先前的方法。最后，一种用于Ptychographic图像重建的新深度学习框架在单位圆上对相位进行建模，提高了速度和准确性，同时避免了常见的伪影。
AI
arXiv cs.CV
TIER_1English(EN)·Chialoon Cheng (Advanced Robotics Centre, National University of Singapore, Singapore), Kaijun liu (Independent Researcher), Zhiyang Liu (Advanced Robotics Centre, National University of Singapore, Singapore), Marcelo H Ang Jr (Advanced Robotics Centre, N·
arXiv:2604.28064v1 Announce Type: new Abstract: This comprehensive review examines the evolution and the current state of the art in three-dimensional (3D) reconstruction techniques in manufacturing applications. The analysis covers both traditional approaches and emerging deep l…
arXiv cs.CV
TIER_1English(EN)·Wenxiao Li, Faqiang Wang, Yuping Duan, Li Cui, Liqiang Zhang, Jun Liu·
arXiv:2604.28159v1 Announce Type: new Abstract: Topological features play an essential role in ensuring geometric plausibility and structural consistency in image analysis tasks such as segmentation and skeletonization. However, integrating topology-preserving learning based on s…
Topological features play an essential role in ensuring geometric plausibility and structural consistency in image analysis tasks such as segmentation and skeletonization. However, integrating topology-preserving learning based on simple points into deep learning tasks remains ch…
This comprehensive review examines the evolution and the current state of the art in three-dimensional (3D) reconstruction techniques in manufacturing applications. The analysis covers both traditional approaches and emerging deep learning methods, showing a critical research gap…
arXiv cs.CV
TIER_1English(EN)·Carson Yu Liu, Jun Cheng, Chien-Chun Chen, Steve F. Shu·
arXiv:2604.26664v1 Announce Type: cross Abstract: Traditional iterative reconstruction methods are accurate but computationally expensive, limiting their use in high-throughput and real-time ptychography. Recent deep learning approaches improve speed, but often predict phase as a…
arXiv cs.CV
TIER_1English(EN)·Shai Bagon, Matan Kichler, Mark Sheinin·
arXiv:2604.26678v1 Announce Type: new Abstract: Optical vibration sensing enables recovering the scene sound directly from the surface vibration of nearby objects, turning everyday objects into `"visual microphones''. However, most prior methods had focused on capturing the vibra…
Optical vibration sensing enables recovering the scene sound directly from the surface vibration of nearby objects, turning everyday objects into `"visual microphones''. However, most prior methods had focused on capturing the vibrations of specific objects with highly favorable …
Traditional iterative reconstruction methods are accurate but computationally expensive, limiting their use in high-throughput and real-time ptychography. Recent deep learning approaches improve speed, but often predict phase as a Euclidean scalar despite its $2π$ periodicity, wh…
