The advancement of 4D (i.e., sequential 3D) generation opens up new possibilities for lifelike experiences in various applications, where users can explore dynamic objects or characters from any viewpoint. Meanwhile, video generative models are receiving particular attention given their ability to produce realistic and imaginative frames. These models are also observed to exhibit strong 3D consistency, indicating the potential to act as world simulators. In this work, we present Video4DGen, a novel framework that excels in generating 4D representations from single or multiple generated videos as well as generating 4D-guided videos. This framework is pivotal for creating high-fidelity virtual contents that maintain both spatial and temporal coherence. The 4D outputs generated by Video4DGen are represented using our proposed Dynamic Gaussian Surfels (DGS), which optimizes time-varying warping functions to transform Gaussian surfels (surface elements) from a static state to a dynamically warped state. We design warped-state geometric regularization and refinements on Gaussian surfels, to preserve the structural integrity and fine-grained appearance details, respectively. Additionally, in order to perform 4D generation from multiple videos and effectively capture representation across spatial, temporal, and pose dimensions, we design multi-video alignment, root pose optimization, and pose-guided frame sampling strategies. The leveraging of continuous warping fields also enables a precise depiction of pose, motion, and deformation over per-video frames. Further, to improve the overall fidelity from the observation of all camera poses, Video4DGen performs novel-view video generation guided by the 4D content, with the proposed confidence-filtered DGS to enhance the quality of generated sequences. In summary, Video4DGen yields dynamic 4D generation with the ability to handle different subject movements, while preserving details in both geometry and appearance. The framework also generates 4D-guided videos with high spatial and temporal coherence. With the ability of 4D and video generation, Video4DGen offers a powerful tool for applications in virtual reality, animation, and beyond.
Download full-text PDF |
Source |
---|---|
http://dx.doi.org/10.1109/TPAMI.2025.3550031 | DOI Listing |
PLoS One
March 2025
Department of Physics, Portland State University, Portland, Oregon, United States of America.
The ability of microbial active motion, morphology, and optical properties to serve as biosignatures was investigated by in situ video microscopy in a wide range of extreme field sites where such imaging had not been performed previously. These sites allowed for sampling seawater, sea ice brines, cryopeg brines, hypersaline pools and seeps, hyperalkaline springs, and glaciovolcanic cave ice. In all samples except the cryopeg brine, active motion was observed without any sample treatment.
View Article and Find Full Text PDFIEEE Trans Med Imaging
March 2025
The detection of cardiac phase in ultrasound videos, identifying end-systolic (ES) and end-diastolic (ED) frames, is a critical step in assessing cardiac function, monitoring structural changes, and diagnosing congenital heart disease. Current popular methods use recurrent neu ral networks to track dependencies over long sequences for cardiac phase detection, but often overlook the short-term motion of cardiac valves that sonographers rely on. In this paper, we propose a novel optical flow-enhanced Mamba U-net framework, designed to utilize both short-term motion and long-term dependencies to detect the cardiac phase in ultrasound videos.
View Article and Find Full Text PDFSurg Endosc
March 2025
Department of Gastroenterology and Hepatology, Digestive Endoscopy Medical Engineering Research Laboratory, Wuhou District, West China Hospital, Sichuan University, No. 37, Guo Xue Alley, Chengdu City, 610041, Sichuan Province, China.
Background: Endoscopic submucosal dissection (ESD) is a crucial yet challenging multi-phase procedure for treating early gastrointestinal cancers. This study developed an artificial intelligence (AI)-based automated surgical workflow recognition model for esophageal ESD and proposed an innovative training program based on esophageal ESD videos with or without AI labels to evaluate its effectiveness for trainees.
Methods: We retrospectively analyzed complete ESD videos collected from seven hospitals worldwide between 2016 and 2024.
J Am Geriatr Soc
March 2025
New England Geriatric Research, Education, and Clinical Centers (GRECC), VA Boston Healthcare System, Boston, Massachusetts, USA.
Background: Older adults with multiple chronic conditions face significant challenges with their health. Patient Priorities Care (PPC) is an Age-Friendly approach that explores 'what matters' by identifying values, care preferences, and health priorities, and aligning healthcare based on patients' health outcome goals.
Methods: Patient priorities care was implemented in four clinical settings (Hospital in Home, a transitional care case management program and in two embedded clinics within specialty care settings) within a large academically affiliated Veteran Affairs hospital system.
Ann Surg
March 2025
Amsterdam UMC, location University of Amsterdam, Department of Surgery, Amsterdam, the Netherlands.
Objective: The aim of this study is to identify learning curves for robotic gastro-enterostomy (RGE) during RPD and the predictive value of the objective structured assessment of technical skills (OSATS) score for DGE according to the Birkmeyer et al and UPMC method.
Summary Of Background Data: In some series, robotic pancreatoduodenectomy (RPD) has been associated with increased risk of delayed gastric emptying (DGE). It is unclear whether this is attributable to learning curve.
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!